the effect of procurement on c f d s p -p p m i p · 2014-06-19 · economics (nie) – namely,...

THE EFFECT OF PROCUREMENT ON

COMPETITION AND FLEXIBILITY : DETERMINING THE SUITABILITY OF

PUBLIC -PRIVATE PARTNERSHIPS IN

MAJOR INFRASTRUCTURE PROJECTS

Pauline, Teo Poh Lian BSc(Hons)(BLG), MPM

A thesis submitted in partial fulfilment of the requirements for the degree of

Doctor of Philosophy

Civil Engineering and Built Environment School

Science and Engineering Faculty

Queensland University of Technology

2014

i

Keywords

Production costs and benefits, transaction costs, major infrastructure, procurement,

competition, flexibility, expressions of interest, value-for-money, Public-Private

Partnerships

ii

Abstract

Government plays a pivotal role in addressing market failure, that is, to prevent

inefficient allocation of resources in the market, resulting in an inequilibrium, such

as lack of competition or monopoly. This is especially important in the provision of

major infrastructure, and is evidenced by the fact that governments are constantly

searching for ways to improve procurement processes that prevent market failure and

enhance Value-for-Money (VfM). More specifically, governments have the ability to

minimise the likelihood of market failure ex ante (or pre-contract) by leveraging the

technological advantages of the market (such as proprietary or tacit knowledge and

resources), and ex post (or post-contract) if the market firm (or private sector firm)

holds-up government because of a change in the works.

Nevertheless, delivering and demonstrating the efficacy of procurement in

addressing market failure and – ultimately – VfM, has remained a challenge, not

least because of difficulties involved in establishing reliable estimates of costs and

benefits in whole-life terms, the intractability of data (particularly in the operational

stage), and a lack of comparative empirical studies across all modes of procurement.

Collectively, this lack of theory and data is evidenced by the fact that procurement

selection remains very much a vexed question.

In light of the above, indirect approaches to procurement selection have greater

appeal. However, the most well-established indirect approach – the Multi-Attribute

Utility Approach (MAUA) – suffers from its own range of weaknesses, which mostly

arise from its lack of theoretical grounding and tautological stance. This research

addresses these weaknesses, and develops a new decision-making model to

determine the appropriate procurement approach – PPP or other – for the provision

of major infrastructure projects.

In the process of developing the model, the research harnesses the relative

strengths of a number of Nobel Prize-winning theories from the New Institutional

Economics (NIE) – namely, Transaction cost theory and Transaction cost economics

– to explain the effect of procurement on Expressions of Interest (EoI) as a key

indicator of competition and flexibility (the scope of contract and/or the ability of the

contractual arrangement to deal with variations in the works) and, in turn, a critical

proxy of market failure and VfM. That is, EoI is indicative of the attractiveness of a

iii

project to the market, with low EoI indicating pre-contract market failure in terms of

oligopoly pricing constraints and high EoI indicating post-contract market failure in

terms of potential negative opportunistic behaviour as a result of lack of flexibility. It

then combines these theories with Resource-based theory (RBT) to delineate key

dimensions of procurement (comprising of size of the project, the level of bundling,

and the nature of exchange relationship).

This research adopts the ontological perspective of realism and uses survey and

case study methods. The statistical analysis of results from a survey of major road

and health projects – comprising of a representative sample of 87 projects (worth

AUD 32 billion) across both Road and Health Sectors in Australia in the five year

period from 2006-2010 – confirm the reliability and validity of using EoI as the

dependent variable and proxy for VfM. The analysis of four case studies of major

road and health infrastructure projects and the analysis of a nationwide survey of

civil and building contractors supports the hypothesis developed to test the model.

This research, and its resultant procurement decision-making model, represents

the first deployment of a combination of (selected) NIE theories and RBT to

determine procurement mode in pursuance of VfM. As such, it is a potentially

alternative procedural framework to MAUA-based procedures, such as Australia’s

National PPP Guidelines (Volume 1: Procurement Options Analysis), and other

MAUA-based procedures around the world. In summary, the a priori model deploys

a novel integration of dominant microeconomic theory in the specific context of

social and economic infrastructure. The model is empirically tested using EoI as a

valid and reliable indicator that the configuration of the key procurement dimensions

(for either PPP or non-PPP mode) selected for the project is more likely to set the

project on a path to superior VfM relative to competing modes.

iv

Table of Contents Keywords ................................................................................................................................................. i

Abstract ................................................................................................................................................... ii

Table of Contents ................................................................................................................................... iv

List of Figures ..................................................................................................................................... viii

List of Tables ......................................................................................................................................... ix

List of Abbreviations ............................................................................................................................ xii

List of relevant publications and presentations by the author .............................................................. xiv

Statement of original authorship ........................................................................................................... xv

Dedication ............................................................................................................................................ xvi

Acknowledgements ............................................................................................................................. xvii

CHAPTER 1: INTRODUCTION ....................................................................................................... 1

1.1 Background to research ............................................................................................................... 1 1.1.1 Capitalism and market failure ........................................................................................... 1 1.1.2 Market failure and transaction costs ................................................................................. 1

1.2 Challenges of value-for-money.................................................................................................... 3

1.3 Current research in procurement selection approaches ................................................................ 6 1.3.1 Weaknesses in multi-attribute utility theory ................................................................... 10

1.4 Procurement selection approach in Australia ............................................................................. 14

1.5 PPP or non-PPP delivery ........................................................................................................... 16

1.6 Research aim .............................................................................................................................. 21

1.7 Research objectives.................................................................................................................... 21

1.8 Justification for research ............................................................................................................ 22 1.8.1 Theory ............................................................................................................................ 22 1.8.2 Research method ............................................................................................................ 23 1.8.3 Practical .......................................................................................................................... 24

1.9 Methodology .............................................................................................................................. 24 1.9.1 Methodology versus methods ......................................................................................... 24 1.9.2 Ontological perspective – Realism ................................................................................. 25 1.9.3 Epistemological position ................................................................................................ 25

1.10 Delimitations of research ........................................................................................................... 26

1.11 Thesis outline ............................................................................................................................. 29

1.12 Summary .................................................................................................................................... 30

CHAPTER 2: LITERATURE REVIEW ......................................................................................... 33

2.1 Introduction................................................................................................................................ 33

2.2 Transaction cost theory .............................................................................................................. 33 2.2.1 Origins and development ................................................................................................ 33 2.2.2 Make-or-buy analysis ..................................................................................................... 34

2.3 Transaction cost economics ....................................................................................................... 36 2.3.1 Origins and development ................................................................................................ 36 2.3.2 Make-or-buy analysis ..................................................................................................... 41 2.3.3 Exchange relationship analysis ....................................................................................... 42

2.4 Resource-based theory ............................................................................................................... 44 2.4.1 Origins and development ................................................................................................ 44

v

2.4.2 The make-or-buy decision .............................................................................................. 48

2.5 Integration of Transaction cost economics and Resource-based theory in the analysis of make-or-buy decisions .................................................................................................................................... 52

2.6 Property rights theory ................................................................................................................ 54 2.6.1 Origins and development ................................................................................................ 54 2.6.2 Bundling ......................................................................................................................... 55

2.7 Agency theory ............................................................................................................................ 56 2.7.1 Origins and development ................................................................................................ 56 2.7.2 Nature of exchange analysis ........................................................................................... 59

2.8 Review of New Institutional Economics literature in construction economics .......................... 62 2.8.1 Make-or-buy analysis (Transaction cost theory, Transaction cost economics and

Resource-based theory) .................................................................................................. 62 a) Transaction cost theory and Transaction cost economics ........................................... 62 b) Resource-based theory and integration with Transaction cost theory ........................ 65

2.8.2 Exchange relationship analysis (Transaction cost theory and Transaction cost economics) ...................................................................................................................... 69

2.8.3 Bundling of design, construct, operations and maintenance activities (Property rights theory) ................................................................................................................... 73

2.8.4 Exchange relationship analysis (Agency theory) ............................................................ 75

2.9 Summary .................................................................................................................................... 78

CHAPTER 3: THEORETICAL DEVELOPMENT ........................................................................ 81

3.1 Introduction ................................................................................................................................ 81

3.2 Procedures in the first-order procurement decision-making model ........................................... 81 3.2.1 Overview ........................................................................................................................ 81 3.2.2 Stage 1/Task A: Activity analysis (first-order) ............................................................... 83 3.2.3 Stage 1/Task B: Make-or-buy analysis (first-order) ....................................................... 85

a) Applying the integrative framework ........................................................................... 85 b) Developing the combination of the value variable in RBT and the frequency variable in TCE ............................................................................................................... 89 c) Key steps .................................................................................................................... 91

3.2.4 Stage 1/Task C: Market analysis (first-order) ................................................................. 92 3.2.5 Stage 2: Bundling analysis (second-order) ..................................................................... 94 3.2.6 Stage 3: Exchange relationship analysis (third-order) .................................................... 97 3.2.7 Outcomes and merits of first-order procurement decision-making model .................... 100

3.3 Approach to testing the first-order procurement decision-making model ................................ 103 3.3.1 Developing the measurement of competition and flexibility as a dependent

variable and indicator of market failure and proxy for value-for-money...................... 103 3.3.2 Market failure ex ante, and pre-contract market failure ................................................ 107 3.3.3 Measure of competition and post-contract market failure ............................................ 108 3.3.4 Development of hypothesis and approach to testing .................................................... 109 3.3.5 Refutability and hypothesis testing ............................................................................... 112

3.4 Summary .................................................................................................................................. 113

CHAPTER 4: RESEARCH DESIGN AND METHODS .............................................................. 115

4.1 Introduction .............................................................................................................................. 115

4.2 Research design ....................................................................................................................... 115 4.2.1 Generally ...................................................................................................................... 115 4.2.2 Survey method rationale ............................................................................................... 117 4.2.3 Case study method rationale ......................................................................................... 118 4.2.4 Merits of multiple methods ........................................................................................... 118 4.2.5 Reliability and validity ................................................................................................. 119

4.3 Questionnaire design and administration ................................................................................. 120 4.3.1 Actual procurement and actual competition (survey of road and health projects) ........ 120

a) Purpose and content .................................................................................................. 120

vi

b) Questionnaire design and pilot meetings .................................................................. 121 c) Population and sampling frame ................................................................................ 123 d) Interviews and administration .................................................................................. 124 e) Data entry and coding procedures ............................................................................ 125 f) Response rate and representation of data .................................................................. 126

4.3.2 Theoretical competition (survey of civil and building contractors) .............................. 130 a) Purpose and content .................................................................................................. 130 b) Questionnaire design and pilot meetings .................................................................. 132 c) Population and sampling frame ................................................................................ 133 d) Improving response rates (reducing non-response bias) .......................................... 135 e) Administration and data collection ........................................................................... 139 f) Non-response bias analysis ....................................................................................... 140 g) Response rates and representation of data ................................................................ 141 h) Data entry and coding procedures ............................................................................ 144

4.4 Components of case study design (theoretical procurement using the procurement model) ... 144 4.4.1 Research question ......................................................................................................... 144 4.4.2 Proposition .................................................................................................................... 144 4.4.3 Unit of analysis ............................................................................................................. 145 4.4.4 Logic linking case study data to hypothesis in the approach to selecting and

testing case studies ....................................................................................................... 145 4.4.5 Criteria for judging quality of research design ............................................................. 146 4.4.6 Design of case study protocol and questionnaire instrument ........................................ 147

a) Unstructured and exploratory interviews .................................................................. 147 b) Semi-structured interviews ....................................................................................... 149 c) Design of questionnaire instrument for TCE and RBT variables ............................. 152 1. Value or capacity ................................................................................................. 153 2. Rarity.................................................................................................................... 154 3. Costly-to-imitate .................................................................................................. 155 4. Asset specificity ................................................................................................... 156 5. Uncertainty ........................................................................................................... 158 6. Frequency ............................................................................................................. 159

4.4.7 Measurement scale ....................................................................................................... 165 4.4.8 Administration of case studies ...................................................................................... 165

a) Case study selection ................................................................................................. 165 b) Case study data collection ........................................................................................ 166 c) Structured interviews ................................................................................................ 167 1. Administering Road Case #R1 in review mode ................................................... 168 2. Administering Road Case #R2 in review mode ................................................... 168 3. Administering Health Case #H2 in review mode ................................................. 169 4. Administering Health Case #H1 in review mode ................................................. 169

4.5 Ethics ....................................................................................................................................... 169

4.6 Summary .................................................................................................................................. 170

CHAPTER 5: ANALYSIS OF SURVEY AND CASE STUDY DATA ....................................... 171

5.1 Introduction.............................................................................................................................. 171

5.2 Questionnaire survey of major road and health projects .......................................................... 171 5.2.1 Actual procurement ...................................................................................................... 171 5.2.2 Actual competition ....................................................................................................... 174 5.2.3 General relationship between procurement and competition ........................................ 178 5.2.4 Key patterns from actual procurement, and indications of scope to improve VfM ...... 178 5.2.5 Summary and key patterns ........................................................................................... 180

5.3 Road Case Study #R1 (Theoretical procurement) ................................................................... 181 5.3.1 Stage 1/Task A: Activity analysis................................................................................. 182 5.3.2 Stage 1/ Task B: Make-or-buy analysis ........................................................................ 183 5.3.3 Stage 1/Task C: Market analysis .................................................................................. 188 5.3.4 Stage 2: Bundling analysis ........................................................................................... 191 5.3.5 Stage 3: Exchange relationship analysis ....................................................................... 192 5.3.6 Procurement strategy .................................................................................................... 193

vii

5.3.7 Hypothesis testing (including theoretical competition) ................................................ 194

5.4 Road Case Study #R2 (Theoretical procurement) .................................................................... 194 5.4.1 Stage 1/Task A: Activity analysis ................................................................................. 195 5.4.2 Stage 1/Task B: Make-or-buy analysis ......................................................................... 196 5.4.3 Stage 1/Task: C Market analysis .................................................................................. 206 5.4.4 Stage 2: Bundling analysis ............................................................................................ 206 5.4.5 Stage 3: Exchange relationship analysis ....................................................................... 209 5.4.6 Procurement strategy .................................................................................................... 211 5.4.7 Hypothesis testing (including theoretical competition) ................................................ 212

5.5 Health Case Study #H2 (Theoretical procurement) ................................................................. 212 5.5.1 Stage 1/Task A: Activity analysis ................................................................................. 213 5.5.2 Stage 1/Task B: Make-or-buy analysis ......................................................................... 216 5.5.3 Stage 1/Task C: Market analysis .................................................................................. 226 5.5.4 Stage 2: Bundling analysis ............................................................................................ 227 5.5.5 Stage 3: Exchange relationship analysis ....................................................................... 230 5.5.6 Procurement strategy .................................................................................................... 231 5.5.7 Hypothesis testing (including theoretical competition) ................................................ 231

5.6 Health Case Study #H1 (Theoretical procurement) ................................................................. 232 5.6.1 Stage 1/Task A: Activity analysis ................................................................................. 233 5.6.2 Stage 1/Task B: Make-or-buy analysis ......................................................................... 235 5.6.3 Stage 1/Task C: Market analysis .................................................................................. 242 5.6.4 Stage 2 Bundling analysis ............................................................................................. 242 5.6.5 Stage 3: Exchange relationship analysis ....................................................................... 245 5.6.6 Procurement strategy .................................................................................................... 246 5.6.7 Hypothesis testing (including theoretical competition) ................................................ 246

5.7 Overall results of hypothesis testing ........................................................................................ 247

5.8 Discussion of findings .............................................................................................................. 248

5.9 Summary .................................................................................................................................. 253

CHAPTER 6: CONCLUSIONS ...................................................................................................... 255

6.1 Introduction .............................................................................................................................. 255

6.2 Scope and certainty of contributions to knowledge ................................................................. 258 6.2.1 New Institutional Economics and Resource-based Theory .......................................... 258 6.2.2 Research methods ......................................................................................................... 260 6.2.3 Procurement selection literature ................................................................................... 261

6.3 Implications and recommendations for practice ...................................................................... 263

6.4 Limitations of the model .......................................................................................................... 264

6.5 Recommendations for future research ..................................................................................... 265

REFERENCES .................................................................................................................................. 267

Appendix A – List of MAUA-based procurement approaches ........................................................... 299

Appendix B – Survey of major road and health projects (final version) ............................................. 301

Appendix C – Coding of survey of major road and health projects .................................................... 311

Appendix D – Survey of civil and building contractors (final version) .............................................. 313

Appendix E – Notification letter for survey of civil and building contractors .................................... 329

Appendix F – Cover letter for survey of civil and building contractors .............................................. 331

Appendix G – Coding of survey of civil and building contractors ...................................................... 333

Appendix H – Case study questionnaire instrument (initial version) for semi-structured interviews.335

Appendix I – Case study questionnaire instrument for health and road projects ............................... 341

Appendix J – Case study database ..................................................................................................... 357

viii

List of Figures

Figure 1.1: Value map and value improvement curve ............................................................................ 4

Figure 1.2: MAUA and VfM in nominal terms..................................................................................... 13

Figure 1.3: New model versus current approach ................................................................................... 19

Figure 1.4: New Institution Economics Theories .................................................................................. 19

Figure 2.1: Simple contractual schema ................................................................................................. 39

Figure 2.2: Efficient governance ........................................................................................................... 43

Figure 2.3: The TCR trinity .................................................................................................................. 71

Figure 2.4: Summary of external exchange relationship ....................................................................... 72

Figure 2.5: Goals and methods matrix .................................................................................................. 76

Figure 2.6: Contract selection based on uncertainty of project and risk allocation ............................... 77

Figure 3.1: Tasks in the first-order procurement decision-making model ............................................ 82

Figure 3.2: Exchange relationship ......................................................................................................... 98

Figure 3.3: First-order procurement decision-making model and VfM in nominal terms .................. 102

Figure 3.4: VfM against competition .................................................................................................. 106

Figure 3.5: Summary of theoretical development ............................................................................... 110

Figure 3.6: Procurement-competition/flexibility hypothesis ............................................................... 110

Figure 3.7: Multiple case studies’ replication design .......................................................................... 112

Figure 4.1: Research design ................................................................................................................ 117

Figure 4.2: Respondent profile in survey of civil and building contractors ........................................ 142

Figure 4.3: Case study design ............................................................................................................. 146

Figure 4.4: Sample of submitted road and health projects .................................................................. 165

Figure 5.1: Road and health EoIs ........................................................................................................ 176

Figure 5.2: Road and health EoIs and commencement dates .............................................................. 177

Figure 5.3: Road and health EoIs and overall capital value ................................................................ 177

Figure 5.4: Increasing rationalisation of procurement ........................................................................ 252

Figure 6.1: Summary of research framework ...................................................................................... 256

ix

List of Tables

Table 1.1 Costs of internalisation and externalisation ............................................................................. 5

Table 1.2 Categorical list of procurement approaches ............................................................................ 7

Table 1.3 Examples of client priorities for procurement selection .......................................................... 9

Table 1.4 Australian states’ procurement guidelines............................................................................. 15

Table 1.5 Mixed PPP reviews and evaluations ..................................................................................... 16

Table 2.1 Summary of research on the integration of TCE and RBT ................................................... 53

Table 2.2 Summary of behaviour, hybrid and outcome-based contracts ............................................... 62

Table 2.3 Summary of singular transaction cost approach in procurement selection............................ 64

Table 2.4 Research on the integration of TCE and RBT in the construction industry .......................... 66

Table 2.5 Summary of research on efficient exchange relationship in the construction industry ......... 70

Table 3.1 Detailed measurement of activity levels................................................................................ 86

Table 3.2 SCP analysis in relation to activity levels ............................................................................. 93

Table 3.3 Summary of outcome, hybrid and behaviour-based contracts in PAT .................................. 99

Table 3.4 Strengths of first-order procurement decision-making model ............................................. 102

Table 4.1 Research methods and data collection instruments ............................................................. 116

Table 4.2 Design of questionnaire survey of major road and health projects...................................... 121

Table 4.3 Pilot meetings for design of questionnaire in survey of major road and health projects ..... 122

Table 4.4 List of budget paper documents .......................................................................................... 123

Table 4.5 List of interviews following submission of questionnaire survey of major road and health projects .................................................................................................................... 125

Table 4.6 List of statewide budget papers reviewed ........................................................................... 126

Table 4.7 Response to survey of major road and health projects categorised by sector ...................... 127

Table 4.8 Response to survey of major road and health projects categorised by project location ...... 127

Table 4.9 Response to survey of major road and health projects categorised by project timing ......... 128

Table 4.10 Response to survey of major road and health projects categorised by project size ........... 128

Table 4.11 Survey of civil and building contractors – Pilot meetings ................................................. 132

Table 4.12 Development of survey questionnaire for civil and building contractors .......................... 133

Table 4.13 Breakdown of mailed questionnaires ................................................................................ 135

Table 4.14 Summary of TDM approaches applied in the survey of civil and building contractors .......................................................................................................................... 139

Table 4.15 Administration of questionnaire in survey of civil and building contractors .................... 140

Table 4.16 Summary of eligible contractors in each state ................................................................... 141

Table 4.17 Response to survey of civil and building contractors by sector ........................................ 142

Table 4.18 Response rate for surveys of civil and building contractors .............................................. 143

Table 4.19 Unstructured and exploratory interview register ............................................................... 148

Table 4.20 Case study questionnaire design and semi-structured interview register .......................... 150

Table 4.21 Case study protocol and questionnaire design ................................................................... 152

x

Table 4.22 Summary of selected case studies ..................................................................................... 166

Table 4.23 Data collection register (in chronological order)............................................................... 167

Table 4.24 Case study meeting register (in chronological order) ........................................................ 167

Table 5.1 Actual procurement types across road and health sectors ................................................... 172

Table 5.2 Single or multiple contracts across road and health sectors ................................................ 173

Table 5.3 Payment terms across road and health sectors .................................................................... 173

Table 5.4 EoI, ItT, and ST in the road sector ...................................................................................... 175

Table 5.5 EoI, ItT, and ST in the health sector ................................................................................... 175

Table 5.6 Make-or-buy analysis for design activities for #R1 ............................................................ 184

Table 5.7 Make-or-buy analysis for construction activities for #R1 ................................................... 185

Table 5.8 Make-or-buy analysis for operations and maintenance activities for #R1 .......................... 187

Table 5.9 Summary of make-or-buy analysis for #R1 ........................................................................ 188

Table 5.10 Market structure analysis of key activities for #R1 ........................................................... 189

Table 5.11 Main design activity bundle and main construction activity bundle for #R1 .................... 191

Table 5.12 Exchange relationship analysis for #R1 ............................................................................ 192

Table 5.13 Make-or-buy analysis for design activities for #R2 .......................................................... 197

Table 5.14 Make-or-buy analysis of construction activities in cut-and-cover tunnels for #R2 ........... 198

Table 5.15 Make-or-buy analysis for construction activities in driven tunnel for #R2 ....................... 200

Table 5.16 Make-or-buy analysis of construction activities for at-grade road, bridge and ramp structures, and two bus stations (#R2) ............................................................................... 201

Table 5.17 Make-or-buy analysis of other construction activities in multiple parts of #R2 ............... 203

Table 5.18 Make-or-buy analysis for operations and maintenance activities of #R2 .......................... 203

Table 5.19 Summary of internalised activities for #R2 ....................................................................... 204

Table 5.20 Summary of externalised activities for #R2 ...................................................................... 205

Table 5.21 Bundles of Level 7 and Level 4b activities for #R2 .......................................................... 207

Table 5.22 Main design activities bundle for #R2 .............................................................................. 208

Table 5.23 Main construction activities bundle for #R2 ..................................................................... 208

Table 5.24 Exchange relationship analysis of #R2 ............................................................................. 209

Table 5.25 Make-or-buy analysis of design activities for #H2 ........................................................... 217

Table 5.26 Make-or-buy analysis of Level 7 construction activities for #H2 ..................................... 218



Table 5.29 Make-or-buy analysis of operation activities for #H2 ....................................................... 222

Table 5.30 Make-or-buy analysis of maintenance activities for #H2 .................................................. 223

Table 5.31 Summary of externalised activities that match practice for #H2....................................... 225

Table 5.32 Summary of externalised activities that do not match practice for #H2............................ 226

Table 5.33 Main design activities bundle for #H2 .............................................................................. 228

Table 5.34 List of separate design activities for #H2 .......................................................................... 228

Table 5.35 Main construction activities bundle for #H2 ..................................................................... 229

Table 5.36 Main maintenance activities bundle for #H2 .................................................................... 230

xi

Table 5.37 List of separate key maintenance activities for #H2 .......................................................... 230

Table 5.38 Exchange relationship analysis of #H2 ............................................................................. 230

Table 5.39 Make-or-buy analysis of design of new installation works for #H1.................................. 237

Table 5.40 Make-or-buy analysis of construction activities 21 to 52 for #H1 .................................... 238

Table 5.41 Make-or-buy analysis for construction activities 53 to 68 for #H1 ................................... 239

Table 5.42 Make-or-buy analysis of the implementation of operations for #H1................................. 240

Table 5.43 Make-or-buy analysis of the implementation of maintenance for #H1 ............................. 241

Table 5.44 Summary of Levels 4b, 6 and 7 activities ......................................................................... 241

Table 5.45 List of Level 4b and Level 7 design activities for #H1 ..................................................... 242

Table 5.46 Main Design Activity Bundle #1 for #H1 ......................................................................... 243

Table 5.47 List of individual key design activities for #H1 ................................................................ 244

Table 5.48 Main maintenance activities bundle for #H1 ..................................................................... 244

Table 5.49 Exchange relationship analysis for #H1 ............................................................................ 245

Table 5.50 Testing the procurement-competition/flexibility hypothesis ............................................. 247

xii

List of Abbreviations AC Alliance Contracting

AHP Analytic Hierarchy Process

AV Audio Visual

BEMS Building Engineering Maintenance Services

BMS Building Management System

BMCS Building Management Control System

CM Construction Management

CO Construct Only

CT Computed Tomography

DC Design and Construct

DCM Design, Construct, and Maintenance

DCOM Design, Construct, Operations and Maintenance

ECI Early Contractor Involvement

EoI Expression of interest

EWIS Emergency warning and intercommunication system

FFE Fixtures, furnishings and equipment

FIP Fire Indicator Panel

GFC Global Financial Crisis

GCS Guaranteed Construction Sum

GFA Gross floor area

HM Her Majesty

HVAC Heating, ventilation, and air-conditioning

ItT Invited to Tender

ITS Intelligent Transportation Systems

MAUA Multi-Attribute Utility Approach

MC Managing Contractor

xiii

NIE New Institutional Economics

NSW New South Wales

OECD Organisation for Economic Co-operation and Development

O&M Operations and Maintenance

OT Open Tender

PAT Principal-agent Theory/ Agency theory

PDP Project Definition Plan

PFI Private Finance Initiative

PM Project management

POA Procurement Options Analysis

PPP Public Private Partnership

PRT Property Rights Theory

PSC Public Sector Comparator

QLD Queensland

RBT Resource-based Theory

RFP Request for Proposal

SA South Australia

ST Submitted Tender

TC Transaction Cost Theory

TCE Transaction Cost Economics

TCR Trust-commitment-relationship

UK United Kingdom

VfM Value-for-money

VIC Victoria

VRI Value, rarity and imitability

VRIO Value, rarity, imitability and organisation

WA Western Australia

xiv

List of relevant publications and presentations by the author

Teo, P., Bridge, A., & Gray, J. (2013, 5-9 May 2013). A new first-order decision-making model for the procurement of public sector infrastructure: Procedures and Testing. In Proceedings of the 19th CIB World Building Congress, Brisbane 2013: Construction and Society.

Teo, P., Bridge, A., Gray, J., & Rowlinson, S. (2012, 23 – 25 January 2012). Towards testing a new first-order decision making model for the procurement of public sector major infrastructure. In Proceedings of International Conference on Facilities Management, Procurement Systems and Public Private Partnership (CIB W070, W092 and TG72).

Teo, P., & Bridge, A. (2012). Project 1: Construction capacity and competition in the Australian major road and health infrastructure market - Survey of Civil Contractors and Building Contractors. Working paper, confidential publication to ARC, Queensland University of Technology.

Teo, P., Bridge, A., Gray, J., & Jefferies, M. C. (2011, 5-7 September 2011). Developing a research method to test a new first-order decision making model for the procurement of public sector major infrastructure. In Proceedings of Association of Researchers in Construction Management Annual Conference (ARCOM 2011).

Teo, P., Bridge, A., & Gray, J. (2011). Towards testing a new first-order decision making model for the procurement of public sector major infrastructure. Working paper, Confidential publication to ARC, Queensland University of Technology.

Teo, P., Bridge, A., & Jefferies, M. C. (2010). Delivering value for money in the procurement of public sector major infrastructure: A new first-order decision making model. In Proceedings of 2010 International Conference On Construction & Real Estate Management (pp. 306-311).

xv

Statement of original authorship

The work contained in this thesis has not been previously submitted to meet

requirements for an award at this or any other higher education institution. To the

best of my knowledge and belief, the thesis contains no material previously

published or written by another person except where due reference is made.

Signature: Poh Lian Teo, Pauline

Date: 17 – 6 – 2014 4

QUT Verified Signature

xvi

Dedication

To my devoted Mother,

Soh Geok Hua,

and in fond memory of my loving Father,

Teo Lee Min,

with love

xvii

Acknowledgements

My heartfelt thanks and gratitude extend to all the people who have contributed

either intellectually or emotionally throughout the four years of my doctoral studies.

Without their support, this thesis would definitely not have been possible. I would

like to take this opportunity to acknowledge the following people who have been an

integral part of this research.

First and foremost, I thank and express my deepest gratitude to my Principal

Supervisor, Dr Adrian Bridge, for his commitment and constant supervision,

relentless passion, words of encouragement, and continuous support throughout my

PhD candidature. I would also like to acknowledge Adrian for his advice on applying

the integrative framework of vertical integration which is central to the development

of the model. I also thank my associate supervisors, Professor Martin Skitmore for

his insightful advice in relation to my research methodology and statistical analysis,

and Mr Jason Gray for his support for the research and his words of encouragement.

I am also thankful to Professor Steve Rowlinson, Mr Marcus Jefferies and Mr Dale

Gilbert for their support and encouragement.

I acknowledge the support of the following organisations for their financial

contribution to the research, namely: Australian Research Council; Partnerships

Victoria, Department of Treasury and Finance, Victoria; New South Wales Treasury;

Queensland Treasury and Trade; Western Australia Department of Treasury and

Finance; South Australia Department of Treasury and Finance; Queensland

Department of Employment, Economic Development and Innovation; Construction

Industry Institute Australia; Aurecon; and Coffey Commercial Advisory.

I would like to thank the following people whose assistance has contributed

significantly to the research:

• Mr Ross Guppy [former Deputy Chief Engineer (Contracts, Technical Research

& Capability), Department of Transport and Main Roads] for his tremendous

support for the entire research process

xviii

• Mr Jon Oxford [former Director of Research and Development (Contracts,

Technical Research and Capability), Department of Transport of Main Roads] for

his support and encouragement

• Mr Steve Hogan [former Principal Advisor (Engineering Program), Department

of Transport and Main Roads] for sharing his wealth of knowledge and expertise

in the road sector, his great enthusiasm, and his generosity in spending precious

time in providing comments on the design of the case study questionnaire

instrument

• Mr Robert Phillips [former Principal Project Manager, Health Law and Order

Portfolio, Project Services, Department of Housing and Public Works] and Mr

Keith Deppeler [former Contract Superintendent, South Coast Region, Project

Services, Department of Housing and Public Works] for their expert advice on

health-related activities, and their remarkable efforts and commitment in

commenting on the research design

• All the respondents and participants in the survey of major road and health

projects and the survey of civil and building contractors, and the many others

who helped me with this in one way or another

My special thanks go to my close friends, Kaveh Bevrani, who is always there

for me (especially during difficult times), and Nicole Tan who is always

encouraging, understanding, and helpful.

Last but not least, I express my immense gratitude to my parents who have

given me so much in life and showered me with unconditional love and care. I also

thank my younger brother, Teo Jun Xiong, and his wife, Ooi Kim Sze, who have

shown me love, support and encouragement.

Thanks are also due

Chapter 1: Introduction 1

Chapter 1: Introduction

1.1 Background to research

1.1.1 Capitalism and market failure

Adam Smith (1776), regarded as the father of capitalism and economics,

depicts a utopian world of capitalism in which resources will be efficiently allocated

by the “invisible hand” in a free market, without any government intervention. He

asserts that “ every individual naturally inclines to employ his capital in the manner

in which it is likely to afford the greatest support to domestic industry, and to give

revenue and employment to the greatest number of people of his own country”

(Smith, 1776, p. 455), and is “led by an invisible hand to promote an end which was not

part of his intention” (Smith, 1776, p. 456).

According to the analogy of the invisible hand, while pursuing self-interest,

individuals unintentionally serve the public interest more than if they had actually

intended to. According to neoclassical economics theory, this efficient allocation of

goods and services is referred to as Pareto optimality or Pareto efficiency, where no one

party can be made better off without detriment to another party. While this promotes a

mundane morality of the market (Lee & Clark, 2013), such a utopian scenario does not

always occur and, when the pursuit of private interests does not lead to an efficient use

of society’s resources or a fair distribution of society’s goods, then market failure occurs

(Weimer & Vining, 1999, p. 41). This concept of market failure is central to welfare

economics and is defined by Bator (1958, p. 351) as “the failure of a more or less

idealised system of price-market institutions to sustain desirable activities or to estop

‘undesirable activities’.”

1.1.2 Market failure and transaction costs

Contrary to Smith’s much cited analogy, Stiglitz (1985) argues that the ‘invisible

hand’ does not exist. The common drivers of market failures include the undersupply

of public goods, negative externalities such as climate change, non-competitive

markets or natural monopolies, information asymmetries, and undesirable income

distributions (Zerbe & McCurdy, 1999). Given that market failures, in particular

externalities, are ubiquitous and occur in every sector (Nelson, 1987), government

2 Chapter 1: Introduction

has the important role of addressing market failure and ensuring a just distribution of

resources through robust regulatory frameworks and public policy.

The effects of market failure can be extensive, as was evidenced in the Great

Depression of the 1920s and the Global Financial Crisis (GFC) in 2008; these

catastrophic events were both examples of the failure of regulatory framework

(Acharya, Cooley, Richardson, & Walter, 2011). In accepting Friedman’s (1962, p.

38) observation that “the Great Depression like most other periods of severe

unemployment, was produced by government mismanagement rather than by any

inherent instability of the private economy”, market failure can be viewed as the

government’s inability to regulate and protect the economy (Winston, 2006).

At a microanalytical level, Williamson (1985, p. 19) argues that market failure

is fundamentally caused by transaction costs. He supports this argument by referring

to Kenneth Arrow’s observation that “it is better to consider a broader category than

that of transaction costs, which in general impede and in particular cases completely

block the formation of markets” (Arrow, 1969, p. 59). The origins of transaction cost

theory can be traced back to Coase (1937) in “The nature of the firm”, where he

recognises the existence of market transaction costs. In another influential paper, the

‘Problem of social costs’ (1960), he further explains that the allocation of resources

in the economy would be efficient in a market of zero transaction costs, and that

market failure would not exist.

Transaction costs, including opportunistic costs, proposed by Williamson

(1985), have been substantially ignored in neoclassical economics theory. In fact,

neoclassical economics – or the marginalist approach – has often been criticised for

ignoring market failure and being unrealistic in its assumptions of perfect

competition and Pareto optimality. Harold Demsetz (1969) refers to this practice of

comparison with the idealistic benchmark of Pareto efficiency, rather than

institutional comparison, as ‘the nirvana approach’. Hence, the transaction cost logic

might be able to provide a more well-grounded concept for understanding market

efficiencies, and for decreasing the chance of market failure.

In the provision of public goods and infrastructure, and in pursuance of

achieving value-for-money (VfM) for tax-payers, governments have the important

role of addressing issues of market failure; specifically, they have a responsibility to

maintain an efficient non-monopolistic market, and to minimise internal and external


transaction costs. As an example of this challenge, governments are compelled to

actively search for improvements in procurement processes.

1.2 Challenges of value-for-money

Governments in many countries are facing significant demands in providing

infrastructure to meet the needs of increasing population growth, migration and

demographic changes. The global demand for infrastructure is increasing

exponentially and the Organisation for Economic Co-operation and Development

(OECD) (2006) estimated USD 53 trillion to be required globally between 2007-

2030. In Australia, Infrastructure Partnerships Australia (2010) compiled a pipeline

of infrastructure projects that amounts to more than AUD 770 billion (in 2007 terms)

over the next ten years. The funding and delivery of this new infrastructure, while at

the same time achieving VfM, is presenting severe challenges for governments

around the globe, especially in the wake of the GFC and the Eurozone debt crisis.

VfM is an economic concept which can be considered as achieving the

effective and efficient procurement of infrastructure. The UK National Audit Office

(2004, p. 3) defines procurement as, “the whole-life process of the acquisition of

goods, services and works..., beginning when a potential requirement is identified

and ending with the conclusion of service contract or ultimate disposal of an asset”.

This is consistent with the HM Treasury (2008, p. 35) definition of VfM as,

“securing the best mix of quality and effectiveness for the least outlay over the

period of use of the goods or services bought. It is not about minimising upfront

prices...” The ‘best mix’ can be interpreted as the best ratio between benefit and cost

(return/cost) or f(benefit/cost) (Johnson, 1990). Furthermore, HM Treasury (2008)

adds that VfM is a relative concept and can be measured in comparison with other

outcomes. Figure 1.1 depicts the relative concept of VfM; that is, any relative or

alternative position associated with sectors or positions A, B, C, D, or E, compared to

the initial position shown, would constitute an improvement in VfM. The alternative

positions aim to improve value by increasing benefits and/or decreasing costs

compared to the original position and is consistent with the underlying principle of

value management and which is to achieve the required functions specified by the

clients at the lowest possible overall cost, consistent with the requirements for

performance (Shen, 1993; Shen & Yu, 2012).


Figure 1.1 conveys the relative concept of VfM in simple terms; however,

demonstrating VfM in the provision of infrastructure is still a major challenge for all

governments. While this challenge arises from the scale, urgency and complexity of

infrastructure itself, it is more fundamentally embedded in the uniqueness of each

piece of infrastructure. This uniqueness creates difficulties in developing empirical

comparative evaluations across different procurement modes. In seeking to

demonstrate VfM then, the emphasis is on the importance of developing reliable

estimates of costs and benefits associated with alternative procurement modes.

Key: • Alternative position: Type A value improvement - Increase in cost, greater increase in benefit

compared to initial position • Alternative position: Type B value improvement - Same cost, increase in benefit compared to

initial position • Alternative position Type C value improvement - Decrease in cost, increase in benefit compared

to initial position • Alternative position Type D value improvement - Decrease in cost, same benefit compared to

initial position • Alternative position Type E value improvement - Decrease in benefit, greater decrease in cost

compared to initial position

Figure 1.1: Value map and value improvement curve (Adapted from Bridge, 1999, p. 76)

Estimating costs and benefits of alternative modes of procurement is

challenging for a number of reasons. For instance, costs are whole-of-life costs and

require predictions of future repairs and maintenance. Furthermore, beyond near-

term capital costs and whole-of-life costs (production costs), transaction costs that

comprise of internal transaction costs and external transaction costs are much less

observable and more difficult to estimate. These internal and external transaction

infrastructure project costs occur in activities that are both internalised and

externalised by government. Some of these transaction costs are shown in Table 1.1.

E

D

C

B

Initial position

Benefit

Cost

Value curve

A


Table 1.1 Costs of internalisation and externalisation (Based on Teo, Bridge, & Jefferies, 2010) and adapted from “Value for Money: Procuring Infrastructure,” by Bridge et al. (2010). Copyright 2010, Reprinted with permission from IGI Global.

Costs of internalised activity Costs of externalised activity Production costs

• Direct costs of resources (salaries and on-costs; capital costs of equipment, buildings; operating costs)

• Costs of mistakes caused by internalised activity, or by reworking by internal resources and reworking by external firms

• Prices; service charges; patronage costs; contracts (etc.) with external firms

External transaction costs

• Market transaction costs incurred in obtaining internal resources (staff; capital equipment; working space; consumables, etc.)

• Market transaction costs involved in operating and managing internal resources; e.g., cost of Human Resource Department in managing staff

• Market transaction costs incurred in finding, bidding and negotiating prices with external firms; in executing external contracts; in contract management and administration

• Costs associated with appropriation of better terms on the occurrence of a variation (hold-up) at any stage of the project’s construction and/or operation

Internal transaction costs

• Bureaucracy costs associated with separability and lack of compliance (lack of certainty) with contracted cost, and with quality and time performance requirements

• Bureaucracy costs associated with ownership/costs of low power incentives (incentive misalignment/lack of positive production investments and induced negative investment or quality/ performance shading)

• Loss of in-house knowledge, capability and competence

Market transaction costs generated in the engagement and administration of

both internal resources and external private sector firms are rarely captured in full by

government, and elude accurate estimation. In terms of bureaucracy costs, recent

benchmarking studies – most notably, the study by Raisbeck, Duffield and Xu (2010)

– are beginning to shed light on the relative performance of various procurement

modes.

In this study, the relative performance or certainty of outcomes from Public

Private Partnerships (PPP) versus traditional procurement is measured in terms of

percentage change in time and cost in three time periods, from formal approval to the

point at which the project is delivered and begins its operation. This kind of empirical

work indicates the benefits of procurement modes that incorporate more single point

responsibility (associated with less separability through less internalisation) across a

greater service scope including design, construction, operations, and maintenance. In

turn, this creates greater incentive alignment that induces greater positive production

investment in securing time and cost certainty, and helps displace negative

investment directed either at meeting contractual obligations to the letter only or at

justifying the avoidance of contractual obligations (quality/performance shading).


External transaction costs associated with hold-up cannot be effectively

estimated in monetary terms ex ante (pre-contract). Hold-up refers to the

phenomenon associated with the quasi-rents and is the threat of the contractor or

consortium delivering the project behaving in a negative opportunistic way when a

variation in the works occurs. Variations can occur during construction and/or during

operations and maintenance and might see the contractor or consortium seeking to

appropriate better terms (time and/or monies) for one or more of these variations.

Hold-up can be prevented or reduced in its severity by selecting an appropriate

procurement approach. If the procurement mode successfully addresses hold-up, then

no such transaction costs related to hold-up ex post (post-contract) will arise;

nevertheless, the mere threat or likelihood of hold-up can inform the successful

selection of a procurement approach.

As well as presenting difficulties in estimating production and transaction

costs, alternative procurement modes also present difficulties in estimating their

related benefit outcomes. This is because alternative modes might incorporate

different design proposals, which include considerations (such as aesthetics and

environmental impacts) which are subjective by definition and, thus, difficult to

evaluate in an objective manner. Moreover, other benefits or performance outcomes

– such as absolute or minimum time to deliver the infrastructure – can also attract

different utility values from various stakeholders.

These difficulties in attempting to estimate actual production costs and

transaction costs are exacerbated by a dearth of infrastructure empirical cost and

performance data in whole-life terms. Thus, this is likely to be, at least in part, the

reason for reliance on indirect approaches to guide procurement selection. To date,

indirect approaches have predominately focused on matching project performance

outcomes with the relative merits of alternative procurement approaches in order to

select the preferred procurement mode. Fundamentally, this and similar approaches

are mostly described as the multi-attribute utility approach (MAUA).

1.3 Current research in procurement selection approaches

Since the 1970s, there have been around 900 studies of procurement systems

(Chang & Ive, 2002b). Love, Skitmore and Earl (1998), cited in Chang and Ive

(2002b), claim that these studies regard MAUA as the most appropriate technique for


examining the criteria of clients and the preferences of expert weightings for each

method in the most objective way. Table 1.2 (below) briefly summarises some of the

key MAUA-related studies and depicts the evolution of this approach from a simple

to a more complex method of scoring. (More details of each method are given in

Appendix A.)

Table 1.2 Categorical list of procurement approaches (Adapted from Ambrose & Tucker, 2000; Love, Davis, Edwards, & Baccarini, 2008)

Procurement selection approaches

Literature Description

Simple scoring or weighting

Nedo (1985); Franks (1990); Griffith and Headley (1997); NSW Department of Commerce (2006); Love et al. (2011)

Use of a rating system using client priorities for nine criteria

Multi-attribute utility approach (MAUA)

Skitmore and Marsden (1988); Bennett and Grice (1990); Singh (1990); Liu (1994); Chan (1995); Love et al. (1998)

Use of multi-attribute utility analysis based on NEDO with a rating system and weighting of client priorities, and including discriminant and concordance analysis; other works were based on Skitmore and Marsden (1988)

Delphi Approach Chan et al. (2001); Manoliadis et al. (2009)

Application of Delphi method in selection of procurement systems for construction projects

Analytical hierarchy process (AHP) - mathematical model

Alhazmi and McCaffer (2000); Cheung et al. (2001); Altabtabai (2002); Khalill (2002)

The relative weightings of the selection criteria are assessed using AHP (improves the objectivity of the weightings or ranking systems with MAUA or simple scoring method)

Expert System (Computer-based)

Brandon et al. (1990) – ELSIE; Moshini and Botros (1990) – PASCON; Kumaraswamy and Dissanayaka (2001) – Decision support system; Luu et al. (2005) – Case-based reasoning; Molenaar & Songer (2001) – Web-based decision support system

Use of computer expert system to evaluate and generate alternative procurement strategies based on client requirements, attitude towards project control and risk taking, and constraints of time and costs

Matrix-based Dell’Isola et al. (1999); SRD Consulting (2000); Construction Industry Institute, CII (2001); Sidwell et al. (2003); Ambrose and Tucker (2000)

A variety of matrix-based models designed to help select a procurement system based on scores of performance of procurement system, project outcomes, or project attributes

Framework-based Coorperative Research Centre for Construction Innovation, CRC (2008); Grimsey & Lewis (2009)

Generally, comprises of a series of logical steps to arrive at a procurement mode; includes simple weighting of client priorities and scoring procurement methods to achieve priorities

Before MAUA, the method of matching project criteria or client requirements

with the relative merits of alternative procurement modes was more simplistic, using

a simple ranking and scoring method for a limited number of procurement options

(Franks, 1990; Griffith & Headley, 1997; National Economic Development Office

(NEDO), 1985). Skitmore and Marsden (1988) were the first to apply MAUA to

procurement selection. The MAUA starts with a subjective ranking of criteria, which

are mindful of both the client’s requirements and the nature of the project, and which


are applied to a range of attributes that are considered to be important. Eight

attributes emerges as the more widely accepted attributes; namely: 1. speed (early

completion); 2. price or time certainty; 3. flexibility (to accommodate change in

design); 4. quality standard (aesthetics; compliance with specification); 5. complexity

(of building); 6. risk allocation (transfer of risk); 7. price competition; and 8.

responsibility (single point). The rankings are then rationalised and multiplied by a

utility factor which represents the extent to which a procurement method satisfies

each attribute. The procurement method with the highest weighted total is taken to be

the most appropriate method. MAUA reduces the issue of subjectivity with respect to

the weighting of criteria and is an improvement on the simple scoring method.

Since Skitmore and Marsden’s (1988) first application of MAUA, research into

procurement selection approach has focused on improving the objectivity of the

weightings and on refining or increasing the number of project performance

attributes or selection criteria. Various techniques for weighting and scoring project

criteria have evolved, and range from a more complex MAUA scoring method

(Ambrose & Tucker, 2000; Love, Skitmore, & Earl, 1998; Singh, 1990), to Delphi

methods (Chan, Yung, Lam, Tam, & Cheung, 2001; Manoliadis, Pantouvakis, &

Christodoulou, 2009). Matching key criteria with various procurement modes

subsequently became more sophisticated, with the development of computerised

decision-support systems – such as ELSIE, and Pascon – and a matrix-based

framework (Dell'Isola, Licameli, & Arnold, 1998; Sidwell, Kennedy, Bennet, &

Chan, 2003).

The MAUA approach reached its peak with the development of complicated

mathematical and statistical methods, such as the analytic hierarchy process (AHP)

and fuzzy logic, to include more quantitative and qualitative attributes [as seen, for

example, in the empirical research of Alhazmi and McCaffer (2000); Cheung et al.

(2001); Al-Tabtabai (2002); and Al Khalil (2002)]. With the wide range of

techniques and approaches available, and acknowledging the limitations of various

types of MAUA-based techniques, Ibbs and Chih (2011) went so far as to develop a

framework to help decision-makers systematically select a suitable MAUA-based

technique.

As well as exploring MAUA’s various techniques and other variations,

research has also focused on refining the types of project criteria used. This body of


research includes similar types of project attributes (while sometimes expressed in

different terms); however, the number and categorisation of these attributes varies in

different studies. More recent MAUA-based research has moved away from

modifying the generic factors – including speed, price or time certainty, flexibility in

design, quality standard, complexity, risk allocation or transfer, price competition,

and responsibility – to include a range of criteria that are of a different nature; the

latter include technological aspects, community involvement and natural disasters

(Kumaraswamy & Dissanayaka, 1998; Luu, Ng, & Chen, 2003). Some of these

selection criteria are summarised in Table 1.3 (below).

In brief, the original MAUA approach has developed to become increasingly

mathematical and complex, and requires the use of expert knowledge for its

implementation. Despite this, MAUA has not been able to resolve three key

weaknesses: issues related to its implementation, a charge of being tautological, and

weaknesses in risk assessment, which are further discussed in the next section.

Table 1.3 Examples of client priorities for procurement selection (Adapted from Love, Davis, Edwards, & Baccarini, 2008, p. 759)

Kumaraswamy and Dissanayaka (1998)

Level of design competition, level of price competition, economy, value for money, life cycle costs, cost certainty, speed, time certainty, urgency to complete project, urgency to commence construction, importance of intermediate milestones, aesthetic value, durability, innovations, quality assurance, construction risks allocation, design risks allocation, financial risk allocation, other risk allocation, need for mid-project design changes, need to be kept informed, need to be involved, need to assign single point responsibility, need to delegate decision-making, desire for good communication, health and safety concerns during construction, importance of planning, importance of controls, technology transfer/exchange, technology innovations, operational guarantees, design life certainty, maintainability, constructability, reduction in environmental impacts, disputes (and claims) minimisation

Luu, Ng and Chen (2003)

Client experience, client type, client’s in-house technical capability, client’s financial capacity, client’s willingness to take risks, client’s willingness to be involved, client’s trust towards other parties, client’s requirement for highly serviced or technically advance building, client’s requirement for aesthetic building, client’s requirement for on-time completion, client’s requirement for within-budget completion, client’s requirement for low maintenance cost, client’s requirement for low operational cost, client’s requirement for value for money, project size, project types, building construction type, project site location, unknown site risk factors, known factors likely to cause problems, usage of pioneering technology, market’s competitiveness, technology feasibility, regulatory feasibility, materials availability, experienced contractor availability, labour productivity, inclement weather, natural disasters, industrial actions, objection from local lobby groups, objection from neighbour, political constraints, cultural differences

NSW Department of Public Works (2006)

Design development flexibility, extent of design input by the agency, flexibility of scope resolution, ability to address complexity, ability to address uncertainty, ability to address the extraordinary, cost/time with brief quality, flexibility with the design brief, flexibility with scope, agency, design and technology change, impact of design change, brief/design realisation, risk/cost, package coord/interface risks, risk with design extra costs, designer continuity, contractor design responsibility, optimising life cycle costs, optimising maintenance design and defects minimisation, contractor maintenance responsibility, completion timing certainty, completion timing minimised, minimum time pre-contract, flexibility with timing changes, flexibility with cashflow control, early start to design, staged design allowed, early start to construction, staging flexibility, delay effect of one contract on others, capital cost minimised, end cost versus budget certainty, value for money for special projects, risk of contractual claims, extent of management/effort for agency for general projects, risk contingency in tender prices, minimising tender costs, minimising tender process costs quality certainty/outcomes/risk, quality of management, choice of contractors, availability of contractors, simplicity of contract, reliance of relationships, novation/relationship complexity


1.3.1 Weaknesses in multi-attribute utility theory

Despite the plethora of tools and techniques that have been developed,

MAUA’s application has been limited. Sweeney (2009, p. 61), having tested five

different projects using utility factors optimised using the Delphi approach, identified

issues with MAUA’s implementation; these include:

• The lack of a generally expert-agreed and accepted set of criteria

• Difficulty in identifying enough skilled experts to reliably contribute to the

scoring process

• Difficulty in reaching consensus on the set of utility scores for each

procurement method

• The lack of a comprehensive range of procurement methods in the research

• Narrow margins of difference in the scores produced (with this limited

differentiation leading to less reliable selection)

• Available analyses of a limited range of trial projects only

Due to the uniqueness of each project and the constantly changing needs of

clients based on internal and external demands, the criteria type and weighting will

invariably change (Love, Davis, Edwards, & Baccarini, 2008, p. 756). Thus, Love et

al. (2008) recommend that decision-makers should only make use of these selection

criteria as a guide to understanding the basic attributes of a procurement system, and

not as the sole basis for selection of the procurement method (Love, Davis, Edwards,

& Baccarini, 2008; Luu, Ng, & Chen, 2003). This might be due in part to the

fundamental weakness of MAUA that arises from its tautological nature.

From a scientific perspective, and notwithstanding its (above) technical issues,

a fundamental MAUA problem arises from its tautological nature. The latter is

associated with its notion of matching client requirements (project outcomes) with

the relative merits of alternative procurement modes in order to select the preferred

mode (Chang & Ive, 2002b; Teo, Bridge, & Jefferies, 2010). Most importantly, this

weakness creates a disconnect between the risk management plan and procurement

selection, renders the approach susceptible to non-economic influences, and makes

the achievement of VfM (in relative terms) an unlikely outcome.

In other words, the preferred procurement mode is being defined as a subset of,

or (effectively) in the same terms as, the desired outcomes of the project. It is,


therefore, tautological. A tautology is a statement of a relationship that is true by

logic, as in Popper’s (1959) p-q example; that is, if cause (read ‘procurement mode’)

and effect (read ‘desired project outcomes’) are defined in the same terms, or if cause

or effect are defined as a subset of each other, then the relationship is circular and

considered to be a truism that is not falsifiable. In other words, MAUA lacks

scientific maturity and amounts to a largely prescriptive approach that can be

decoupled from the risk management plan if deemed justifiable, and for reasons other

than an economic.

To further explain this phenomenon: Given that the project outcome and

procurement mode are effectively synonymous in MAUA, the client has effectively

and automatically selected the procurement mode upon determining the project

outcome. Although references can be made to data gathered as part of developing a

pre-existing risk management plan, this data can now be viewed in an ex post manner

and not as a determinate of the procurement mode; that is, the selection of the

procurement mode, and when regarded as one contract, automatically determines the

size of the project and (largely) the allocation of risks. In this case, there may be a

natural inclination to focus on more visible outcomes from the project on the opening

day of the asset, as opposed to benefits and/or cost that are difficult to assess across

the lifecycle of the asset (as previously mentioned in relation to both PPPs and non-

PPPs in Section 1.2).

A focus on these shorter term production benefits correspondingly diminishes

the concern for transaction costs incurred both internally as part of engaging the

market, and externally as transaction costs arising from ex post opportunism by the

private sector partner if there is any variation in the services required by government.

The major problem here is the danger of leaving the procurement decision open to

non-economic influences – either political or financial pressures – due to the non-

scientific nature of MAUA. This can be seen, in practice, from the broad focus on

limited visible attributes at the start of the operation of the new facility; that is, a

focus on short-term production benefits (such as time) or costs in absolute terms that

are associated with organising the whole project at the second-order level.

In summary, this weakness in MAUA results in a disconnect between any risk

management plan and procurement selection, and the approach is susceptible to non-

economic influences such as political and/or financial concerns which may not


always equate to a purely economic imperative across the whole-of-life of the asset.

On the other hand, if prepared to accept a tautology, the MAUA can be argued to be

a useful tool in ex post justification of a particular procurement mode and in

delivering VfM, albeit in terms of a narrower definition of the latter.

Consider, for example, an extreme case in which a government agency is faced

with the need to satisfy a dominant or key project outcome that is visible on the

opening day of the asset; say, for illustrative purposes, this visible outcome is ‘the

earliest time for starting operations’, and ‘the earliest time for construction

completion’. In this case, the government agency may proceed from this outcome

(effect) upstream to the procurement mode (cause) that is selected on the basis of the

procurement mode’s relative merits; say, for illustrative purposes, Managing

Contractor.

The selected procurement mode might then proceed to deliver the desired key

outcome, but at the same time may deliver lower whole-life outcomes or benefits

than those achievable by other procurement modes, given the outcome or benefit

trade-off that exists among different procurement modes (Ive & Chang, 2007). The

selected procurement mode may well also represent an inferior approach to

economising on the sum of production costs and transaction costs, given a lack of

attention to production improvements resulting from incentive alignment and costs

arising from incomplete contracting (Chang & Ive, 2002b; Sweeney & Duffield,

2006).

Despite this, the procurement mode selected may still succeed in terms of the

key performance outcome. Logically, then, this should render the selected

procurement a success, as all other benefits beyond the key performance outcome

should be set at near zero. On the basis that the selected procurement mode is largely

the only mode able to deliver the key outcome required, then all other procurement

modes will result in value deterioration (value curve moving towards the right, as

seen in Figure 1.1) relative to the benefit/cost position achieved by the selected

procurement mode. Hence, accepting a tautology in this case is useful, as the

government agency responsible for the delivery may be judged substantially on

whether or not the procurement mode was successful in delivering the key outcome.

Economically, however, this scenario creates a narrow or nominal approach to

VfM, in so far as VfM is being framed in terms of whether or not the selected

Chapter 1: Introduction

procurement mode delivered the key p

the start of operations

Thus, this agency should then be relieved of having to demonstrate VfM in relative

terms, as previously defined.

based on MAUA that

relative merits of alternative procurement

mode.

In addition to undermin

seeks to identify risks

and educational guess

required. Subsequent

example, by a consultant rather than the contractor

guessing of the extent to whic

lack of information and a great

project, there is a great amount of risk involved in construction. E

a study conducted by Fly

and fifty-eight transportation infrastructure projects

to USD 90 billion, indicate that in nine out of ten transportation projects, the actual

construction costs had been underestimated.

In summary, the

risk analysis, are:

procurement mode delivered the key production or performance outcomes vis

the start of operations; that is, VfM is measured in nominal rather than


defined. Figure 1.2 depicts the notion of VfM in nominal terms

based on MAUA that seeks to match client requirements (project outcomes) with

relative merits of alternative procurement modes in order to select

Figure 1.2: MAUA and VfM in nominal terms

undermining any risk analysis being carried out, current practice

risks at an early stage of the project when information is lack

educational guesses concerning the nature and extent of potential risk

. Subsequent risk assessment, if conducted by the wrong experts

example, by a consultant rather than the contractor – might then involve

he extent to which the potential risk events might occur.

lack of information and a great number of uncertainties in the conceptual phase of

is a great amount of risk involved in construction. E

nducted by Flyvberg, Holm and Buhl (2002) with a sample of two hundred

eight transportation infrastructure projects from five continents

90 billion, indicate that in nine out of ten transportation projects, the actual

truction costs had been underestimated.

In summary, the three key weaknesses of current practice in MAUA

13

performance outcomes visible at

rather than relative terms.


picts the notion of VfM in nominal terms,

to match client requirements (project outcomes) with the

modes in order to select the preferred

risk analysis being carried out, current practice

information is lacking,

the nature and extent of potential risks are

if conducted by the wrong experts – for

then involve second

occur. Because of the

of uncertainties in the conceptual phase of a

is a great amount of risk involved in construction. Empirical results of

a sample of two hundred

from five continents amounting

90 billion, indicate that in nine out of ten transportation projects, the actual

practice in MAUA, including


1. Issues with implementation, arising from a lack of generally agreed and

accepted criteria and difficulty in reaching consensus on utility scores for each

procurement method

2. Risk analysis being overshadowed by a main concern or criterion; that is,

matching key opening day outcome(s) to a procurement mode expressed at the

organisational or second-order level undermines a full or effective risk

analysis, and results in activities not being allocated to the party with the best

resources to manage risk

3. Too focused an approach to risk analysis in attempting to identify individual

risks (for example, guessing when information is missing at the schematic

stage and consultants attempting to assess size or value of activity risks –

issues that are better understood by other parties such as contractors).

In order to elaborate on these weaknesses, the next section discusses the

procurement selection approaches being applied in practice in the Australian context.

1.4 Procurement selection approach in Australia

Given the difficulties in estimating production and transaction costs, and the

weaknesses of current research associated with MAUA, it may not be any

coincidence that current practice in the procurement of new infrastructure is mainly

at a prescriptive level and lacking in scientific maturity. For example, in the

Australian context, the National PPP Policy Framework seeks to enable public and

private sectors to work together to improve public service delivery through private

provision of public infrastructure and related services, and requires jurisdictions to

apply the framework to the procurement of PPP projects. The framework recognises

that no delivery method is presumed to be more efficient than another and, in its

Volume 1 Procurement Options Analysis (POA), the framework provides an

approach to assessing the viability of PPP delivery against other procurement

methods. The approach in the POA comprises of five steps:

• Step 1: Data gathering (objectives; risks; unique project characteristics; agency and market capability)

• Step 2: Shortlist delivery models (consider suitability of: PPP; Alliance; Managing Contractor)

• Step 3: Validation (what precedent exists for the project? What does the market think?)


• Step 4: Delivery model options analysis (Which model best achieves requirements and objectives and reduces risk?)

• Step 5: Preferred delivery model (structure preferred model; consider risk; approve; conduct gateway review)

Step 1 of the POA identifies any risks that might occur in the project as part of the

data gathering process and assess the appropriate risk allocation or transfer to

government or private party. Step 2 uses four criteria – namely, scale, scope, whole-

of-life opportunities and risk – to begin to direct the procurement in one of three

main directions; that is, PPP or Alliance, Managing Contractor, or other procurement

mode. Step 4 evaluates the shortlisted procurement modes against project objectives,

using data gathered in Step 1. The POA does not prescribe an evaluative method to

be used in Step 4; however, it does recognise that some departments in Australia

have developed approaches (such as assessment matching, weighted tables and other

tools) that reflect their particular project requirements. Some of these approaches are

summarised in Table 1.4 (below).

Table 1.4 Australian states’ procurement guidelines

State Procurement guidelines NSW NSW Public Private Partnerships Guidelines (NSW Treasury, 2012) QLD Public Private Partnerships – Policy (Queensland Government - Department of

Infrastructure and Planning, 2008a); and Value for Money Framework, Public Private Partnerships guidance material, Supporting document – Business case development (Queensland Government - Department of Infrastructure and Planning, 2008b)

SA Procurement Policy Framework (State Procurement Board, 2013) VIC Investment Lifecycle and High Value/High Risk Guidelines (Department of Treasury and

Finance Victoria, 2012) WA Infrastructure Procurement Options Guide (Centre for Excellence and Innovation in

Infrastructure Delivery (CEIID), 2010)

The evaluative methods in Step 4 of the National PPP Policy Framework and

related state procurement guidelines in Table 1.4, use semi-quantitative measures

(rankings or perceptual scales and non-monetary criteria) of benefits/costs of

alternative procurement modes. In plain terms, these evaluative methods seek to

systematically match client requirements (project outcomes) with the relative merits

of alternative procurement modes, in order to select the preferred mode. These

methods are akin to MAUA. However, due to the weaknesses of MAUA (discussed

above), other factors such as time and cost may be driving the selection of the

procurement mode, and the current practice is susceptible to non-economic

influences and undermines any risk analysis that is being conducted. It may not be a


surprise then that the determination of when or when not to go down a PPP path

remains very much a vexed question.

1.5 PPP or non-PPP delivery

Despite having led the Private Finance Initiative (PFI) and having undertaken

more PFI or PPPs than any other developed country in the world, a report from the

National Audit Office in the UK noted, amongst its key findings, that "There is no

clear data to conclude whether the use of PFI has led to demonstrably better or worse

value-for-money than other forms of procurement" (House of Lords, 2010, p. 6).

This lack of comparative data creates uncertainty surrounding the relative merits of

PPPs versus other non-PPP procurement modes. This is being further exacerbated in

the UK by the lack of transparency and accountability in PPPs (Thomas, 2011), to

the extent that ‘currently the form of public accounts means that the overall costs of

infrastructure are shown in a number of different budgets and there is no single point

of control for these budgets’. This situation is most acutely felt in the face of cuts in

recurrent spending in conjunction with a funding model that places PPPs and non-

PPPs in the same budget, and competing for similar resources.

Although PFI are being promoted in the UK as a means of reducing ongoing

costs in the contract, Thomas (2011) indicates that the debate surrounding

procurement selection may become even less reasoned and skewed by prejudice

and/or short term considerations. Similarly, in a review of all key international PPP

evaluative studies, including studies in Australia, Hodge and Greve (2009) conclude

that the review of the merits of PPPs is mixed, and indicate contradictory results with

respect to their effectiveness and VfM, as illustrated in Table 1.5.

Table 1.5 Mixed PPP reviews and evaluations

Supporting case for VfM of PPP Questioning whether PPP achieves VfM

Unclear/neutral/mixed reviews

Generally: Arthur Andersen & LSE Enterprise (2000); Savas (2000); NAO (2000); Macdonald (2002); Pollitt (2002); Grimsey and Lewis (2004); Pollitt (2005); Allen Consulting Group (2007); and Raisbeck, Duffield and Xu (2010)

Generally: Greve (2003); Edwards et al. (2004); Shaoul (2005); Boardman et al. (2005); Pollock et al. (2007); and Leviäkangas (2007); NAO (2011, p. 6); Thomas (2011)

Generally: Walker and Walker (2000); Teisman and Klijn (2001); Fitzgerald (2004); Ghobadian et al. (2004); Hodge (2005); and Blanc-Brude et al. (2006)

Specific to either hospitals, schools, or correctional facilities: Auditor-General of NSW (2006); NAO (2010)

Specific to either hospitals, schools, or correctional facilities: Bloomfield et al. (1998); Pollock et al. (2002); Audit Commission (2003)

Specific to either hospitals, schools, or correctional facilities: NAO (2003)


Bingham and Felbinger (2002) describe four levels of PPP/VfM evaluation

from the strongest (Level 4) to the least strong (Level 1) as follows:

• Level 4: Experimental control group (explanatory)

• Level 3: Comparison groups

• Level 2: Time series (for example, gateway reviews) and

• Level 1: Before or after comparison (for example, reconciliation against business case or Public Sector Comparator)

Hodge and Greve (2009) note the limitations of both the evaluations supporting

VfM, and those questioning VfM in PPPs. Specifically, they note that the empirical

studies are mostly Level 1 evaluations and, of the Level 3 evaluations conducted, the

focus is on the performance attributes in the construction stage and the period up to

the opening or first day of the facility; in other words, there is a lack of Level 3

comparative evaluations through the operational stages. Hence, the evaluations of the

comparative merits of alternative procurement approaches are not conclusive.

In addition, it is not surprising that there appears to be uncertainty in terms of

early identification of projects that are suitable for a PPP. This early identification of

suitability as part of a preliminary evaluation is critical to avoiding the unnecessary

expense of travelling too far down the PPP path (which includes the development of

the Public Sector Comparator (PSC) as part of the development of a PPP business

case).

The House of Lords Select Committee on Economic Affairs (2010) questions

VfM from PPPs in certain types of infrastructure. This is based on key VfM

determinants concerning the size of transaction (associated with competition issues

which impact on prices and innovation) and flexibility (associated with issues

concerning hold-up arising from changes, and manifested in factors such as variation

flows and restrictions to asset management). On the issue of considering the long-

term stability, size and complexity of projects in determining their suitability for

private finance, the House of Lords (2010, pp. 31, para. 152) recommends that,

The projects most suitable for private finance are those where the requirements can be clearly specified at the outset and which are of a size that consortia of private sector companies can take on their balance sheets.

KPMG and University College London (2010) argue that robust evidence-

based decision-making can only be produced through comparable and quantitative

data on the performance of various operational arrangements on a like-for-like basis.


Thus, comparable quantitative data on cost and performance across PPPs and non-

PPPs needs to be systematically collected to facilitate this apple-with-apple

comparison. This data can then be used to develop and test a theory that both

explains cost and performance differentials, and equates to Bingham and Felbinger’s

(2002) Level 4 type evaluation. In this way, we can move towards a state of scientific

maturity and use the developed and tested theory to guide the selection of

procurement in future projects. Such comparable and quantitative cost and

performance data does not currently exist; however, KPMG and University College

London (2010) have begun to generate such data on the performance of various

operational arrangements across both PPPs and non-PPPs.

In summary, the difficulties in measuring or estimating production and

transaction costs seriously undermine attempts to develop empirical or estimated

comparators to objectively determine which projects are suitable for a PPP path.

Furthermore, the tautological nature of current procurement selection approaches in

major infrastructure, and the uncertainties in the criteria for selecting a suitable

procurement approach (PPP or non-PPP), indicate that the existing research provides

practitioners with insufficient guidance in selecting a procurement approach that is

more likely to deliver VfM in relative terms. Moreover, even in a world in which all

historical cost and performance data for extant infrastructure were available, there is

still no procurement model that could predict the best empirical cost/performance

ratio in the next new project as this would undoubtedly incorporate new and unique

features unaccounted for in any established database.

It is clear, therefore, that there is an important opportunity and need to develop

a new procurement decision-making model that addresses MAUA’s weaknesses and

can be used to determine the procurement of major infrastructure, including the

identification of projects that are suitable or unsuitable for PPP. Progress can now be

made in procurement by developing an alternative approach to determine the effect

of procurement selection on VfM; that is, by developing indirect or non-monetary

measures of the benefits/costs of alternative procurement modes, (particularly at an

early stage of project development, and deploying the New Institutional Economics

(NIE) theories and Resource-based theory (RBT), (as illustrated in Figure 1.3) that

begin with project attributes at a first-order level.


Figure 1.3: New model versus current approach

The NIE paradigm belongs to an emerging field of micro-economics, and is

closely related to the make-or-buy decision. The RBT, on the other hand, is a

dominant theory in the Strategic Management field. NIE is broadly divided into two

categories: incentive alignments and economies of transaction costs (as shown in

Figure 1.4). Transaction cost theory (TC) and Transaction cost economics (TCE),

both Nobel prize-winning theories, belong to the measurement and governance

category, while both Property rights theory (PRT) and Principal-agent or Agency

theory (PAT), belong to the incentive alignment category.

Figure 1.4: New Institution Economics Theories

Source from “The Economic Institutions of Capitalism: Firms, Markets, Relational contracting” by O.E. Williamson, Copyright © 1985 by Oliver E. Williamson. Reprinted with permission of Simon &

Schuster Publishing Group from the Free Press Edition. All rights reserved.

The Transaction cost approach emphasises transaction cost economising of

transactions within a firm (Williamson, 1981, p. 548), and is related to the effective

adaptation and the elimination of waste at the first-order (or transaction) level

(Williamson, 1991, p. 276). Williamson (2000, p. 597) categorises both PRT and

Measurement

Incentives

Transaction cost

Efficiency

Governance (Ex post governance)

Principal-agent or Agency theory (Ex ante incentive alignment)

Property rights (Ex ante incentive alignment)

Procurement

Mode

Risk Analysis Project

Attributes

New Institutional Economic

Theories and Resource-

based Theory

Procurement Mode

Government Resources

Private Sector Resources

Project Attributes

Project Performance Outcomes

(long-term)

VfM

Relative Terms


(short-term)

VfM Nominal Terms


transaction cost approaches as first-order level and further extends these to second-

order level economising (dominated by TCE) to select an appropriate governance

structure (either market, hybrid or hierarchical). Finally, Williamson (2000)

categorises PAT, which is concerned with efficient risk alignment, as third-order

economising.

In summary, current research and practice in procurement selection appears to

lack scientific rigour as a sound basis to secure VfM across whole-of-life of the asset.

Consequently, there is scope for developing a procurement decision-making model

that deploys NIE theories and RBT, and which starts from the first-order

measurement and characteristics of the project and provides a clear connection

between risk analysis and project attributes. Moreover, in the context of determining

the suitability or otherwise of PPPs, the model can potentially explain the effect of

procurement on Expressions of Interest (EoI) as a key indicator of competition and

flexibility (the scope of contract and/or the ability of the contractual arrangement to

deal with variations in the works) which, in turn, is a critical proxy of pre-contract

and post-contract market failure. The development of EoI as a dependent variable

and critical proxy of market failure and VfM is further discussed in Section 3.3.

As mentioned earlier, competition and flexibility have been found by the House

of Lords in the UK to be critical in crystallising not only lower prices, but also

innovations that can increase utility and reduce whole-life costs. At the same time,

the House of Lords has also found that a balance needs to be struck, as too much

competition becomes counterproductive and can be an indicator of market failure

post-contract; that is, the market can be signalling their gains from hold-up arising

from variations in the PPP long-term contract. Thus, if there is need for a high level

of flexibility that cannot be managed effectively by the contract, and which creates

the prospect of costly variations (that have the effect of increasing the attractiveness

of the project to the market), this undermines PPP as an efficient mode of

procurement.

In PPPs then, EoIs are reflective of both the prospect of sufficient competition

and the attenuation of hold-up delivered by the approach to procurement taken. On

this basis, a procurement decision-making model derived from NIE offers the

potential to identify whether projects are suited or not suited to PPPs and, in so

doing, raises the chance that the procurement approach selected (either PPP or non-


PPP mode) can deliver superior utility and whole-life cost outcomes; that is, VfM

relative to the alternative competing modes.

1.6 Research aim

The aim of the research, therefore, is to: Develop a first-order procurement

decision-making model based on NIE theories and RBT, and to test this model using

EoI as a key indicator of competition and flexibility, and a critical proxy of market

failure, and VfM. More specifically, this research uses an integrative framework of

vertical integration – comprised of TC, TCE and RBT – as the core of the new

procurement model. It also includes other theories from the NIE – namely, PRT and

PAT – to determine the suitability of a PPP or non-PPP approach. The aim of the

model is to deliver superior utility and whole-life cost outcomes; in other words,

VfM relative to the alternative competing modes.

1.7 Research objectives

In order to address the research aim, the objectives of the research are to

develop a first-order procurement decision-making model derived from NIE theories:

• Objective 1: to integrate TC, TCE and RBT (in a first-order stage) with respect

to the make-or-buy decision to determine the size and scope of externalised

activities within a piece of infrastructure

• Objective 2: to determine (in a second-order stage) an optimal bundle of

externalised activities which seeks to: i) minimise bureaucratic costs through

greater single-point contact with a private sector firm across a wide range of

activities (that is, the transfer of control to a private sector firm based on PRT);

and ii) minimise hold-up through the greater use of internalised management (or

agents) and control over private sector firms

• Objective 3: i) to deploy TCE (in a third-order stage) to determine the most

efficient type of exchange relationship for each externalised bundle or contract

(to further minimises external transaction costs); and ii) to deploy PAT to create

greater alignment between principal and agent (which also reduces transaction

costs, and promotes production benefits beyond this cost reduction)

• Objective 4: to develop EoI as a key indicator of competition and flexibility

(according to experiential evidence from the House of Lords in the UK), as a


critical proxy for market failure and VfM in the context of the suitability or

otherwise of PPPs.

The research hypothesis, therefore, is:

Actual competition is expected to be in the optimum range of competition (5 to

8 EoI) in cases in which actual procurement matches the theoretical procurement

(informed by the first-order procurement decision-making model) and outside the

optimum range in cases where there is an appreciable mismatch between actual

procurement and the theoretical procurement.

The following section gives the nature and extent of theoretical contributions to

knowledge, therefore providing justification for the research. The specific details and

scope of contributions to knowledge are given in the concluding Chapter Six.

1.8 Justification for research

1.8.1 Theory

Literature relating to the dominant (and Nobel prize-winning) theory of

procurement from the NIE field (that is, TC and TCE), and to the related dominant

Strategic Management theory (that is, RBT), has been substantially ignored in the

construction literature, and its application in construction-related sectors is only just

becoming evident [for example, in Bridge and Tisdell (2004), Bridge (2008), Chang

and Ive (2007a), Jin and Doloi (2008) and Sweeney (2009)]. Duffield (2009) affirms

the potential for TCE to contribute to the discussion of the most appropriate

procurement strategy. Much more significantly, however, the chief protagonists of

transaction costs (TCE) and production costs (RBT) are all pointing towards the

integration of both perspectives in order to develop more accurate explanations of

key aspects of procurement, including the make-or-buy decision (read risk

allocation).

A review of the construction economics literature reveals that Bridge and

Tisdell (2004) and Bridge (2008) represent the only research that has developed and

empirically tested the integration of TC, TCE and RBT in a single framework with

respect to the make-or-buy decision (risk allocation). A number of leading scholars

in the field of Construction Management and Economics have supported Bridge and

Tisdell's approach to integrating TCE and RBT; for example, Kumaraswamy et al.

(2008), Walker (2007) and Bröchner (2008, 2011). Indeed, Ball (2007, p. 221)


considers that Bridge and Tisdell have developed an "ingenious" approach to

drawing together the TCE and RBT literature.

Moreover, Chang (2013) in his recent critique of justifying the approach to

combining TCE and RBT in the procurement of PPP has effectively conceded that

these theories can be integrated. This view supports the approach taken in the

procurement model (as explained in Section 2.8.1b). Brahm and Tazijan (2013) also

cite Bridge and Tisdell’s approach as leading edge in terms of its integration of the

theories of the firm; at the same time and notably, Brahm and Tazijan (2013) remain

silent with respect to Chang’s critique of Bridge and Tisdell’s approach to

integration.

Furthermore, Bridge, Tiong and Wang (2010) propose that Bridge and

Tisdell’s framework can be applied in the context of infrastructure procurement.

However, in order for the framework to be deployed in practice, it requires major

development to match the full spectrum of procurement modes, and refinement of the

description of TCE and RBT attributes. Bridge, Wang and Tiong (2010, p. 308)

conclude that, by applying the integrative framework to the procurement of

infrastructure, there will be “a substantial advancement and very significant

contribution to the knowledge in the field of procurement of major infrastructure”.

1.8.2 Research method

This research represents the first empirical research to apply TC, TCE and

RBT in relation to the differential capability and competence of the focal firm and

the market firms with respect to an entire new infrastructure project; that is, the first

empirical research to develop and test the integrative framework as a means of

assessing firm boundaries or make-or-buy decisions within a firm and with regards to

a new project.

In achieving the above, this research: i) further develops the approach to

measuring the activity as the unit of analysis in the first-order stage of the new

model; ii) operationalises the theoretical constructs underpinning the framework,

such as the measurements of TCE (asset specificity, frequency and uncertainty) and

RBT (value, rarity and imitability) variables; iii) develops the combination of the

value variable in RBT and the frequency variable in TCE; iv) develops EoI as the

dependent variable, a key indicator of competition and flexibility, and a critical


proxy of market failure and VfM; and, finally, v) develops a method to test the

procurement decision-making model, including the integrative framework.

1.8.3 Practical

Fundamentally, the procurement model is designed to focus the analysis on the

production activities of design, construction, operations and maintenance at a first-

order level, and predicts a procurement approach by deploying production

cost/benefit theory from the field of Strategic Management, and theories concerning

transaction costs from NIE. In this way, it identifies a procurement approach that

represents an efficient configuration of risk allocation, bundle(s) of externalised

activities, and the nature of the external exchange relationship with each externalised

contract(s) for the project concerned.

The key departure of the procurement model from current practice or MAUA is

its analysis of make-or-buy decisions within the project at the individual activity or

first-order level, rather than the MAUA analysis of organisation or management

across activities which are considered second-order level. Through this analysis, the

procurement model also addresses MAUA’s weaknesses and provides a more

reliable method of risk allocation. This model also provides a clear connection

between data gathering (including risks associated with project characteristics in

conjunction with the relative capability and competence of government versus

private sectors relative to the project) and the potential to be incorporated into PPP

guidelines; for example, Australia’s National PPP Policy Framework – Volume 1:

Procurement Options Analysis (POA) and similar PPP guidelines around the world.

1.9 Methodology

1.9.1 Methodology versus methods

‘Methodology’ represents the philosophical approach and underlying

assumptions deployed in the research, and contrasts with the term ‘methods’ which is

associated with the procedures or steps employed to collect and analyse data. The

various theories employed in this research belong to the Social Sciences. It is

important for research in the Social Sciences, more so than for research in the

Natural Sciences, to articulate its ontological position and epistemological approach

and assumptions. In simple terms, the main distinction between methodology and

methods, is as follows: ‘methodology’ represents the theory or the broad principles


or philosophy with respect to the conduct of the research and the application of

theory; ‘methods’, on the other hand, are the procedures or steps used to collect and

analyse data during the course of the research (Jayaratne & Stewart, 1991).

1.9.2 Ontological perspective – Realism

Like most other research in the Social Sciences, this research adopts the

ontological perspective of realism in its treatment of the theories deployed. Realism

acknowledges that reality is objective and that theories of science help to ‘make

sense of what would otherwise be inscrutable or unmeaning empirical findings’

(Kaplan, 1964, p. 302). These observations comprise of patterns or sequences in

observable events. These events exist independently of one’s identification and are,

therefore, objective and realistic; observations of such events are considered as

empirical evidence. Knowledge can only be gained if these observations manifest in

a form or pattern, rather than as a series of unrelated or irregular events. It is the

challenge in this research to discover the science and theory to explain these events.

In the same vein, and in relation to the topic of this research, Williamson has been

emphasising the operationalisation of TCE to test and develop ‘a predictive theory of

economic organisation’; that is, to discover the regularities or the ‘logic of the match

where transactions are aligned with governance structure in the most transaction-cost

economising manner’ (Williamson, 2005a, p. 6).

1.9.3 Epistemological position

Philosophically, epistemology is defined as the possible ways of gaining

knowledge of social reality, whatever it is understood to be (Blaikie, 2009, p. 8).

Construction Management research has been criticised for the dominance of

positivist epistemology and the preference for quantitative methods such as statistical

analysis and modelling (Seymour, Crook, & Rooke, 1997). Seymour, Crook &

Rooke (1997) argue that given the ‘object’ of the study in Construction Management

research are mainly people; this requires that more interpretive methods be

employed. In response, Runeson (1997, p. 299), argues that this critique on research

method is “anti-scientific” and defends positivism as still “the best way to reduce

subjectivity” and “instrumental in substantial advances in science”.

This debate regarding positivism or interpretivism in the literature indicates a

confusion in the epistemological positions adopted in Construction Management


research (Dainty, 2008). Furthermore, within the Social Science literature, positivism

has been criticised for not being able to understand unobservable factors – for

example, deep structures in society – and does not satisfactorily explain causal

relationship (Brown, 2007). Given that Construction Management draws theory from

both the Natural and Social Sciences; epistemologically, therefore, its research can

include both positivist and interpretive approaches. This coincides with the concept

of ‘critical realism’ proposed by Bhaskar (1978) in his seminal work which combines

positivism and interpretivism in its theoretical assumptions, and which stands out as

a suitable epistemological position. Bhaskar (1978) explains that a critical realist

perspective provides causal explanations similar to positivism, and understands the

nature of social reality through the process of interpretivism.

Based on the concept of critical realism, both observable and unobservable

events or characteristics – such as characteristics concerning negative opportunistic

behaviour (TCE) and characteristics concerning differential capability and

competence (RBT) – can be measured. The sequential steps involved are generally:

identification of a phenomenon; construction of explanations or propositions;

empirical testing and validation of the proposition or alternative propositions; and,

finally, explanation of the generative mechanism of the phenomena (Bhaskar, 1978,

p. 4). These steps provide a structure for understanding and interpreting the data, and

clarify both the directions for testing the model, and the linkages between theory and

data.

1.10 Delimitations of research

Bounded by the research aims and objectives, the research findings are

delimited in terms of the sectors involved, the capital value of projects, their location,

and their timeframe.

1.10.1 Sector

In response to the research aim, the focus of the research is on the procurement

of public sector infrastructure assets. Broadly, infrastructure can be defined

according to the Oxford dictionary (Infrastructure, 2013) as ‘the basic physical and

organisational structures and facilities, such as, buildings, roads, and power

supplies, which are needed for the operation of a society or enterprise’.

Infrastructure is, therefore, critical in driving economic development and social


enhancement. More specifically, infrastructure assets refer to the physical structures,

facilities and networks which provide essential services to the public; for example,

economic support services can include transportation structures, energy and utility

companies, communication entities, and social services such as educational facilities

and hospitals (Chambers, 2007).

In the Australian context, approximately AUD 169 billion is required to

finance proposed state-wide projects on the infrastructure priority list, and more than

500 other projects are not included due to a low benefits/costs ratio (Infrastructure

Australia, 2008). Road and rail projects account for more than half of the critical

projects, amounting to approximately 90 percent of the priority list. In the 2009-2010

budget, the Australian government committed to spending AUD 27.7 billion on

expanding and improving land transport infrastructure, and AUD 14.7 billion on

social infrastructure, including healthcare services.

Given the importance of transportation networks and healthcare services in

Australian major infrastructure procurement, this research is specifically delimited to

the construction of new roads, highways and bridges in the road sector, and to the

construction of new hospitals and healthcare facilities (not including refurbishment

or information technology projects).

1.10.2 Capital value of project and location

Projects with a minimum of AUD 50 million can be considered as major

infrastructure. This threshold has been indicated by the recommendation [in the

National Public Private Partnerships (PPP) Policy Framework and National PPP

Guidelines (2009)] to assess any project with a capital cost greater than AUD 50

million in terms of its suitability for PPPs. In addition, preliminary data collection of

projects in budget papers from five mainland states in Australia – New South Wales,

Queensland, South Australia, Victoria and Western Australia – indicates a relatively

large proportion of projects below AUD 50 million in project value; this could

indicate that this market is sufficiently efficient and price competitive. On this basis,

and with regard to the research aim of preventing market failure in major

infrastructure procurement, it seems logical to limit the scope of research to projects

from the five mainland Australian states that are worth a minimum of AUD 50

million.


1.10.3 Timeframe

The period between July 2005 and June 2010 inclusive, which includes the

period before and after the GFC in 2008, has been selected as the timeframe for this

study. It provides a once-in-a-lifetime opportunity to surface the effects of

procurement and tendering policy and practice on competition in the context of a

significant change and swing of EoI towards public sector work due to a

disproportionate downturn in private sector work that can be observed pre- and post-

GFC.

In summary, the scope of the research includes any road and hospital (or

healthcare) project completed or proposed (current, deferred or discontinued) with a

capital value over AUD 50 million, and in which EoIs were established in the five

year period between July 2005 and June 2010 inclusive.

1.10.4 Contractor sample

Based on the above limitations on the type and size of projects, it follows that

the targeted sample of contractors includes civil contractors and building contractors

operating in the five mainland Australian states, who are capable of constructing

infrastructure projects of value greater than AUD 50 million in the period between

July 2005 to June 2010 inclusive.

1.10.5 Summary of delimitations

The scope of the research is summarised as follows:

1. The procurement of road and hospital (or healthcare) infrastructure projects in

five mainland Australian states: New South Wales, Victoria, Western

Australia, Queensland and South Australia

2. Any road and hospital (or healthcare) project, completed or proposed (current,

deferred or discontinued) with a capital value over AUD 50 million, in which

(at least) EoIs were established in the period between July 2005 and June 2010

inclusive

3. Civil and building contractors capable of projects worth AUD 50 million or

more


1.11 Thesis outline

Following this chapter, Chapter Two provides an overview of the thesis,

summarises the fundamental concepts of each NIE and RBT theory and, based on

empirical research, highlights the strengths and weaknesses of each theory in

explaining key aspects of procurement; namely, make-or-buy decision, level of

bundling, and nature of exchange relationship. Chapter Two also examines the

progression of the development of the theory, and gives the current status of the

integration of TCE and RBT. Finally, to establish the knowledge gap, the literature is

examined for applications of NIE and RBT theories in relation to the key aspects of

procurement in construction, and justifies their potential for application to key

aspects of major infrastructure procurement.

In Chapter Three, the first-order procurement decision-making model, which

addresses the weaknesses of current research and practice, is developed. The

applications of NIE theories, including the integration of TCE and RBT, are

incorporated into the key steps of the model, and these steps are described in detail.

Finally in this chapter, EoI as a valid and reliable indicator for VfM (in terms of

avoiding market failure), and the procurement-competition/flexibility hypothesis to

test the model are developed.

Chapter Four discusses the approach to testing the model, and documents the

processes involved in the design and administration of both the survey and case study

research methods. With regards to the survey method, the design of the data

collection instruments – mainly, a survey of major road and health projects and a

nationwide survey of civil and building contractors – are detailed, including the

specific details of the sampling, administration, response rates and coding of

submitted data. With regards to the case study method, the approach to selection of

case studies, the steps in designing the case study questionnaire instrument – which

involves operationalisation of RBT and TCE variables and includes the method of

selection and administration of case studies – are presented in detail.

The analysis of the survey of road and health projects and the results of the

case studies are presented in Chapter Five. The analysis explores and establishes the

relationships or effect of key aspects of procurement on competition and flexibility.

Subsequently, the outcomes of deploying the model in all case studies are evaluated

against the hypothesis. The results from the survey of civil and building contractors


are then used in theoretical competition to validate and corroborate the results of the

hypothesis testing.

Finally, Chapter Six summarises the conclusions drawn from the results of the

hypothesis testing and survey results; explores the implications of the findings for

practice; discusses the contributions of this study to both theory and research

methodology; explains the research limitations; and finally, provides

recommendations both for further research and implementation of the findings.

1.12 Summary

This chapter began with a critical analysis of current issues in the practice of

procurement selection and justified the rationale for a first-order procurement

decision-making model by identifying the research gap in the literature and

inadequacies in practice. It then presented a brief description and justification of the

research methodology, the delimitations of the research, and an outline of the thesis.

The weaknesses of MAUA indicate that there is scope for the development of a

procurement decision-making model that starts with a consideration of the

characteristics of a project, and provides a clear framework for the analysis of risk

and associated project attributes. It is apparent that current practice is lacking the

scientific and economic theory to inform procurement, and fails to consider the long

term benefits/costs that affect VfM. In addition, the procurement selection is at the

project level; that is, at a second-order level of analysis. Furthermore, there is no

mention of ways in which to achieve economies in aspects of make-or-buy, no

mention of bundling, or of the nature of the exchange between client and the

contractor.

In order to move beyond current weaknesses in research and practice, a

comprehensive procurement selection guide can have the conditions surrounding the

project as the starting point of analysis as its focus, rather than the project

performance outcomes. These conditions include the technological and physical

attributes of the project, and the capabilities and competencies of both the

government and the private sector relative to the project. The project conditions

represent what is to be measured, while a priori theory guides how these conditions

are measured and the manner in which these measurements inform procurement

selection.


Having established NIE theories and RBT as the core of the a priori theory in

this research, the relevance of TC, TCE, RBT, PRT and PAT in the development of a

procurement decision-making model in the construction industry is studied in detail

in the following chapter. The review of each theory begins with a literature review,

an overview of the theory, and related significant literature. Key points from this

literature are then highlighted, and the relevance and application of each theory with

respect to aspects of infrastructure procurement is explained.

Chapter 2: Literature review 33

Chapter 2: Literature review

2.1 Introduction

As seen in Chapter 1, procurement selection is compromised by limitations in

its current research and practice, a lack of empirical data, and weaknesses in MAUA.

In this chapter, it will also be seen that the classical economics theory of production

does not fully account for the complex nature of construction procurement. This is

because it relies on simple market exchange or spot transactions, and pays little

attention to a range of transaction costs that might promote internalisation or vertical

integration arising from the desire to avoid post-contract market failure. In contrast to

classical economics that does not recognise opportunism, and only considers the

firm’s production function in technological terms, the NIE theories might be able to

better explain construction procurement by using alternative modes of efficient

governance structures.

In order to address Research Objectives 1, 2 and 3, this chapter delineates and

connects key aspects of procurement and the various theoretical positions from the

extant literature associated with the emerging NIE theories; namely, Transaction cost

theory (TC), Transaction cost economics (TCE), Property rights theory (PRT) and

Principal-agent or Agency theory (PAT). Additionally, Resource-based theory (RBT)

from the field of Strategic Management, is also considered and deployed in terms of

procurement. Finally, a review of the NIE theories and RBT in the Construction

Economics literature is provided.

2.2 Transaction cost theory

2.2.1 Origins and development

Professor Ronald Coase was the first scholar to recognise the existence of

transaction costs. This recognition led to the emergence of the firm in his seminal

paper, “The nature of the firm” (1937). Coase was awarded the Nobel Prize (1991)

for his discovery and clarification of the transaction costs in the institutional structure

and functioning of the economy.

34 Chapter 2: Literature review

Contrary to the mainstream belief held by economists that the economic system

“works itself” and is coordinated by the price mechanism, or Hayek’s “marvel of the

market”, Coase (1937, p. 38) argues that there is a cost associated with using the

price mechanism and “the most obvious cost of organising production through the

price mechanism is that of discovering what the relevant prices are” Transaction

costs of using the market can include discovering costs, negotiation costs and

contracting costs.

Coase (1988b, p. 19) observes that a firm emerges in the economic system if

transactions can be organised internally at a lower cost than if the same transactions

are carried out through the market. According to Coase, the choice between the

market and the firm can be determined through the comparison of the cost of

organising transactions within the firm (management or bureaucracy costs) and

market transaction costs. In other words, Coase associates the concept of transaction

costs with the study of the firm and market organisation, and explains that the

existence of the firm lies in its ability to capture efficiencies of transaction costs

during production and exchange; this is known as the ‘Transaction cost theory’. TC

not only explains the existence of the firm, but also its size and scope; that is, the

decision to make-or-buy within the firm.

Despite this important insight, TC has been “much cited and little used”

(Coase, 1988a, p. 67). The “Nature of the firm” was neglected for more than thirty

years (since its publication in 1937) before it experienced a revived interest in the

1970s when a number of economists began to explain business practices in the

presence of transaction costs; for example, Williamson (1975), Klein, Crawford and

Alchian (1978), and Alchian and Demsetz (1972). TC has since come a long way and

is now an important part of the study of institutional arrangements (Klein, 2005).

2.2.2 Make-or-buy analysis

The decision whether to carry out an activity within or outside the firm is

known as the make-or-buy decision, and determines the extent to which the firm is

vertically integrated (Besanko, Dranove, Shanley, & Schaefer, 2007, p. 120). More

specifically, the make decision – or internalisation – is a mode of operation in which

the firm is able to exert direct control over resources and is wholly responsible for an

activity. The buy decision – or externalisation – on the other hand, comprises all

other modes of operation.


Coase’s main argument is to include transaction costs in economic analysis –

costs that have been ignored by economists who “neglected the main activity of

running the business” (Coase, 1991, p. 38). In the “Nature of the firm”,

the key idea is on the comparison of the costs of coordinating the activities of factors of production within the firm with the costs of bringing about the same result by market transactions or by means of operations undertaken within some other firm (Coase, 1991, p. 38).

For example, the greater the external transaction costs of obtaining the supply of the

resource from the market, or the greater the savings through an increase in the

frequency of organising similar transactions (that is, greater efficiency or lower

internal organisational or management costs) which reduce production costs, the

greater is the incentive for the firm to integrate with the resource. More specifically,

the make-or-buy decision is focused on the comparison of organisational or

management costs (internal transaction costs) and market transaction costs (external

transaction costs) at the margins (Demsetz, 1988). As the firm expands, it

experiences diminishing returns to management, and the limit to the size of the firm

is set when its costs for organising a transaction are equal to the cost of carrying it

out through the market (Coase, 1937, p. 396). That is, the costs of organising an

additional transaction within the firm (internalisation) are equal to the costs of

carrying out that same transaction on the market, or organising it within another

market firm (externalisation).

Other subsequent theories relating to the concept of transaction costs, such as

TCE and PRT, have been developed with the aim of operationalising TC theory.

However, Coase did not deviate from what had been originally published in 1937,

and differentiated between TC and TCE. For instance, Coase recognises that

hazardous behaviours associated with asset specificity can be controlled with suitable

contract provisions; for example, a customer’s ownership of transaction-specific

assets employed by a supplier or mitigated by incentives to continue a mutually

advantageous relationship (for example, the prospect of future business

relationships), as opposed to vertical integration recommended by other NIE

economists (Winter, 1988, p. 169).

Coase calls for the operationalisation of TC through an investigation of the

influence of the interrelationships of market transaction costs and organisation or


management costs in the study of pricing practices, contractual arrangements and

organisational forms (Coase, 1988a, p. 47).

2.3 Transaction cost economics


TCE is a microeconomic theory of the firm, which originates from TC. In

1975, Williamson attributed the limited interest in TC theory to the lack of

operationalisation of the Coasian framework, and has since expounded and advanced

the theory. He incorporates the concept of bounded rationality and opportunistic

behaviour into the study of the nature of the firm, and deals with the problem of

hold-up and maladaptations to disturbances by assigning transactions to governance

structures in the most discriminating way (Williamson, 1985, p. 18).

The problem of hold-up has been widely discussed in the literature (Alchian &

Demsetz, 1972; Grossman & Hart, 1986; Klein, 1988; Milgrom & Roberts, 1988;

Williamson, 1971, 1985). TCE assumes that human agents are subject to bounded

rationality and self-interest seeking with guile; in other words, opportunism.

Williamson makes the important contribution that, given that idiosyncratic

investments or sunk costs (read ‘asset specificity’) are made by one party in a

contract, this party then becomes vulnerable to the opportunistic behaviour of the

other party, and a bilateral dependency is created. This bilateral dependency can

create the potential for opportunistic behaviour ex post. Because the predictive nature

of TCE is mainly associated with asset specificity, the concept of opportunism is one

of the key factors in its study. Indeed, TCE has also been known as the ‘moral hazard

approach’ and the ‘asset specificity approach’ (Alchian & Woodward, 1988). Coase

(1988a, p. 43) points out that asset specificity, or its associated opportunism, is not

found in the Coasian framework, and that this distinguishes TCE from TC.

Williamson summarises TCE succinctly when he explains that governance

(organisation theory) is the overarching concept and TCE is the “means by which to

breathe operational content into governance and organisation” (Williamson, 2010,

p. 215), and ”any issue that arises or can be reformulated…and be examined to

advantage in transaction cost economising terms” (Williamson, 2010, p. 221). In

other words, TCE can be applied to complex contracts and economic organisations

from a lens of contract perspective, with the emphasis on economising transaction


costs (Williamson, 2008). It is applicable to the study of vertical boundaries of the

firm and make-or-buy decisions; that is, it can explain why some transactions are

vertically integrated in the firm, while others are organised through the market.

TCE asserts that transactions differ in their attributes and are aligned with

governance structures (which differ in their adaptive strengths and weaknesses) to

accomplish the most transaction cost economising outcomes (Williamson, 2008, p.

9). Williamson defines ‘governance structure’ as an “institutional framework in

which the integrity of a transaction or related set of transactions is decided”

(Williamson, 1996, p. 11). In TCE terms, transactions differ in three key attributes:

asset specificity, uncertainty, and frequency.

Asset specificity refers to idiosyncratic investments that are durable and

transaction-specific, and that cannot be readily redeployed without sacrificing

productive value if the transaction or contract is terminated prematurely (Williamson,

1985, p. 54). Empirical studies have shown a significant relationship between

contractual arrangements and asset specificity [Joskow (1985) and Crocker and

Masten (1988)]. Results show that as asset specificity increases, firms tend to

subcontract less and integrate more (Maher 1997). Asset specificity is also the key

factor in creating bilateral dependency and the most important of the three TCE

attributes in determining hold-up (Williamson 1985) arising from the opportunistic

behaviour of the vulnerable contractual party, and is core to the TCE theory. This is

because these ex post investments (or sunk costs) create a bilateral dependency

between buyer and supplier, and hence create the potential for opportunistic

behaviour; for example, non-performance leading to a hold-up situation. The

following are the six types of well-established asset specificities in the TCE

literature, and need to be considered with each transaction to ascertain the most acute

asset specificity pertaining to a particular transaction.

1. Site specificity: investments in plant and equipment that are specific to the site

of the activity, and costly to move once set-up

2. Physical asset specificity: investments in specific technology that is not usable

for other activities; for example, equipment, inventory, machinery and software

3. Human resource specificity: investments in human knowledge and skills

specific to the activity (for example, time taken to familiarise oneself with an

activity)


4. Dedicated production asset specificity: investments to increase capacity; for

example, increase in plant capacity, that cannot readily be used for other

purposes

5. Temporal (or timeline) specificity: high costs (time and/or money) of replacing

the internal resources or external firm providing the activity if the activity sits

on the critical path or timeline

6. Trademark specificity: investments in building and protecting reputation; for

example, with respect to a particular group of products or a particular market

segment

The second attribute, uncertainty, refers to the disturbances affecting the

transaction and, unlike asset specificity and frequency, poses a contractual hazard

and requires additional governance or contractual safeguards. Contractual hazards

arise when the transaction is complex, and writing a complete contract is

prohibitively costly and time-consuming. Even if a transaction is uncertain or

complex, yet still requires investments that are non-specific, it might still be

relatively easy to mediate the transaction through market governance, as alternative

sellers are readily obtainable in the market (Maher, 1997, p. 168). However, if the

uncertain or complex transaction is coupled with high asset specificity or

idiosyncratic investments, then the transaction requires additional governance and

contractual safeguards to counter potential opportunistic behaviour.

Williamson (1985, p. 57) classifies two types of uncertainties: environmental

uncertainty (exogenous) caused by random acts of nature, and behavioural

uncertainty (endogenous), or the difficulty of evaluating performance. Similar to

asset specificity, the different types of uncertainty falling under the two categories

need to be considered in order to identify the most acute uncertainty pertaining to the

transaction.

The function of the frequency attribute is to consider the comparative

economies of both transaction and neoclassical production costs (Williamson, 1985,

p. 61). The frequency attribute is related to set-up costs and the impact of reputation

on generating the continuity of workflow (Williamson, 2008). Frequency is

employed so that the cost of specialised governance can be more easily recovered

across large transactions of a recurring kind. This justifies the size and extent of the

resource(s) required for each transaction. If a transaction is occasional rather than a


core activity of the firm, then it is not economical to internalise or vertically

integrate. Only frequency of a recurring kind generates enough economies of scale

and economies of learning to support or recover the cost of the internalised or unified

governance structure. In other words, frequency is related to the comparative ability

to aggregate demand in the market, or to generate enough workflow internally to

generate economies of scale and scope. However, the guidance in literature on

measuring frequency is general in nature; more detail is required to define or

determine the level of occurrence for transactions to be considered as either

‘recurrent’ or ‘occasional’.

Given the relationship between the behavioural assumptions of bounded

rationality and opportunism and the three key attributes of transaction, Williamson

(1985, p. 33) further explains the governance structure of a transaction in a firm in

broad terms, using a simple contractual schema. This schema has been improved

since 1985, and now includes hybrid contracting or governance; that is, contracting

or governance between the extremities of firm (make) and market (buy) that is more

encompassing than the focal firm or market dichotomy. Figure 2.1 (below) illustrates

the updated contractual schema.

Figure 2.1: Simple contractual schema

Source from “Outsourcing: Transaction cost economics and supply chain management,” by O.E. Williamson, 2008, Journal of Supply Chain Management, 44(2),5-16. Reprinted with permission from

John Wiley and Sons © 2008 Institute for Supply Management, Inc.™

In this diagram, k represents asset specificity, and s denotes safeguards in a

contractual arrangements. Where investment in technology or assets is general in

nature and can easily be redeployed in the market, asset specificity is low and

denoted by k=0. Since bilateral dependencies do not exist, then the choice is classical

market contracting or market governance at Node A.

Administrative support

s=0

D (Internal organisation/ firm)

A (Unassisted market)

s>0

k=0

C (Hybrid contracting)

B (Unrelieved hazard)

Market support k>0


When asset specificity is greater than zero (k>0) (say, investments in a special-

purpose technology), this creates a bilateral dependency and one of the contracting

parties becomes vulnerable to the opportunistic behaviour of the other party. If there

is no contractual safeguard (S=0), then the specialised investment is exposed and

poses a contractual hazard. This problem is solved by either: i) the provision of

contractual safeguards (s>0) – such as penalty clauses in contract agreements to

prevent hold-up – and taking the transaction out in the market using contracts with

credible threats, or hybrid contracting at Node C; or ii) internal organisation within

the firm at Node D. To illustrate this, given vertical integration as the last resort and

with a high degree of investment idiosyncrasy (such as the construction of a major

steel bridge), the firm may favour a more complex governance structure with

consequentially higher transaction costs at Node C; this represents hybrid

contracting.

Since the mid-1980s when Williamson (2009 Nobel Prize for Economics)

published his ground-breaking book on the subject (Williamson, 1985), TCE has

gained much prominence, and now appeals to a wide variety of disciplines and fields.

The disciplines that apply TCE have been classified into Economics and Business,

Finance and Accounting, Business Strategy, Organisation Theory, Law and Public

Policy, Health Economics and Policy, and Agricultural Economics and Policy

(Macher & Richman, 2008).

There has been extensive empirical research of TCE over a wide variety of

disciplines and studies, and TCE has been regarded as an empirical success story.

There are approximately 3500 articles and 900 empirical research studies of aspects

of TCE reported in articles and book chapters (Macher & Richman, 2008). Six major

empirical reviews of TCE (Boerner & Macher, 2002; Joskow, 1988; Lyons, 1996;

Masten & Saussier, 2000; Rindfleisch & Heide, 1997; Shelanski & Klein, 1995)

report a remarkable congruity between theory and evidence (Williamson, 2005b, p.

33). An extensive review of the empirical research also shows corroborative results

on TCE’s predictive nature (Macher & Richman, 2008). Geyskens, Steenkamp and

Kumar (2006, pp. 519, 531) remark that “Transaction cost theory has become the

predominant theoretical framework for explaining organisational boundary

decision” and that “Today transaction cost theory stands on a remarkably broad

empirical foundation”.


In relation to procurement, TCE has been employed in the decision to make-or-

buy and in the decision on the nature of exchange relationship. These applications

are discussed in Sections 2.3.2 and 2.3.3 (below).

2.3.2 Make-or-buy analysis

As mentioned previously (Section 2.2.1), the make-or-buy analysis determines

the extent of vertical integration; that is, it determines the decision to either outsource

services to external market firms or to provide or produce these services by

employing internal resources. Based on TC in maximising profits, the decision on

whether to externalise (market governance) or internalise (vertically integrate) turns

on the comparative costs of both governance and production costs (Williamson,

1985, p. 93). First, given fixed outputs and negligible economies of scale and scope,

the decision turns on the comparative governance costs of using the market or the

internal organisation. Second, given similar outputs and comparative governance

costs, the decision turns on the comparative production costs of economies of scale

and scope; that is, on whether the market or the firm is better at aggregating demand.

In other words, the firm will vertically integrate, if the overall transaction costs of

engaging the market are larger than the costs of internal organisation, and vice versa.

However, upon examining the ‘black box’ of the firm, it turns out that positive

transaction costs are ubiquitous, and the black box proves to be a Pandora’s Box.

This is because the measurement of transaction costs is far from a straightforward

process. For instance, the transaction cost of hold-up that has been avoided cannot be

measured unless we are able to go back in time to measure it (if it had occurred). For

this reason, Williamson focuses his attention on the key dimensions or attributes of

transactions – namely, asset specificity, frequency and uncertainty – to operationalise

the concept of transaction costs. Hence, the TCE approach makes it possible to

determine the organisational boundary of a firm by specifying the conditions for

organising a transaction within the firm, and the conditions to outsource the activity

to an external firm.

Conventionally, TCE simultaneously informs the make-or-buy and the nature

of exchange relationship decisions; that is, the make decision is synonymous with

vertical integration within a firm in a hierarchy governance structure, and the buy

decision is synonymous with a market governance structure. In other words, the

nature of the exchange relationship is considered as part of the make-or-buy decision.


However, the acceptance of hybrid governance indicates that there is scope to

consider make-or-buy and the nature of exchange relationship as two separate

decisions, rather than combining the two in a single step in the conventional method

(Bridge, 2008).

2.3.3 Exchange relationship analysis

After making the economic decision as to whether to internalise an activity

within the firm or externalise an activity to a market firm, the focal firm needs to

craft an entire governance structure. This includes the nature of the exchange

relationship with the internalised resource or the market firm, in order to oversee the

completion of the particular activity (Bridge, 2008, p. 42). When a dispute occurs in

a transaction, the norms under which the exchange relationship generally operate will

play an important role in determining the parties’ reactions to each other’s behaviour

during and after disputes (Kaufmann & Stern, 1988, p. 535).

The nature of exchange relationship is characterised by two contractual norms:

discrete or relational (MacNeil, 1978). Classical contracting is mainly applied to

transactions of a discrete nature, where the identities of contracting parties are

irrelevant and terms and conditions of the contract can be clearly specified; on the

other hand, neoclassical contracting and relational contracting are applied to

transactions of a relational nature, such as third party involvement (arbitration), long-

term contracts, and relational contracts.

Based on Macneil’s three-way classification of contracts (classical,

neoclassical and relational), and assuming that uncertainty is present but not to a

substantial degree, Williamson (1985) has developed a map of efficient governance

(See Figure 2.2). In this governance map, four modes of governance structures –

market, trilateral, bilateral, and unified governance – are matched to transactions of

varying asset specificity and frequency in an economising way.


Figure 2.2: Efficient governance

Source from “The Economic Institutions of Capitalism: Firms, Markets, Relational contracting” by O.E. Williamson, Copyright © 1985 by Oliver E. Williamson. Reprinted with permission of Simon &

Schuster Publishing Group from the Free Press Edition. All rights reserved.

Market governance is aligned with classical contracting, such as spot market or

contingent-claims contracts, where buyers and sellers meet to exchange standardised

or non-transaction specific goods. Market governance can be applied to transactions

of low asset specificity or of a standardised kind, regardless of the frequency of the

transactions. Reve and Levitt (1984) indicate that transactions of low asset specificity

and of recurrent nature can be handled effectively using classical or market

governance, and are protected against opportunism by the existence of market

alternatives and legislation.

Bilateral and unified governance are aligned with relational contracting, where

parties in bilateral governance retain their autonomy and transactions are organised

internally in unified governance. Bilateral and unified governance can be applied to

transactions that are recurrent and of a non-standardised kind; that is, transactions

that are mixed or highly specific. In other words, when transactions are recurrent and

investments are highly- or semi-specific (idiosyncratic), relational contracting in the

form of bilateral governance or employment contracts (related to organisation

governance) becomes the dominant governance mode (Maher, 1997; Reve & Levitt,

1984; Winch, 2001, 2002).

Trilateral governance is aligned with neoclassical contracting and can be

applied to transactions that are occasional and of a mixed or highly specific kind.

When transactions are occasional, then a highly specific governance structure, such

as unified or hierarchical governance, cannot be supported. Furthermore, given that

the transaction has some degree of asset specificity, classical contracting is more

effective because of the assistance of a third-party in resolving disputes; in other

words, trilateral governance. Maher (1997, p. 149) comments that,

Investment characteristics

Nonspecific Mixed Idiosyncratic

Trilateral governance (neoclassical contracting)

Bilateral governance Unified governance (relational contracting)

Mar

ket

gove

rnan

ce

Cla

ssic

al c

on

trac

ting

Rec

urr

ent O

ccas

ion

al

Fre

qu

ency


where all future contingencies cannot be identified such as in long term contracts and where appropriate adaptations may not be evident until the contingency arises, the transaction would break down under classical contracting. Firms in this case could move either towards more standardised products or vertical integration or alternatively implement a different contracting scheme which preserves trading but provides for additional governance referred to as neoclassical contracting. In his later work, Williamson acknowledges the emergence of a new form of

governance, that is, inter-firm cooperative arrangements – such as partnerships, joint

ventures and alliances – and classifies these as hybrid governance. This lies between

the firm and market dichotomy, and tends towards a more relational nature when

compared to pure market governance (Williamson, 1991). This shows that aligning

governance or exchange relationship with the classic three-way classification is not

straightforward, and there are opportunities for presenting a continuum of exchange

relationship, ranging from relational to discrete.

2.4 Resource-based theory


Since the mid-1980s, RBT has emerged as a dominant theory in the Strategic

Management field. RBT focuses on implementing strategies to achieve a sustainable

competitive advantage, so as to obtain better than normal economic performance or

economic rents (Barney, 1986). In contrast to TCE as a theory of the firm, RBT

focuses on the continual search for, and acquisition of, competitive advantage and

economic rents. TC and TCE, on the other hand, focus mainly on efficient

governance structures and the minimisation of transaction costs (Madhok, 2002).

RBT has gained prominence with the seminal contributions of Wernefelt

(1984), Rumelt (1984), and Barney (1986), who draw upon the work of Edith

Penrose, Philip Schelnik and Alfred Chandler. Other influential literature that

contributes to the resource-based approach, includes Lippman and Rumelt (1982),

Nelson and Winter (1982), Teece (1980), and Connor (1991). However, the RBT

gained the most significant momentum with Barney’s contributions (1986, 1991) in

formalising the RBT. Based on the work of Penrose (1959), Wernerfelt (1984), and

Rumelt (1984), Barney’s framework analyses the sources of sustainable competitive

advantage in terms of organisational strengths and weaknesses, and external

opportunities and threats (Barney, 1991).


RBT perceives the firm as bundles of productive resources (Penrose, 1959),

where resources and products are two sides of the same coin (Wernerfelt, 1984). That

is, even though a firm’s performance is driven by the firm’s products, it is ultimately

driven by the resources that are required for their production (Newbert, 2007). In

general, resources may refer to: all assets (tangible or intangible); capabilities;

organisational processes; and the information and knowledge controlled or owned by

a firm which enables it to conceive and implement strategies that improve its

efficiencies and effectiveness (Barney, 2007, p. 133).

The terms ‘resources’ and ‘capabilities’ are used interchangeably in the

literature (Barney, 2007, p. 135). Barney (1991, p. 101) defines ‘resources’ as

anything that enables the firm to conceive and implement strategies that improve its

efficiencies and effectiveness; and ‘capabilities’ as the ability of firms to exploit

resources to gain a competitive advantage. However, the generic nature of the

definitions of resources and capabilities, has resulted in confusion of terminologies in

the RBT literature (Foss, 1997b). For instance, Wernefelt (1984) only refers to

resources and describes them as a strength or weakness of a firm. Langlois (1992),

on the other hand, only talks about capabilities; Teece et. al. (1997) only refer to

dynamic capabilities; and Prahalad and Hamel (1990) are mainly concerned with

core competencies.

Following Prahalad and Hamel’s (1990) influential paper, Barney (2007, p.

135) distinguishes core competencies from resources and capabilities and reserves

the term ‘core competencies’ for discussion on diversification of the firm. Prahalad

and Hamel identify core competencies of the firm as a distinctive source of

competitive advantage or asset, and define it as the ‘collective learning of the

organisation, especially on ways to coordinate diverse production skills and integrate

multiple streams of technologies’ (Prahalad & Hamel, 1990). Core competencies are

a function of tacit understanding and resources that a firm accumulates over time

(Mahoney, 1995).

Teece et al. (1994, p. 19) differentiate two types of competence; namely,

organisational or economic competence and technical competence. Organisational or

economic competence involves: 1) allocative competence; that is, deciding what to

produce and how to price it; 2) transactional competence; that is, deciding whether to

make or buy, and whether to do so alone or in partnership; and 3) administrative


competence; that is, designing organisational structures and policies to enable

efficient performance. Meanwhile, Teece et al.’s (1994, p. 19) second type of

competence, technical competence, includes the ability to develop and design new

products and processes to operate facilities effectively.

Capabilities and competencies have important tacit dimensions and knowledge.

Barney (2001a, 2001b) acknowledges close links between RBT and evolutionary

theory, and accepts concepts drawn from evolutionary theory, such as divisions in

labour (Smith, 1776), routines and learning (Nelson & Winter, 1982), capabilities

view (Langlois, 1992), core competence view (Prahalad & Hamel, 1990; Teece,

Rumelt, Dosi, & Winter, 1994), knowledge-based view (Grant, 1996; Kogut &

Zander, 1993), and the dynamic capabilities approach (Foss, 1993; Teece, Pisano, &

Shuen, 1997). Barney (2007) considers these as part of, or examples of, resources

and capabilities. The original static analysis of competitive advantage has now

evolved into a dynamic resources approach to RBT, which stresses the importance of

a firm’s competencies or core competencies in addressing changing environmental

conditions, and being an important source of competitive advantage.

RBT assumes that the firms are mostly heterogeneous in terms of resources and

capabilities, where different firms possess different bundles of resources; that is, the

concept of ‘resource heterogeneity’. It also assumes that these resources are limited,

valuable, rare, and costly-to-imitate; that is, the concept of ‘resource immobility’.

The firm is organised in order to exploit these firm-specific resources and

capabilities to gain a sustained competitive advantage.

Barney (1991) developed the VRIO (value, rarity, non-imitability and

organisation) framework which aims to identify resources that are potential sources

of competitive advantage (strengths) and competitive disadvantage (weaknesses).

Barney (1991, p. 102) asserts that ‘a firm is said to have sustained competitive

advantage when it is implementing a value creating strategy (value) not

simultaneously being implemented by any current or potential competitors (rarity),

and when these other firms are unable to duplicate the benefits of this strategy

(imperfect imitability)’. Fundamentally, in order to gain potential sustained

competitive advantage, the firm’s resources must possess key traits of value, rarity,

and imperfect imitability.


The value of a resource lies in its ability to enable a firm to devise and

implement strategies that improve the firm’s efficiency and effectiveness, and to

exploit opportunities and neutralise threats. The typical approach of most empirical

studies is to measure value in terms of economic performance of the firm, or

competitive advantage (Newbert, 2007). However, Priem and Butler (2001) charge

the value dimension of RBT with tautology, and Barney (2001a) agrees that the

value variable is exogenous to RBT and not fully parameterised. Furthermore, given

that valuable, rare and non-imitable resources on their own do not generate

competitive advantage, Newbert (2007, p. 124) argues that Barney has moved on to

also include organisation as a factor in generating competitive advantage. Due to the

challenges of measuring value, resource-based scholars, including Newbert (2007, p.

140), suggest the need to devise alternative definitions of the value variable so as to

avoid tautology.

The rarity of resources refers to the level of the firm’s current and potential

competition. That is, if a valuable resource is owned and implemented by a large

number of firms such that all firms are capable of exploiting the resource in the same

manner – in other words, the resources are not rare – this gives no competitive

advantage to any firm. However, the firm does enjoy a competitive advantage if the

number of firms that implement the resource is fewer than that required to generate

perfect competition in the market, thus generating the potential for sustainable

competitive advantage. Rarity is often measured by showing the extent of resources

or the incidence of market firms competing in the same market, or by measuring

changes in the demand for services (other things being equal).

The extent to which resources are imperfectly imitable refers to resources that

other firms cannot completely duplicate due to: a) unique historical position or path

dependencies where capabilities have been developed over a period of time; b)

causal ambiguity, that is, the source of how the resource is being acquired or

developed is either unknown or unclear; or c) social complexity, that is, the

competitive advantage is based on complex social phenomena that other firms are

unlikely to imitate. Furthermore, strategically equivalent substitutes for valuable

resources that are rare or not completely imitable should not exist simultaneously.

Non-imitability is measured by estimates of the difficulty and/or cost of acquiring

and implementing the tacit nature of new technologies.


In addition to VRI, the firm also needs to be organised in a way that exploits

the competitive potential of the firm’s resource. Thus, the organisation attribute in

VRIO refers to the internal attribute of the firm that brings together value, rarity and

imitability attributes, and might include formal reporting structures, management

practice and compensation policies. It determines the extent to which competitive

advantage can be achieved and is a moderating factor which, in isolation, does not

generate competitive advantage. Organisational components of the firm are

complementary to capability and resources (Barney, 2007).

RBT literature appears to be silent on the nature of exchange relationship

between firm and market. That is, having made the make-or-buy decision, RBT does

not offer any explanation or clarification of the nature of the exchange relationship,

and the firm is able to choose either a discrete or relational exchange (Bridge, 2008).

Fundamentally, RBT is concerned with the exploitation of idiosyncratic and

costly-to-imitate resources which may potentially provide the firm with a

competitive advantage. Accordingly, the VRI framework provides a systematic

framework to analyse the strengths and weaknesses of the firm by focusing on the

value, rarity and imitability of its resources in achieving a sustainable competitive

advantage. The following section now discusses how Strategic Management

researchers have applied RBT and the VRI framework to the issue of make-or-buy.

2.4.2 The make-or-buy decision

Given that the very existence of a firm is based on a continual search for

economic rents, RBT can be considered in terms of determining firm boundaries, that

is, the make-or-buy decision (Foss, 1993). RBT contends that a firm should

internalise an economic activity when it has the competitive advantage of a valuable,

rare and costly-to-imitate resource; conversely, the firm should externalise the

activity when another firm has the valuable, rare and costly-to-imitate resource.

This heterogeneity of resources or capabilities and competencies among firms

implies that they have varying or differential capabilities, and the relative capabilities

of the firm and market help to determine the level of integration and the boundaries

of the firm; that is, to determine the make-or-buy decision (Foss, 1993, 1997a; Foss

& Knudsen, 2003; Langlois, 1992; Peteraf, 1993; Teece, Rumelt, Dosi, & Winter,

1994). For instance, two similar firms producing a similar type of product might have


a different level of integration due to differential capabilities with the market

(Langlois, 1992). As the relative capabilities or competencies between market and

firm change, the boundaries of the firm change accordingly. This means that the

differential level of external or internal capabilities between firm and market in an

economy will be important to the level of vertical integration of a firm.

Making the make-or-buy decision based on existing internal capacity alone,

however, can be criticised as a form of tautology. Coase (1988b, p. 19) agrees with

this tautological charge in saying that,

the limit to the size of the firm would be set when the scope of its operations had expanded to the point (that is to expand or exclude certain resources or capabilities), at which the costs of organising additional transactions within the firm exceeded the costs of carrying out the same transactions through the market or in another firm (that is differential capabilities of firm versus market).

In other words, even though the relative capabilities or resources indicate the level or

extent of vertical integration in a firm, it is not a strong determinant for the make-or-

buy decision (Barney, 1999). By making the make-or-buy decision based on existing

capacity alone – that is, the decision to internalise when the firm has the capacity and

externalise when the firm does not have the capacity – can be considered to be

circular reasoning and, therefore, a form of tautology. Barney (1999) argues that it is

ultimately the VRI attributes of the resources or capabilities, in terms of

organisational strengths and weaknesses and external opportunities and threats, that

determine the level of competitive advantage and the extent of vertical integration of

resources within the firm.

RBT focuses on the firm-specific competencies or resources that provide the

competitive advantage and promote vertical integration (internal capabilities); for

example, technical competence (Barney, 1986, 1991) and organisational competence

(Foss, 1993, 1997a; Kogut & Zander, 1996; Madhok, 1996; Teece, Pisano, & Shuen,

1997). Firms with superior technical competence or resources will earn superior

profits, while firms with marginal technical competence or resources are likely to

only break-even. In terms of organisational competence, the more successful a firm

is in organising an economic activity in a manner that the market cannot, or the more

successful it is in carrying out a similar activity with greater frequency, the more

efficient it becomes in administering or organising the activity, and the more

profitable it becomes in internalising the activity (Connor, 1991; Foss, 1993, 1997a;

Kogut & Zander, 1996; Madhok, 1996; Nelson & Winter, 1982; Poppo & Zenger,


1998; Teece, Pisano, & Shuen, 1997). For example, Langlois and Robertson (1989)

and Langlois (1992) examine Ford’s case of vertical integration of car parts

manufacture in the early stages of the automobile industry in the 1920s, when other

automobile makers were mainly assemblers only and purchased parts from their

suppliers. With process innovation and a greater knowledge of its own needs, Ford

had a cost advantage over its suppliers (Langlois, 1992, p. 114), and was able to

make use of its own resources to mass produce car parts (vertically integrate) at a

lower cost than the cost of obtaining them from the market.

In terms of externalisation, Strategic Management researchers propose that

non-core activities that are not part of the business strategy can be outsourced

(Connor, 1991; Madhok, 2002; McIvor, Humphreys, & McAleer, 1997; Venkatesan,

1992); that is, the coherence with, or relatedness of these activities to the core

business activities determines whether they are outsourced (Teece, Rumelt, Dosi, &

Winter, 1994). Teece, et al. (1994) explain that firms tend to externalise activities

that are not related to the main business activities or existing main asset base, and

more frequently combine related activities within the same corporation. Echoing

Coase’s argument that “the costs of organising and the losses through mistakes will

increase with an increase in… the dissimilarity of the transaction” (Coase, 1937, p.

397), Madhok (2002) highlights the importance of routine – learning and tacit

knowledge; in other words, over-extension of activities into domains which are too

diverse and dissimilar not only dilute the strength of a firm’s competence, but also

increase the costs of organising those activities in-house due to lack of expertise and

experience in the area. In other words, if the activity is not a core business of the

firm, the firm is likely to externalise the activity to a market firm which is more

competent in the activity, and is unlikely to integrate the activity into the business

structure in the short term.

Similarly, Barney (2002, p. 208) proposes that when another firm has valuable,

rare and costly-to-imitate resources and capabilities that are too costly or not

economical to acquire (external capabilities), non-hierarchical governance or

externalisation can be preferred in spite of significant threats of opportunism. RBT

holds that if the valuable resource that the market owns is extremely rare and is not

imitable in the short term, or if the cost of acquiring or developing the resource or

capability is too high – due to causal ambiguity, historical context or path


dependencies, or social complexity (which result in a monopoly or oligopoly market)

– externalisation of the activity becomes the preferred economic option. This is the

case regardless of potential opportunistic market behaviour. Langlois (1992, p. 108)

summarises this situation,

All firms must rely on the capabilities owned by others, especially to the extent those capabilities are dissimilar to those the firm possesses. A firm could … acquire dissimilar capabilities complementary to the ones they already own. But there is no particular reason to do so unless there are specific transaction costs impeding contractual arrangements. And there are generally costs to owning dissimilar assets, especially when the acquiring firm cannot use or sell their full capacity.

In other words, RBT suggests that if the firm does not have a particular

profitable capability or competence, it can either: 1) outsource or externalise an

activity or transaction to a market firm, using either market governance or

intermediate governance; 2) develop these capabilities on its own via hierarchical

governance or integration; or 3) simply acquire the firm and use hierarchical

governance (Barney, 2007). However, due to the costliness of acquiring a capability,

it may or may not be profitable to simply acquire the resource (issue of value). In the

long run, profitability depends on whether there is a large enough workload to drive

the acquisition or develop the capacity (issue of frequency); that is, the frequency

needs to be high enough to offset or justify the cost of acquiring the capability. This

indicates that there is a close relationship between the frequency variable in TCE and

the value variable in RBT, and that the make-or-buy analysis is essentially a

production cost question.

In terms of early studies of the relationship between RBT and TCE, Langlois

(1992) was one of the early researchers to associate governance costs (or TCE) and

the relative competence approach (or RBT), in the consideration of boundary

decisions. Having examined the automobile industry, and in particular the vertical

integration of Ford, Langlois proposes dynamic governance costs as the costs of

transferring capabilities or knowledge, to provide a better explanation of vertical

integration rather than TCE attributes.

Argyres (1996) highlights some of the anomalies in TCE by pointing out that

RBT is better at explaining the rationale for firms externalising activities even when

the potential for hold-up is high. Argyres examines a multidivisional firm producing

a wide variety of industrial products and identifies that the make-or-buy of some

production activities is dominated by resource-based explanations, while some are


dominated by transaction cost variables. Argyres proposes using a combination of

the transaction cost approach and the relative capabilities or competence approach to

help in explaining make-or-buy more completely. Madhok (2002) suggests that both

the characteristics of transactional conditions and the attributes of the firm’s own

capabilities should be considered in firm decisions. Meanwhile, Jacobides and

Winter (2005) acknowledge that there is now substantial empirical research that

applies both the transaction cost approach and the capabilities or competence

approach to the study of firm boundaries.

Even though make-or-buy has been studied and tested extensively in TC, TCE

and RBT literature, there is not yet a unified theory that synthesizes TC and/or TCE

and/or RBT in make-or-buy analysis into a formalised monistic framework.

However, there has been some significant progress in the integration of TCE and

RBT in the analysis of make-or-buy decisions.

2.5 Integration of Transaction cost economics and Resource-based theory in the analysis of make-or-buy decisions

Both Williamson and Barney highlight the complementary strengths of TCE

and RBT and point to their integration (Barney, 2002; Williamson, 1999). In terms of

the make-or-buy decision, TC compares internal transaction costs (such as

management costs) with external transactions (such as resources) (Demsetz, 1988).

Coase asserts that “...the more efficient solution depends on the absolute relation of

the costs of different arrangements” (Coase, 1991, p. 70), while Williamson differs

from this TC view by stating that “transaction costs are economised by assigning

transactions to governance structures in a discriminating way” (Williamson, 2005b,

2008). While TCE focuses on the potential effects of hold-up on governance, both

TC and TCE can be applied in determining the most efficient exchange relationship;

for instance, between government and the private sector.

Anomalies in the empirical research of TCE imply weaknesses in the theory

(Argyres, 1996; Jacobides & Winter, 2005; Madhok, 2002), and Williamson and

Coase now point out the need to incorporate firm differences and inherent

idiosyncrasies into the analysis. Williamson acknowledges the limitations of their

assumptions, which do not account for internalisation in pursuance of production

benefits beyond the market, and falls in the purview of RBT. Furthermore,

Williamson is beginning to accept that TCE is a static theory and now encourages


pluralism of theories to include RBT, such as learning and knowledge-based

approach (Williamson, 1999). Given that the theories are complementary and not

contradictory in nature, adopting pluralism is beneficial in further enhancing the

understanding of a complex phenomenon (Groenewegen & Vromen, 1996).

Indeed, there is a growing body of empirical research that justifies and supports

theoretical pluralism and integration of TCE and RBT. In general, researchers have

tested TCE and RBT and found support for each theory in different aspects of make-

or-buy. Based on their findings, they recommend that pluralism of theories provides

better explanations and more accurate predictions. Table 2.1 lists some of the recent

developments in the integration of TCE and RBT (in chronological order).

Table 2.1 Summary of research on the integration of TCE and RBT

Argyres (1996)

A multidivisional firm producing a wide variety of industrial products was examined and make-or-buy decisions explained from a relative capabilities approach, together with a transaction cost approach. With reference to the nature of the activity, Argyres identifies that the make-or-buy of some production activities is dominated by resource-based explanations, while some are dominated by transaction cost variables. Argyres proposes using a combination of TCE and RBT to help to explain make-or-buy more completely.

Poppo and Zenger (1998)

Proposed a series of make-or-buy hypotheses or predictions using transaction cost and resource-based theories, and tested the degree of support for these theories on boundary choice by examining the data on the governance of nine information services functions across 152 companies. The results indicate that the integration of transaction cost, knowledge-based and measurement reasoning is likely to improve the study of the theory of the firm and a theory of boundary choice.

Leiblein and Miller (2003)

Based on the work of Poppo and Zenger, Leiblein and Miller developed an index function model based on insights from transaction cost and resource-based theories. Empirical testing of 469 make-or-buy decisions across 117 semiconductor firms provides strong evidence that transaction-level characteristics, firm-specific capabilities and product-market scope have effects on governance decisions.

Jacobides and Winter (2005)

Given that the boundaries of the firm is dynamic in nature and changes with the industry context, they propose four evolutionary mechanisms of transaction costs and capabilities that shape the vertical scope of the firm over time, and illustrate their application in the mortgage banking industry in United States and in the Swiss watch manufacturing industry.

Jaspers and Jan van den Ende (2006)

Propose that the vertical relationship between a buying firm and a supplier firm can be categorised into 4 dimensions of integration: ownership integration (the extent that the firm owns the upstream component supplier), coordination integration (the intensity of information exchange to align the two stages of production), task integration (the extent that the buying firm performs upstream tasks), and knowledge integration (the extent that the buying firm possesses knowledge about the upstream component). This draws on transaction costs economics, information-processing theory, and learning.

Gulbrandsen, Boge Sandvik, Kåre Haugland, Sven A.(2009)

Develop a vertical integration model that is derived from TCE and the resource-based approach, and is empirically tested with data from the mechanical maintenance services in the hydroelectricity industry. Results show that asset specificity and closeness to present competence are positively correlated with vertical integration, while tacit knowledge is negatively correlated. However, the model is only focused on vertical integration and does not include the progress that has already been made in the relationship between internalisation and asset specificity, where a high level of asset specificity can indicate both vertical integration or internalisation and externalisation.

Boneta, Peris-Ortiz and Gil-Pechuan (2010)

Having examined the complementary nature of TCE and resource-based view in the integration and externalisation of activities with reference to several economic cases in history, Boneta et. al. (2010) propose a theoretical framework comprising of comparative costs, idiosyncratic demands and core competences, and suggest that idiosyncratic demands is the main locomotive in predicting make-or-buy.


Research on the integration of TCE and RBT has progressed and is now

converging on the integration of both TCE and RBT to better explain the in-depth

phenomena in the study of the firm. It is evident now that both transaction costs and

production costs need to be simultaneously considered in making decisions on the

boundaries of the firm.

2.6 Property rights theory


Property rights have been broadly defined as the rights of individuals to use

resources (Alchian, 1965). In economic terms, Libecap (1989) describes property

rights to be the rights to use, to earn income from, and to transfer or exchange assets

or resources. These rights are partitioned and grouped into bundles (or shared

ownership) and are assigned to the transacting party which is most efficient at

utilising the bundles and the incentives that come with them (Alchian, 1965; Alchian

& Demsetz, 1973). The concept of property rights has been applied in broad social

science issues of economic development and social welfare (Alchian, 1965;

Demsetz, 1967; Furubotn & Pejovich, 1972); economics and political science

(Libecap, 1989; North, 1990); and at the individual firm level (Barzel, 1989; Hart,

1995).

The property rights approach gained momentum with the seminal papers of

Grossman and Hart (1986) and Hart and Moore (1990), and has been referred to as

‘the modern Property Rights Theory (PRT)’ (Holmstrom & Roberts, 1998; Kim &

Mahoney, 2005). Hart purports that control and incomplete contracting are important

concepts that can be applied in understanding a number of economic institutions and

arrangements, including boundaries of firms and financial structures (Grossman &

Hart, 1986; Hart, 1995; Hart & Moore, 1990). In Hart’s framework (1995), asset

ownership is defined as being synonymous with residual rights of control; that is, the

rights to use the asset in any way except to the extent that specific rights have been

given away in an initial contract. Grossman and Hart (1986) present a formal theory

of integration based on the optimal allocation of residual rights of control or asset

ownership between parties. The stylised mathematical formula trades off the costs

and benefits of ownership and shows that in cases when complete contracting is


costly to write, it can be optimal for one party to gain ownership over the other party

(Grossman & Hart, 1986).

With its foundations in Coase (1937, 1960), PRT is closely related to TC, and

transaction costs can be defined as the economic costs associated with the transfer,

capture and protection of rights (Barzel, 1989). Williamson acknowledges that

modern PRT is complementary and may be able to help in formalising TCE through

the development of a governance model, with its foundations in Grossman and Hart

(1986) and Hart and Moore (1990) (Tadelis & Williamson, 2010; Williamson, 2010).

The following section discusses the application of PRT in relation to residual rights

of control or make-or-buy, and draws mainly on the work of Oliver Hart (Grossman

& Hart, 1986; Hart, 1995; Hart & Moore, 1990).

2.6.2 Bundling

The modern property rights approach has attempted to explain firm boundaries

based on the optimal allocation of asset ownership, given incomplete contracting

(Grossman & Hart, 1986; Hart & Moore, 1990). Hart (1995) points out that classical

property rights literature focuses on the importance of gaining control and ownership

of an asset, but does not explain why it matters who owns a piece of private property

at an individual firm level; that is, why it is necessary to determine which party

should own which asset. Hart’s work focuses on the boundary and scope of the firm

in the market economy and describes incomplete contracting to both explain and

predict firm-level vertical integration decisions using the PRT approach (Mahoney,

2005).

The emphasis of Hart’s framework is based on the problems of incomplete

contracting and the optimal or efficient allocation of residual rights of control over

assets, in order to explain who owns which assets (Grossman & Hart, 1986, p. 717).

Given the difficulties in predicting all future contingencies ex ante, it is difficult to

write a complete contract to specify the actions of contracting parties when

externalities occur. Even if it were possible to write a complete contract covering all

types of contingencies, it might be too costly to do so. Hence, contracts are written

with gaps, and with provisions missing; that is, incomplete contracting. This can lead

to high re-negotiation costs when externalities that have not been anticipated in the

contract arise. Another problem that is associated with incomplete contracting is

incontractible quality, or elements that could not be anticipated and provided in


contracts; in this case, residual rights of control matter for economic efficiency. In

other words, the party who has the residual rights of control over the resource can

decide the usage of a resource in any way that is not inconsistent with a prior

contract, custom or law (Hart, 1995). Based on Hart’s model, if a complete contract

can be written and every contingency specified (costless contracting), then ownership

does not matter for economic efficiency (therefore, externalise); conversely, if it is

too costly to specify contingencies, then it might be more efficient for the owner to

make decisions (that is, to internalise). The efficient allocation of control or power in

asset ownership determines who owns which asset or resource; that is, the make-or-

buy decision. Hart views make-or-buy decision as an optimal allocation of control

rights among parties to a transaction. Therefore, given that contracts are incomplete,

firm boundaries or allocation of control rights matters for economic efficiency.

Asset ownership or integration creates the economic incentive system that

shapes resource allocation (Mahoney, 2005). It is in the interest of the acquiring

party to look after the asset, and to invest in relation-specific investments (or sunk

costs) to increase its productivity or efficiency. In contrast, if the asset is being hired

or leased, then the party does not have the incentive or the residual control rights to

repair or modify the asset, as the party does not own the asset.

Hart (1995) concludes that the economic benefit of integration is that the

acquiring firm’s economic incentive to make relationship-specific investments

increases; given that it has more residual control rights (bundling or greater extent of

vertical integration), the firm will receive a greater fraction of the ex post surplus

created by these relationship-specific investments. The contractual party that retains

ownership of the asset has the most to gain from the bundling of relationship-specific

assets (Mahoney, 2005).

2.7 Agency theory


Agency or Principal-agent theory (PAT) is applicable to the study of principal-

agent relationships – for example, an employer-employee, lawyer-client, or buyer-

supplier relationship – and addresses the problems that can occur in the relationship

between the principal and the agent. The principal-agent or agency relationship arises

when the principal engages or contracts an agent to perform some service on their


behalf, which involves delegating some decision-making authority to that agent

(Jensen & Meckling, 1976, p. 308). Hence, it has application in a wide range of

fields; for instance, accounting, auditing, law, economics and organisational

behaviour.

According to PAT, due to the divergence or conflicts of interests and goals

between principal and agent, the agent might not always act in the best interest of the

principal, and can make decisions differently from the principal due to different risk

perspectives. In this situation, it becomes difficult or expensive for the principal to

verify that the agent is behaving appropriately. The two main sources of the agency

problem are moral hazard (agent shirking) and adverse selection (misrepresentation

of ability by the agent). In order to reduce loss and protect the principal’s welfare, the

principal can either establish monitoring mechanisms to police the administration of

contracts between principal and agent, and/or provide appropriate incentive systems

in the contract to ensure optimal performance (Godfrey & Hill, 1995). The sum of

agency costs comprises of monitoring expenditures by the principal, bonding or

incentive expenditures by the agent, and residual losses (Jensen & Meckling, 1976).

Given the assumptions of bounded rationality and self-interest seeking with guile,

PAT’s main concern is to determine the most efficient contractual governance of

optimal risk sharing and incentive structure between the principal and agent.

PAT has its origins in the literature of the 1960s and 1970s which discusses the

differences in risk attitudes between cooperating parties; for example, Arrow (1971)

and Ross (1973). Eisenhardt (1989) presents a comprehensive review of this

literature and distinguishes two mainstream approaches that have been developed:

the formal PAT, which is based on mathematical proof and deduction; and positivist

agency theory, which is a more applied version of the agency theory. Principal-agent

theorists, such as Jensen and Meckling (1976), McAfee and McMillan (1986),

Laffont and Tirole (1987), Laffont and Martimort (2002), and Holmstrom and

Milgrom (1991, 1994), focus on determining the optimal contract or payment terms,

by utilising logical deduction and mathematical proof. Positivist agency theory, on

the other hand, is focused on identifying situations in which the principal and agent

have conflicting goals, and then describing the governance mechanisms that limit the

agent’s self-serving behaviour.


PAT emphasises economic selection processes, and favours governance

structures that economise on agency costs (Eisenhardt, 1989). Governance structure

in the PAT literature refers to the mechanisms established in the contract that help to

reduce the divergence of interest between principal and agent, such as incentive-

based performance contracts, or mechanisms that reduce the possibility of the agent

misbehaving, such as monitoring mechanisms or enforcement mechanisms (Godfrey

& Hill, 1995). PAT purports that an adequate alignment of economic incentives at

the outset leads to efficient contracting.

Despite sharing the same antecedents of bounded rationality and self-interest

seeking with guile with TCE, PAT is different from TC, TCE and PRT, as it does not

explain the make-or-buy decision or clarify the boundaries of the firm (Eisenhardt,

1989). Conversely, TC, TCE and PRT do not explain how to ensure that the agents

behave in accordance with the principal’s best interest, and are silent on the

economic incentive problems (Hart 1995). Eisenhardt (1989) points out that PAT

theorises the contract between the cooperating parties, while TCE theorises on

organisational boundaries. That is, TCE takes a pre-contract perspective in guiding

the make-or-buy decision and thereby choosing an efficient governance structure,

while PAT addresses the agency issues that can arise post-contract. TCE and PAT

could be considered in tandem so that the make-or-buy decision is based on

minimising transaction costs (TCE) and production costs (RBT). The contracts could

subsequently be designed to realign the interests of both principal and agent in

minimising agency costs (PAT). In other words, the main concern of PAT is with the

post-contractual minimisation of agency costs after the make-or-buy decision has

been established.

Eisenhardt (1989) urges that the PAT be adopted to investigate the problems

inherent in a principal-agent relationship in a governance structure or in contractual

payment terms. The important point being emphasised here is, in an exchange

relationship between the principal and agent, the overarching nature of the

governance structure of the relationship comes first, and is then followed by the

characteristics of the principal-agent relationship. The following section discusses

the applicability and application of PAT to exchange relationship analysis, which

includes contractual payment terms.


2.7.2 Nature of exchange analysis

In a typical exchange, both the principal and the agent seek to maximise

benefits received from the contract; that is, the principal wishes to minimise total

investment, and the agent wishes to minimise effort. As such, the onus is on the

principal to design a contract that offers incentives for the agent to behave as the

principal desires.

From the positivist stream, research is mainly concerned with describing

mechanisms, such as implementing information systems or outcome-based

incentives, to solve the agency problem of the separation of ownership from

management in large corporations (Fama & Jensen, 1983). The principal-agent

stream, on the other hand, is more directly focused on the contract of the exchange

relationship between the principal and the agent (Eisenhardt, 1989), and predicts

whether the exchange should be governed by outcome-based contracts or behaviour-

based contracts, using mathematical proof and deductions. With careful specification

of assumptions, the most efficient or optimal contracting method (outcome-based or

behaviour-based) is derived from the optimisation of the principal’s expected utility

or payoffs, based on agency variables related to the exchange analysis; these

variables include, risk attitudes of the principal and agent, information systems, task

programmability, and outcome uncertainty (Eisenhardt, 1989).

Other principal-agent researchers have also used empirical data to test and

support the propositions on the predictions of behavioural or outcome-based

contracts derived from PAT. For instance, Floricel and Lampel (1998) tested a

sample of 60 power plant projects, and Kalnins and Mayer (2004) used data from 394

contracts from the information technology services industry. Their results were

consistent with PAT predictions. The following summarises the agency literature

discussion of the two generic types of contracts that help to reduce agency costs:

behaviour-based contracts, and outcome-based contracts (Eisenhardt, 1989; Farrell,

2003; Floricel & Lampel, 1998).

Outcome-based contracts refer to contracts where the project outcomes, rather

than agent inputs, are prescribed. Outcome-based contracts motivate behaviour by

the co-alignment of the agent’s preferences with those of the principal, albeit at the

cost of transferring risks to the agent. The incentive contract and fixed-price contract

are examples of outcome-based contracts, where the principal pays the agent a fixed


price for carrying out the prescribed work. The incentive is created by making the

agent the residual claimant of any cost savings in the execution of the contract, which

is considered as a high-power incentive contract. The principal, on the other hand,

incurs the costs of writing the full outcomes and the costs of risk premiums that are

passed on by the agent for undertaking risks in the outcome-based contract. The

incentive may be eroded by uncertainties in a project, when total cost becomes

difficult to estimate ex ante, and specification of the output difficult to define.

For a comparatively uncertain and complex project, the agent will require a

higher risk premium for using an outcome-based contract rather than a behaviour-

based contract (Farrell, 2003). In addition, Bajari and Tadelis (2001) show that any

subsequent variations to the agreed contractual requirements will become more

costly to renegotiate for the outcome-based contracts, as compared to the behaviour-

based contracts. Outcome-based contracting is suitable in circumstances of

information asymmetry where the principal has limited access to technology and/or

resources, and is also suitable for risk-averse principals because of its certainty of

outcomes.

Behaviour-based contracts, in contrast to the outcome-based contracts, refer to

contracts where the principal completely prescribes and monitors the actions of the

agent. Incentives are created by compensating the agent for fulfilling the terms of the

prescribed behaviour and penalising them for any deviations from it. The principal

incurs the costs of specifying and monitoring the behaviour of the agent, and needs to

have the necessary competence to monitor their performance; that is, the principal

undertakes the risks of completion of the contract. Cost-plus or re-measurement

contracts are examples of behaviour-based contracts, where the agent is monitored

and compensated or reimbursed (plus a fixed fee) for the amount of work carried out

according to their compliance with the prescribed behaviour or performance.

However, cost-plus or re-measurement contracts are considered low-power incentive

contracts, because the agent has little incentive for price reduction or innovation as

they do not appropriate any of the savings.

If there is a greater alignment of interests between the principal and agent, then

the agent will tend to behave in accordance with the interests of the principal

regardless of any monitoring, and the more likely that behavioural contracts will be

used. Hence agency costs for behaviour-based contracts are sensitive to the level of


conflict of interests and the competence of the principal in monitoring and evaluating

agent behaviour (Floricel & Lampel, 1998). Principal-agent researchers also

demonstrate that behaviour-based contracts are preferred to fixed-price contracts

when a project is more complex and uncertain, where the cost of measuring quality

ex post is high, and where the task of estimating costs ex ante is difficult (Kalnins &

Mayer, 2004).

There are inherent risks in both types of contract and both are not optimal

contractual choices (Floricel & Lampel, 1998). Ultimately, the choice is dependent

upon factors such as the level of information asymmetry, the level of competence of

the principal, the level of risk aversion of the principal, and outcome uncertainty.

These, therefore, are important factors to consider in contractual design. Muller and

Turner (2005) suggest a continuum of contracts from fixed-price (outcome-based) to

cost-plus (behaviour-based), where most of the risks lie with the agent at the

outcome-based end of the continuum, and with the principal at the behaviour-based

end of the continuum. Behaviour-based contracts are better known as performance-

based specifications or contracts in the construction industry, while outcome-based

contracts are more commonly referred as method specifications contracts.

A form of hybrid contract comprising of features of both behaviour- and

outcome-based contracts, lies between the two extremes. One example is cost-plus

contracts with a cap, where the agent is compensated for carrying out the prescribed

work up to a maximum value of the agreed fixed price (Kalnins & Mayer, 2004). The

maximum agreed value is adjusted according to any client-initiated variations and

risks beyond the agreed terms of the contract. The incentive is created by linking the

maximum agreed sum with a gain-share or pain-share regime; that is, the firm is

rewarded with part of the cost-savings and responsible for some fraction of the cost-

overrun. The risk allocation is distributed between the principal and agent, and helps

to ensure quality with cost efficiency.

An outcome-based or fixed-price (armslength exchange relationship) contract

is preferred when costs are easy to estimate ex ante; a cost-plus with cap (hybrid)

contract is preferred when there is some level of difficulty in estimating costs; and a

behaviour-based or pure cost-plus contract (relational exchange relationship) is

preferred when costs are extremely hard to estimate (Kalnins & Mayer, 2004).

Projects with extremely hard to estimate costs are related to high risk projects with


more unknowns, for instance, underground works (such as road works) or integration

with existing facilities (such as existing hospitals). Table 2.2 summarises the three

types of contracts, and indicates that PAT can be applied to the exchange relationship

in terms of consideration of risk-sharing and risk distribution, in the form of financial

payment terms.

Table 2.2 Summary of behaviour, hybrid and outcome-based contracts

Behaviour-based Hybrid (outcome and behaviour- based)

Outcome-based

• Cost-plus • Low power incentive • Principal's risk to

completion • Agency cost

o specification of behaviour

o monitoring of behaviour o outcome uncertainty o high project complexity o less goal alignment

• Target outturn costs or guaranteed construction sum linked with gain-share or pain-share regime

• Risks balanced between agent and principal

• Suitable for outcome uncertainty

• Fixed-price • High power incentive • Agent's risk to completion • Agency costs

o specification of outcomes

o verification of outcomes o risk premium o suitability for

information asymmetry o outcome certainty o better goal alignment

The literature on NIE theories on the issue of make-or-buy, on the nature of

exchange relationship and on residual rights of control or bundling has now been

discussed. The following section reviews the development and application of each

theory in relation to aspects of construction procurement; that is, to make-or-buy

analysis, exchange relationship analysis and bundling analysis.

2.8 Review of New Institutional Economics literature in construction economics

2.8.1 Make-or-buy analysis (Transaction cost theory, Transaction cost economics and Resource-based theory)

a) Transaction cost theory and Transaction cost economics

In the Construction Economics literature, Eccles (Eccles, 1981a, 1981b), Reve

and Levitt (1984), and Winch (1989) are pioneers in applying the transaction cost

approach to the study of governance and organisation in the construction industry.

Eccles (1981b) makes the important contribution of noting the existence of

quasifirm in the construction industry. Quasifirm is a type of hybrid mode of

governance, which represents a combination of transacting in the market

(externalisation) and hierarchy (internalisation). Examples of hybrid contracting are

alliances or joint ventures between public and private sectors, which tend towards a

more relational exchange between the client and the main contractor. Reve and Levitt


(1984) apply transaction cost to the relationship between contracting and governance

structures in complex construction projects, and propose a clan and professional

relationship among client, main contractor and subcontractor to generate

transactional advantages. Finally, Winch (1989) advocates the application of TCE at

the activity level rather than at the project level, which is at the heart of TCE. He

makes the incisive observation that,

the key question… is not whether a project is organised as a market or a hierarchy, but why do construction firms choose to contract for construction services (make), rather than employ the capacity to provide those services themselves (buy)… In other words, where is the efficient boundary of the construction firm? (Winch, 1989, p. 337) In relation to the boundary of the construction firm, most construction

researchers focus on the boundaries between main contractor and subcontractor, that

is, on subcontracting or supply chain management in the construction industry. This

research includes: Eccles (1981a, 1981b); Winch (1989, 2001, 2002); Costantino,

Pietroforte and Hamill (2001); Constantino and Pietroforte (2004); Gonzalez-Diaz et

al. (2000); Lai (2000); Bon (1991); Bridge and Tisdell (2004, 2006); Bridge (2008);

and Brahm and Tarziján (2013).

With respect to the boundaries between the client and the main contractor,

various research applies TC and TCE to transaction cost economising aspects of

procurement; or to construction project processes; from the client’s perspective. For

example, the research applies TC and TCE to: construction governance structure

(Winch, 1989, 2001, 2002); risk allocation (Jin, 2009, 2010; Jin & Doloi, 2008; Jin

& Zhang, 2011); mathematical-based procurement models (Bajari & Tadelis, 2001;

Sha, 2011); and procurement selection framework (Bajari & Tadelis, 2001; Chang &

Ive, 2002a; Eriksson, 2006; Ive & Chang, 2007; Sweeney, 2009; Tadelis, 2002).

Most of these studies are either conceptual or exploratory, and include a few case

studies. Only five include empirical data and testing; namely, Eccles (1981b);

Gonzalez-Diaz, Arruñada, and Fernández (1998, 2000); Costantino, Pietroforte and

Hamill (2001); Bridge (2008); and Brahm and Tarziján (2013). However, only

Gonzalez-Diaz, Arruñada, and Fernández (1998, 2000), and Bridge (2008) have

specifically measured the TCE attributes.

Even though TC and TCE have been the dominant microeconomic theories

pertaining to governance and procurement since the 1980s, the application of TC and

TCE to procurement selection or make-or-buy analysis in construction has been

fairly recent, and is seen in the work of Bajari and Tadelis (2001); Bridge (2008);


Chang and Ive (2002b, 2007b); Ive and Chang (2007); Eriksson (2006); Jin and

Doloi (2008); Sweeney (2009); and Turner (2004). Table 2.3 (below) is a summary

of the procurement selection models which are based on TCE only. Jin and Doloi

(2008) and Bridge (2008) integrate TCE with RBT, and this is discussed in greater

detail in the next section.

Table 2.3 Summary of singular transaction cost approach in procurement selection

Bajari and Tadelis (2001)

Develop a mathematical model that involves the trade-off between incentives and transaction costs, and explain the many stylized facts of procurement contracts. Their model shows that cost-plus contracts are preferred to fixed-price contracts when a project is more complex, and can provide some micro-economic foundations for TCE.

Ive and Chang (2007)

Propose a conceptual model of the inconsistent trinity that the client will inevitably need to trade-off with the procurement systems: 1) fastest completion, 2) least vulnerability when making changes, and 3) least vulnerability to non-performance.

Per Erikkson (2006)

Provides a conceptual model of the procurement and governance of transactions from the perspective of the contractor, by integrating industrial buying behaviour and TCE.

Turner (2004)

Develops a decision-making flowchart for selecting contract strategy (payment terms), depending on whether uncertainty is controlled by client or contractor, and whether the project is simple or complex.

Sweeney (2009)

Develops TCE analysis schema based on the TCE approach to predict procurement delivery methods and improve contracting performance; has shown empirical support for TCE schema analysis.

Ive and Chang (2007) propose an inconsistent trinity model where the client

will face trade-off between: 1) fastest completion, 2) least vulnerability when making

changes, and 3) least vulnerability to non-performance. However, the model does not

fully operationalise TCE, as it mainly focuses on measuring process specificity

variable and does not consider other TCE variables such as frequency and

uncertainty. In addition, the model is concerned with minimising cost on opening day

of the asset and not on whole-of-life costs.

Based on the efficacy of different contractual types in terms of incentivisation

and governance, Turner (2004) develops a decision-making flowchart for selecting

contract strategy (payment terms), depending on whether uncertainty is controlled by

the client or contractor, and whether the project is simple or complex. This method is

straightforward and relatively easy to implement; however, it lacks full

operationalisation of TCE attributes.

Sweeney (2009) develops a TCE analysis schema to analyse performance of

different procurement approaches in the construction industry from a TCE

perspective. The TCE analysis schema comprises of four hypotheses and the

empirical results provide support for the predictions in the TCE analysis schema.

This indicates that it could be suitable as a tool to predict procurement delivery


methods and to improve contracting performance. However, the analysis can be

biased towards large complex projects, and requires further work to determine the

threshold value of the project at which the schema is no longer a suitable tool for

prediction.

Eriksson (2006) proposes a conceptual model based on TCE attributes of asset

specificity and frequency, industrial buying behaviour, and PAT to develop a set of

prescribed governance mechanisms, and promote co-operative procurement

procedures to achieve efficient governance. However, this model does not account

for uncertainty, and does not explain how the proposed governance mechanisms are

able to handle opportunistic behaviour.

Even though the above models are associated with TCE, the models do not

follow TCE closely and do not directly measure the TCE key attributes. In general,

the models lack operationalisation of TCE in terms of measurement of its key

attributes such as asset specificity, uncertainty and frequency, and require further

empirical testing for validation purposes.

To summarise, the TCE literature in construction procurement demonstrates

that there is appreciable room for the development of a comprehensive procurement

model that clarifies the operationalisation of TCE attributes. Its application in the

construction industry at the firm level could help to improve the decision-making

process of whether to contract for construction services in the market, or to employ

the firm’s capacity to provide these services. In relation to externalised activities, it

could also help to craft an appropriate governance structure that could potentially

economise on bounded rationality, and safeguard transactions against possible

opportunistic behaviour.

b) Resource-based theory and integration with Transaction cost theory

In the RBT literature, there are not many studies that address the issue of

make-or-buy. The RBT literature proposes the identification of critical resources or

assets, and the outsourcing (buying) of products or services that are non-core or non-

strategic to the business. Having surveyed the literature, Cánez, Platts and Probert

(2000) identify work by Venkatesan (1992), McIvor, Humphreys and McAleer

(1997), and Cox (1997), that applies RBT to the make-or-buy decision. However, the

frameworks proposed have limited applicability in construction procurement.


Corresponding with Williamson’s (1999) call for integrating TCE and RBT, a

number of construction researchers – such as Bridge and Tisdell (2004), Bridge

(2008), Jin and Doloi (2008), and Brahm and Tarziján (2013) – integrate TCE and

RBT in different approaches to the study of make-or-buy and apply these in the

construction field. Table 2.4 is a summary of the integration of TCE and RBT in

construction-related industries.

Table 2.4 Research on the integration of TCE and RBT in the construction industry

Bridge and Tisdell (2004) and Bridge (2008)

Bridge and Tisdell (2004) and Bridge (2008) represent the only research that justifies the complementary nature of the TCE and RBT theories employed in the framework based on Lakato’s Science Research Program. The framework is based on a capability and competence spectrum between the focal firm and independent market firms, where the differential capabilities and competencies across technical and/or organisational competence of the focal firm and independent market firm generate differential competitive advantage. RBT provides an efficiency-based explanation of company performance derived from competitive advantage attributed to a firm’s resources having differential levels of efficiency, competence or capability. The levels are generated by the characteristics of the transaction of the activity. Activities that are valuable and/or rare and/or costly-to-imitate, indicate a greater potential for the activity to generate a competitive advantage, and a greater likelihood that the activity will be internalised, and vice versa. The framework clarifies the scope of different levels of competitive advantage; namely, sustainable competitive advantage, temporary competitive advantage, competitive parity and competitive disadvantage, in explaining and predicting internalisation and externalisation. Additionally, the framework clarifies the relationship between Coase theory (frequency) and the value variable in RBT.

Jin and Doloi (2008, 2009) , Jin (2009, 2010), and Jin and Zhang (2011)

Develop a conceptual model that facilitates the interpretation of risk allocation in PPP projects, from a TCE perspective integrated with the resource-based view of organisational capabilities. The model provides a logical and complete understanding of the process of selecting a risk allocation strategy for a particular risk in a PPP project.

Brahm and Tarziján (2013)

Develop a logistic regression model of vertical integration (dependent variable), which is being defined as a function of: 1) temporal specificity, 2) construction quality, 3) institutional changes in law, and 4) differential capabilities between contractors and subcontractors. It has been tested empirically in the Chilean subcontracting construction industry, where the empirical results conclude that the predictions of transaction costs, insofar as temporal specificity and institutional change, are negatively correlated with capabilities in explaining vertical integration.

Jin and Doloi (2008, 2009) and Jin (2009, 2010) address risk allocation and

risk transfer using the TCE approach. Jin and Doloi (2008) consider that it is

appropriate to address risk allocation from a TCE perspective because any issue can

be formulated as a contractual problem and investigated to advantage in transaction

cost economising terms (Williamson, 1985). Although this framework does not

predict procurement route, it indicates the applicability of TCE in terms of risk

transfer or allocation between client and contractor in association with reducing

costs. For example, a certain risk transfer will be agreed to by both parties in order to

reduce transaction costs. However, this method is the most complex of those

discussed, requiring the knowledge of fuzzy logic (Jin & Zhang, 2011). Teo, Bridge

and Jefferies (2010) and Chang (2013) also agree that the methodology employed in

the framework has a key weakness. That is, because risk management is not a


transaction, utilising ‘risk management responsibility’ as the unit of analysis is not

appropriate, because the unit of analysis should be mediated by a governance

structure (market, hybrid or hierarchy). In Chang’s (2013) recent review of work by

Jin and Doloi (2008, 2009), Chang comments that the framework is deficient in

terms of poor specification of governance structure and misinterpretation of asset

specificity.

The procurement model in this research will address all of Chang’s concerns.

Furthermore, it is interesting to note that, in his critique of Jin and Doloi’s (2008,

2009), Chang has effectively conceded that TCE and RBT can be combined to

address the make-or-buy decision. Thus, Chang is supporting the approach being

taken in the procurement model developed in this research.

Based on an extensive data set from the Chilean construction industry, Brahm

and Tarziján (2013) develop a logit model of vertical integration, aiming to explain a

contractor’s decision to make-or-buy trade activities at an activity level. In their

analysis, they assume that asset and human specificity are generally low, and attempt

to operationalise and measure temporal specificity in terms of switching costs,

number of change orders, and cost of delays. They have developed a logistic

regression model of vertical integration (dependent variable), which is defined as a

function of 1) temporal specificity, 2) construction quality, 3) institutional changes in

law, and 4) differential capabilities of contractors and their sub-contractors. They

have found that temporal asset specificity and hazards caused by institutional change

are related to increased vertical integration by contractors. In addition, the analysis

shows that if the contractors have stronger differential productive capabilities than

the subcontractors, the potential transactional hazards caused by subcontractors are

reduced. They posit that differential capabilities are related to governance choices in

the construction industry, and that the predictive strengths of the theories can be

improved with the inclusion of RBT.

In general, these empirical findings are supportive of the pluralistic view of

integrating TCE and RBT. Brahm and Tazijan (2013) cite and support Bridge and

Tisdell’s approach of integrating TCE and RBT and, notably, are silent with respect

to Chang’s critique of this approach.

Of all the procurement selection research outcomes related to TCE in

construction reported in the literature, only the integrative framework of


internalisation and externalisation developed by Bridge and Tisdell (2004) – where

the make-or-buy decision is the focal transaction from the perspective of a focal firm,

and where (as mentioned) Chang’s concern about using transaction as an appropriate

unit of analysis is also addressed – has a strong potential application to procurement

selection (Bridge, Tiong, & Wang, 2010). Taking into consideration the

developments in the literature, and justifying the complementary nature of the

theories based on Lakato’s Science Research Program, Bridge and Tisdell (2004) and

Bridge (2008) have integrated TC, TCE and RBT to develop an integrative

framework of vertical integration.

This integrative framework of vertical integration considers market capacity

and in-house resources of the firm and, in so doing, goes beyond other TCE-based

approaches. The framework combines the operationalisation of TCE and RBT

attributes in a spectrum of seven levels of capabilities and competencies of the firm

and the market. The framework has been tested and validated empirically in the

mechanical services supply chain industry with respect to the firm’s internalisation or

externalisation decision, and can be applied at the level of the activity. Therefore, it

can contribute to make-or-buy decisions at the activity level. In addition, Bridge

(2008) has also developed and empirically tested the application of TCE on the

nature of the exchange relationship.

The framework offers the facility to focus on conditions concerning the

technological and physical attributes of a project, as well as the capabilities and

competencies of the government and private sector with respect to that project. It

also has the potential to address the aim of this research. The project conditions

represent what is to be measured, while a priori theory guides the way in which these

conditions are measured, and the manner in which resultant measurements inform

procurement selection. For these reasons, the procurement selection is more likely to

incorporate whole-of-life considerations and to achieve a more efficient balance

between production costs or benefits and transaction costs. Indeed, Bridge, Tiong

and Wang (2010) recommend the development, application, and empirical testing of

the integrative framework with respect to major infrastructure procurement and, in so

doing, envisage contributing to significant progress in the area.

In summary, the literature review identifies the potential application of the

integration of TCE and RBT in the construction industry. However, the literature


indicates only limited application of TCE in the make-or-buy decision in the

construction and construction-related sectors. A number of leading scholars in the

field of construction management and economics have supported Bridge and

Tisdell's approach to the integration of TCE and RBT; for example, Kumaraswamy

et al. (2008), Walker (2007), and Bröchner (2008, 2011). Indeed, Ball (2007, p. 221)

considers that Bridge and Tisdell have developed an "ingenious" approach to

integrating the TCE and RBT literature.

The integrative framework of vertical integration has yet to be applied in the

procurement decision of new infrastructure projects, and the framework requires

further matching of a full spectrum of procurement approaches to key production

activities in the construction field. Further refinement of the methods of measuring

TCE and RBT attributes in terms of infrastructure characteristics, and opportunities

to broaden the approach to include bundling and relationship exchange application

within an NIE framework, are also required.

2.8.2 Exchange relationship analysis (Transaction cost theory and Transaction cost economics)

The Construction Economics literature reveals different perspectives on the

exchange relationship between the main contractor and subcontractor. Gunnarson

and Levitt (1982) are of the view that subcontracting is a hybrid governance

structure; that is, a mixture of market and firm. Eccles (1981a, 1981b) considers

subcontracting as an interface between market and firm, while Reve and Levit (1984)

argue that subcontracting is similar to an employment or relational contract and

towards unified governance, and observe that the ‘elements of hierarchy are added to

what appeared to be a pure market relationship’ (Reve & Levitt, 1984, p. 20).

In an effort to clarify the contractual nature of subcontracting by returning to

Coase’s original thesis, Lai (2000) argues that subcontracting is a nexus of Coasian

firms interacting through the main contractor with a main consumer in a Coasian

market. More importantly, it has become clearer that the market and hierarchy are

two limits of a continuum of contractual choices, and that the focus is on determining

the nature of exchange relationship via contractual choice to achieve efficient

governance.

The nature of the relationship between parties can be specified by the contract

signed, and the contracts between client and contractor and between client and


consultant provide the formal basis for construction project governance (Reve &

Levitt, 1984). Correspondingly, hybrid governance is aligned with relational

contracting, such as cost-plus contracts, reimbursable contracts, risk sharing

mechanism and gain-share or pain-share regime. These are less adversarial contracts

than the fixed-price lump sum contracts that characterise the construction industry.

The adversarial, fragmented and inefficient nature of construction industry

practices has led construction management researchers to explore the efficient

exchange between client and main contractor, and main contractor and sub-

contractor. Results indicate a general consensus among construction researchers on

the role of trust and cooperative relationships in reducing opportunistic behaviour

and achieving effective governance, particularly for complex projects with high

uncertainties (Doree, 1997; Kale & Arditi, 2001; Love, Tse, Holt, & Proverbs, 2002;

Lyons & Mehta, 1997; Lyons, 1996; Miller, Packham, & Thomas, 2002; Parkhe,

1993; Rahman & Kumaraswamy, 2004; Walker & Chau, 1999). Table 2.3 (below) is

a summary of the research on efficient exchange in the construction industry.

Table 2.5 Summary of research on efficient exchange relationship in the construction industry

Parkhe (1993)

Empirical data from 111 inter-firm alliances demonstrates that the perception of opportunistic behaviour is greater at the beginning of the alliance, and the guard against opportunistic behaviour tends to lower when trust-dominated understanding between parties develops and helps to reduce transaction costs and improve efficiency.

Lyons (1996) Reports evidence that the type of contractual relationship (classical, neoclassical, or relational contracting) depends on both the frequency with which exchange takes place, and the extent to which the transaction requires specific investment.

Lyons and Mehta (1997)

Analyses the role of self-interested trust and socially-oriented trust in facilitating efficient exchange.

Doree (1997)

Empirical data shows that Dutch municipals use a limited number of preferred contractors and prospects of future work to develop a more continuous and stable relationship with them. These contractors become more flexible, cooperative, and quality oriented as a result.

Kale and Arditi (2001)

Examined the impact of the quality of inter-firm relationships between contractors and subcontractors and found positive correlation between positive relationships and performance.

Walker and Chau (1999)

Selection of governance structure should aim at minimising transaction costs according to the complexity of the project; that is, simple governance structures should be used in conjunction with simple contractual relations and complex structures for complex relations.

Love, Tse, Holt and Proverbs (2002)

Using a case study, demonstrate the effectiveness of cooperative learning alliances in increasing profitability, and reducing risks and transaction costs.

Miller, Packham and Thomas (2002)

Conclude that integration of processes into construction process and relational ties can help to reduce guileful self-interest seeking in subcontractors and main contractors, and lead to efficiency and reduced transaction costs.

Rahman and Kumaraswamy (2004)

Discuss the nature of various team members, and demonstrate how relational contracting principles can be applied to building procurement during the entire project life cycle; for example, the implementation of a joint risk management strategy to address unforeseen risks and incomplete contract conditions.

Bridge (2008) Develops a trust-commitment-relationship (TCR) trinity which uses trust to clarify the relationship between TCE and nature of exchange relationship.


Reve and Levitt (1984) were early pioneers who promoted the development of

trusting relationships between client, main contractor and subcontractor to achieve

efficient governance, using the TCE approach. Their major contribution is that

trusting relationships can “exhibit transactional advantages that economise on

bounded rationality and reduce incentives for opportunism in situations of

complexity and uncertainty”. They term the relationship between client and

consultant as a ‘professional relationship’ and the relationship between consultant

and contractor as ‘clan relationship’. However, clan relationships and professional

relationships are both intangible, informal relationships that are not easily defined

within a formal governance structure, or in contractual terms. This explains the lack

of its application in the determination of an appropriate contractual relations or

governance structure in the literature.

Williamson (1985, p. 406), although recognising the importance of trust in

governing transactions, merely treats trust and commitment as latent variables, and

focuses on the measurement of tangible attributes. The TCE model or schema is

deficient in terms of accounting for the role of trust in contractual relations, as the

model excludes trust in the analysis of efficient exchange on the grounds of

“inordinate difficulties of measuring trust” (Williamson, 1985, p. 406). On the other

hand, Bridge (2008) points out that TCE is incomplete in explaining trust and

commitment and their interaction with the nature of exchange relationship. More

importantly, however, he develops a trust-commitment-relationship (TCR) trinity

which uses trust to clarify the relationship between TCE and nature of exchange

relationship, as shown in Figure 2.3 (below).

Figure 2.3: The TCR trinity

Source from “The determinants of the governance of air conditioning maintenance in Australian retail centres,” p. 111, by Bridge (2008) Queensland University of Technology. Reprinted with permission.

Affective Commitment/ Credible Commitments

Relational Exchange

Potential Nature / Level

of Trust

Calculative Commitment /Credible Threats

Discrete Exchange

Realised Trustworthy Environment

Realised Distrusting Environment

Bonding / Asset Specificity

Uncertainty

Service Quality / Frequency


The TCR trinity illustrates that high asset specificity and high frequency are

likely to develop a high level of trust and commitment. With a lesser reliance on

contractual safeguards; hence, TCE predicts a relational exchange. If the investment

is of high level of asset specificity and uncertainty, but is of low frequency, it

generates a distrusting atmosphere with potential opportunism, and the exchange

relationship is highly discrete using credible threats, such as penalties for non-

performance. By incorporating the TCR trinity into the analysis of exchange

relationship, Bridge (2008) reinvents the method of determining the nature of

exchange using TCE variables, and represents the exchange in a continuum from

‘discrete’ to ‘relational’, as illustrated in Figure 2.4 (below). This is in contrast to

Macneil’s (1978) three-way classification of contractual exchange. The TCR trinity

proposed by Bridge (2008) suggests that the lesser the potential for hold-up, the more

likely that an efficient exchange is preferable; the greater potential for a strong hold-

up suggests either a relational exchange or a discrete inefficient exchange. These

exchange relationships have been empirically tested and validated in the building

mechanical services supply chain (Bridge, 2007, 2008).

Key: +++ = Extremely high incidence of variable ++ = Very high incidence of variable + = High incidence of variable 0 = Low high incidence of variable

Figure 2.4: Summary of external exchange relationship Adapted from “The determinants of the governance of air conditioning maintenance in Australian

retail centres,” p. 113, by Bridge (2008) Queensland University of Technology. Reprinted with permission, and Bridge (2014b).

Coincidentally, the continuum of the nature of relationship in Figure 2.4

(above) broadly matches the three types of contracts – outcome, hybrid and

behaviour-based – (previously) given in Table 2.2. Consistent with Eisenhardt’s

recommendation, the opportunity now appears to connect PAT with the above Figure

2.4, which is based on TCE.

Efficient Exchange (naturally occurring)

+++ Relational +

Inefficient Exchange

+ Discrete +++

Efficient Exchange

+ Discrete +++ External Exchange

TCE Asset specificity +/+++ Uncertainty +/+++ Frequency +/0

0 0 0 +++ +/0 0

+ +++ + +++ +/0 0

a cb

1

3 2


2.8.3 Bundling of design, construct, operations and maintenance activities (Property rights theory)

Given the bundling incentive associated with relationship-specific investments,

PRT has been recently applied to construction infrastructure procurement. This

application includes Hart’s (2003) much-cited development of incomplete

contracting theory with regard to PPP, which examines the economic efficiency of

construction from an incomplete contract perspective. Empirical research which

applies PRT to infrastructure procurement includes the work of Blanc-Brude (2006)

and Grimsey and Lewis (2007). The following section discusses the recent literature

on the economics related to the bundling of operations and maintenance activities

with the design and construction activities of a public sector infrastructure project

from a property rights perspective, which is also closely related to the nature of

exchange relationship.

The allocation of residual control rights to the private firm – that is, bundling to

include operations and maintenance – creates the incentive for the contractor to

invest a larger capital outlay upfront in the construction phase, that could reduce the

whole-of-life maintenance costs of the asset (productive efficiency), so as to gain

greater returns on investments in the long term (Blanc-Brude, Goldsmith, & Välilä,

2006; Grimsey & Lewis, 2007). In PPP, for instance, bundling operations and

maintenance into the construction contract – that is, making the consortium or

service provider responsible for operations or maintenance of the project, or

increasing the level of the consortium or service provider’s ownership – increases the

contractor’s incentive to invest in production and non-production-related

investments. This can help to lower life-cycle maintenance costs of the asset, and

reduces the problem of incontractible quality.

Conversely, in unbundling, the contractor does not internalise any social or

operating costs and there is little incentive for them to invest. Thus, the contractor

only carries out the requisite production and non-production-related investment. If

there is a lack of specification of production quality and more focus on the service

provision, the contractor might have the flexibility to choose a lower productive

investment, or to modify the nature or quality of the asset without violating the

contract. In other words, the contractor might well build in the cheapest possible

way, while staying within the enforceable contract stipulations and ignoring the


adverse effects of any non-contractible quality, as the contractor does not internalise

the externalities. It becomes clear, then, that unbundling can subsequently impinge

on the performance of the asset over the next thirty years, and ultimately, the

government pays for the low quality of the asset in the operations and maintenance

stage. Hence, property rights do not favour unbundled provision where specifications

are not clear at the outset, as is the case in Design and Construct contracts.

In summary, Hart (2003) concludes that unbundling – for example, the use of

separate design, construct, operations and maintenance contracts – is preferable when

the quality of the asset can be well specified at the outset, whereas the quality of

service cannot be. Bundling – that is the bundling of design and/or construct and/or

operations and/or maintenance, such as in PPP – on the other hand, is preferable

when the quality of service can be clearly specified at the outset, whereas the quality

of the building might not be.

Hart shows that the choice of contract (that is, bundled or unbundled) turns on

“whether it is easier to write contracts on service provision than on building

provision” (Hart, 2003, p. C75). More specifically, conventional contracting is

effective if the building performance specifications – as opposed to the services

performance specifications – can be clearly specified; for example, with respect to

the building of schools. Conversely, Hart proposes that bundling (such as PPP) is

effective if the services performance specifications are easier to be well specified

than building performance specifications.

In this respect, Hart’s view is consistent with the House of Lords (2010, p. 31)

recommendation that the projects that are most suitable for private finance are those

where the requirements can be clearly specified. However, Hart’s speculation that

hospitals are more suitable for PPP is contrary to the House of Lords’ view. The

House of Lords (2010, p. 28) deem that hospitals are not suitable to be procured

through PPPs because of major potential for hold-up of services in situations where

change occurs; for example, the result of demand, or the demographics of the local

population. Hold-up has been observed in services that are susceptible to change,

such as patient record keeping, IT services and other non-clinical services. These

services can pose as a serious potential for hold-up, if these are locked into a contract

for twenty to thirty years. Hence, the inflexibility of PPP contracts can result in high

hold-up costs.


From the above, it appears that there are uncertainties in the literature

pertaining to the types of projects that are best procured under PPP, and to the

bundling and unbundling of services in a construction project. Although PPPs are

claimed to be a value-for-money tool (KPMG & Infrastructure Australia, 2010),

there is a lack of evidence or research to support the claim. Thus, the House of Lords

(2010, p. 33) is calling for research into whole-life cost data to compare PPPs with

traditionally procured projects, in order to determine the best procurement approach

for infrastructure projects. The challenge of bundling is to minimise bureaucracy

costs through greater use of single-point contact with a private sector firm across a

wide range of activities (transfer of control to private sector firm) while, at the same

time, minimising hold-up through the greater use of internalised management (or

agent) and control over private sector firms. Thus, there is a tension or a trade-off in

the extent to which both types of transaction costs can be minimised.

In summary, PRT is complementary to TC and TCE theory, and introduces the

concept of ownership and control rights; that is, the bundling of design, construct,

operations and maintenance activities to construction procurement.

2.8.4 Exchange relationship analysis (Agency theory)

In a procurement context dogged by moral hazard, adverse selection, and the

problems of incomplete contracting and asymmetric information in major

infrastructure projects, PAT can contribute to a procurement model by greater goal

alignment between principal and agent, which minimises potential agency or

monitoring costs. The principal-agent relationship exists in the procurement of

infrastructure construction, where the principal (the client or project owner) engages

the agent (main-contractor or subcontractor) to undertake the task of delivering a

facility or infrastructure, such as a hospital or road.

Most construction contracts are incomplete when agreed upon, and this

encourages a divergence of interests of the contracting parties. The agent who

performs on behalf of the principal may or may not perform to satisfaction if

incentives are not in place, and the problem of moral hazard arises. For instance,

supplier (agent) might not be motivated to share critical information with the client

(principal), or the supplier (agent) might be tempted to supply lower quality goods

when buyers (principals) are not able to monitor the quality of goods and services

received (Winch, 2002). The emphasis is to provide ex ante incentive alignment


mechanisms in contractual arrangements to reduce agency costs arising as a result of

adverse selection and potential moral hazard caused by agents (Winch, 2002).

Turner and Simister (2001) propose the selection of appropriate pricing or

payment terms to align the goals of the client and the contractor, and to prevent the

potential for negative opportunistic behaviour of either party. Turner and Simister

(2001, p. 460) observe that different contract types or payment terms can incentivise

or motivate contractors in various ways, and that they can be “expected to perform

differently under different payment terms”. Turner and Simister (2001, p. 460) also

indicate that the final out-turn costs of the same project will be different under

different payment terms. Turner and Simister (2001) develop the goals and methods

matrix (illustrated in Figure 2.5 below) to select suitable project payment terms. This

matrix is based on the level of uncertainty of the project in defining goals;

specifically, it is based on: 1) the complexity of the project (that is, the uncertainty of

the process); and 2) the ability of the client to contribute to the resolution of

problems (that is, uncertainty of the product).

Uncertainty of the product/ Ability of client to contribute

Low High

Uncertainty of the

process/ Complexity

High Fixed price

Design and Build

Cost plus Design and Build

Alliance

Low Re-measurement

Build only

This situation was not researched

Figure 2.5: Goals and methods matrix Source from “Project contract management and a theory of organization,” by J.R. Turner and S. J. Simister, International Journal of Project Management, 19(8),457-64. Reprinted with permission

from Elsevier Limited, copyright 2001.

However, the matrix is incomplete in one quadrant, and is not comprehensive

in terms of the whole range of payment options, such as cost-plus with cap (hybrid)

payment terms (Kalnins & Mayer, 2004). Subsequently, Turner (2004) revised the

matrix into a flowchart process to select contract strategy based on: 1) who controls

the risks, 2) the nature of the project, and 3) the location of the uncertainty. One key

change is the determination of make-or-buy (client, or contractor, or both) as the first

step in the process, as illustrated in Figure 2.6.


Figure 2.6: Contract selection based on uncertainty of project and risk allocation Source from “Farsighted project contract management: Incomplete in its entirety,” by J.R. Turner, Construction Management and Economics, 22(1), 75-83. Reprinted with permission from Taylor &

Francis, Copyright © 2004 Routledge.

The merit of this process lies in the relevance of make-or-buy analysis as the

step before exchange relationship analysis or contract selection. This indicates

similar approach to the model in this research and reaffirms the relevance of

exchange relationship being a subsequent step to make-or-buy. This process also

broadly aligns with the outcome, hybrid, and behaviour-based contracts (previously)

summarised in Table 2.2 (which are based on PAT), and with the exchange

relationship analysis developed by Bridge (based on TCE) (See Figure 2.4). The left

branch looks to be related to behaviour-based contracts or relational exchange

(including internal exchange); the right branch seems to be related to outcome-based

contracts or discrete exchange; and the middle branch appears to be related to hybrid

contracts. However, in the flowchart, risk analysis comes after deciding make-or-

buy, and the latter is based on a simplistic and dichotomous analysis – that is, simple

or complex – that leads to a specific contract type. This creates a lack of flexibility in

Who controls the risk?

Client Both Contractor

Complexity?

Simple

Complex

Management Cost + Incentive

Complexity? Location of risk?

Simple Process

only

Remeasurement Sor or BoQ

Remeasurement BoM

Design and Build fixed price

Complex Product and

Process

Alliance Cost + Gain Share

Prime contracting target price

Neither

Remeasurement or fixed price


the contract selection process; however, there is the potential for further

development.

The nature of the exchange relationship in the procurement model is developed

based on TCE in the following chapter, and this is followed by connecting principal-

agent characteristics with the nature of exchange relationship (based on PAT) to

provide additional information of a better-aligned principal-agent relationship.

2.9 Summary

The literature review of the NIE theories and RBT indicates that there are now

numerous contributions that explain different aspects of the nature and existence of

the firm; however, the search for a unified theory still remains a significant

challenge. Thus, this review also highlights the opportunities to deploy the relative

strengths of each theory in a theoretical framework or set of procedures in the form

of a procurement decision-making model.

It is clear from the literature review that the dominant microeconomics NIE

theories of TC, TCE, PAT and PRT and RBT from the Strategic Management field

are only beginning to be applied in construction-related sectors and aspects; see, for

example, Bridge and Tisdell (2004), Bridge (2008), Chang and Ive (2007a), Jin and

Doloi (2008) and (Sweeney, 2009). More importantly, the literature is showing a

convergence in drawing upon the complementary strengths of TCE and RBT to

better explain the boundaries of the firm. Of these approaches, Bridge and Tisdell’s

(2004) approach is considered to have a logic which is more consistent with TCE and

RBT’s key dimensions, and to be more comprehensive than the works of other

construction researchers.

The review of the PRT literature also demonstrates that there is scope for a new

procurement approach that analyses a project and breaks it into its key production

activities prior to bundling. Key aspects that lack flexibility or have high hold-up

potential can then be removed from a PPP package and be separately procured,

depending on in-house capabilities or resources compared to market capacity and

availability. The model also needs to deal with other troublesome activities that are

caused by a nominated subcontractor.

With the progress made in the application of the integration of TCE and RBT

in the field of procurement in the construction industry, the aim of this (thesis)


research is to further incorporate key contractual relations issues, such as bundling

decisions related to PRT and payment terms related to PAT. The further

incorporation of these key issues will result in a new and comprehensive first-order

decision-making model for major infrastructure procurement.

In summary, there is appreciable scope to develop a new first-order

procurement decision-making model which deploys the integrative framework of

TCE and RBT (Bridge, 2008; Bridge & Tisdell, 2004) in make-or-buy analysis (risk

allocation) and exchange relationship analysis, including PRT in bundling analysis,

and PAT in contractual payment terms. This procurement model deploys production

cost and benefit theory from the field of Strategic Management, and from theories

concerning transaction costs from NIE (including Bridge and Tisdell's approach to

integrating TCE and RBT).

The model serves to identify a procurement approach that represents an

efficient configuration of: risk allocation in terms of delineating internalised and

externalised activities, bundle(s) of externalised activities, and the nature of the

external exchange relationship with each externalised contract(s) for the project

concerned. In turn, the procurement approach informed by the procurement model –

either PPP or non-PPP – is more likely to deliver superior VfM, or a better ratio of

production benefits to production and transaction costs, across the whole-of-life of

the asset, than procurement approaches that are determined by MAUA.

The schematic of this first-order procurement decision-making model is

developed and presented in the next chapter. The model addresses the make-or-buy

decision (risk allocation), the bundling decision (property rights incentives), and the

exchange relationship decision (relational to arms-length exchange) in a novel

approach to articulating a procurement strategy designed to determine the suitability

(or otherwise) of PPPs. The procurement model incorporates the NIE approaches to

facilitate procurement selection in pursuance of VfM in relative terms or, at least, to

serve as a starting point from which departments or agencies can deviate for

justifiable reasons. In particular, this procurement model is based on falsifiable and a

priori theory, and provides a clear connection between data gathering (including

risks associated with project characteristics, and the relative project capability and

competence of the government and private sector) and the procurement approach.

Chapter 3: Theoretical development 81

Chapter 3: Theoretical development

3.1 Introduction

Thus far, the opportunity to address the weaknesses of MAUA using a number

of NIE theories and RBT applied to key aspects of procurement approaches (namely,

make-or-buy, bundling and exchange relationship) has been established. This chapter

now develops the first-order procurement decision-making model by configuring

these theories into a set of procedures in terms of make-or-buy, bundling, and

exchange relationship, and with respect to activities arising from a project.

That said, the model can be broadened to address the possibility of a program

of projects that could change the balance of advantage; say, for example, from

government to market in remote areas. The model can be used in two ways: i) in the

review (or explanatory) mode to evaluate the procurement decision in a completed

project; and ii) once sufficiently tested, in a predictive (or normative) mode to

identify the suitability of PPPs (or otherwise). The model will thus increase the

chance that the selected procurement delivers superior ratio of benefits or front-line

user utility and whole-life costs relative to competing modes of procurement.

Finally, EoI as a key indicator of competition and flexibility and, in turn, as a

critical proxy of market failure and VfM, is developed as the dependent variable in

terms of testing the model and addressing the hypothesis developed in this chapter.

3.2 Procedures in the first-order procurement decision-making model

3.2.1 Overview

The procurement model comprises three stages. Stage 1 involves an iterative

process concerning the three tasks of activity analysis, make-or-buy analysis, and

market analysis. Stage 2 then involves bundling analysis, and Stage 3 involves

exchange relationship analysis which, finally, informs the procurement strategy. The

procurement model is illustrated in Figure 3.1.

82 Chapter 3: Theoretical development

Figure 3.1: Tasks in the first-order procurement decision-making model

The model arrives at a procurement strategy by analysing make-or-buy

decisions within the project at the activity level, rather than analysing the project as a

whole. That is, the model starts with a focus on key production activities of design,

construct, operations and maintenance in Stage 1/Task A: Activity analysis. With this

knowledge of the key activities in hand, Stage 1/Task B: Make-or-buy analysis

deploys the integrative framework of vertical integration to measure the relative

technical and/or organisational capabilities and competencies of each activity across

government and private sectors, to determine the party which can better manage the

activity.

Initially, these tasks in Stage 1 were envisaged in a linear sequence but are now

developed as part of an iterative Stage 1 process. This iterative process allows

activities associated with oligopoly; duopoly; or monopoly market structures to be

identified in Stage 1/Task C: Market analysis. In the first iteration, when an activity

has been generated on the basis of size or scale of work within the activity

concerned, then a second iteration is undertaken involving breaking-up the

activity(ies) to correspond with the capacity limits of the next lower tier of firms

specialising in this activity.

In subsequent Stage 2: Bundling analysis, activities with high hold-up potential

are excluded from bundling with the other activities on the basis that government is

better placed to manage potential hold-up in these activities if it directly engages

Stage 1 (1st-order)

Stage 2 (2nd-order)

Initial Schematic Design

Task A Activity Analysis

Task B Make-or-Buy Analysis

Procurement Strategy

Task C Market Analysis

Exchange Relationship Analysis

Bundling Analysis

Stage 3 (3rd-order)


and/or collaborates with the private sector firms supplying these troublesome

activities. Finally, Stage 3: Exchange relationship analysis determines the most

efficient exchange relationship between the government with the private sector firm

at the head of the supply chain or bundle to determine a procurement strategy.

In summary, the model develops an efficient procurement strategy by

identifying externalised activity, optimal bundling of externalised activity, and

selecting the most efficient exchange relationship with each bundle, based on NIE

theories and RBT. In so doing, the model minimises whole-life costs and maximises

utility or benefits from the operations of the facility, in pursuance of VfM in relative

terms. The following section explains the operational procedures in each stage of the

first-order decision-making model, to arrive at an efficient procurement strategy.

3.2.2 Stage 1/Task A: Activity analysis (first-order)

The procurement model begins by identifying key production activities in the

design, construct, operations and maintenance of an infrastructure project at a first-

order level. That is, the project is broken down into its key activities, using

transaction costs and production cost-benefit logic. According to TCE, a transaction

cost occurs when goods or services are transferred across a technologically separable

interface (Williamson 1985), with distinct technology and distinct knowledge or skill

sets. This creates a natural division of labour, and the extent to which the division of

labour occurs is explained by classical theory of production; that is, the extent of the

market demand that generates scale economies, including the accumulation of

knowledge or learning curve economies. In turn, this justifies investments made in

special purpose technology.

Deploying this logic, which is different from identifying trade packages using a

work breakdown structure, an infrastructure project can be broken down into

activities that correspond with the highest level of market specialisation. In other

words, if there are existing market firms that specialise in an activity that lies within

the boundaries of the project, an activity has been identified. For example, a road

project might comprise a number of technologically separable major work packages

or supply chains, including major and minor civil engineering works, building works,

mechanical and electrical works, and a tolling system. Alternatively, a project may

comprise only one work package or supply chain. Within each work package or

supply chain, there can be several market firms that specialise in some part or all of


the management, and/or design, and/or construction, and/or operations and

maintenance associated with the work package.

The extent to which the project is decomposed into activities depends in the

first instance on the local market and requires some iterative analysis of the project

and the market. This approach to activity analysis should be differentiated from other

methods, such as work breakdown structure, because of the underlying differences in

decomposing the project.

More specifically, the model prompts or guides the user to identify first-order

or key activities which: 1. have distinct technology requiring distinct knowledge or

skill sets, and 2. are non-trivial in terms of significant cost relative to cost of project.

In the process of identifying key activities, it is important to note that the

distinguishing features of various key activities lie in their discrete technological

boundaries, their distinct knowledge base or skill sets, and that their dominant source

of adding value is the technology associated with design, construction, operations

and maintenance. These distinguishing features are distinct from: a milestone in a

program; a schematic design which covers many consultant disciplines; a building

element which covers many trades; an organisational and/or management activity

which includes planning or programming; and supply chain management (including

procurement) as the dominant source of adding value. None of these is considered to

be a key activity.

In addition to the defining rules of first-order key activities, the scope of

activity analysis is delimited to the following activities upstream and downstream in

a project:

• From, and including, activities arising from schematic design (hence upstream

activities – such as planning and surveying a site, geotechnical survey and other

activities that inform the schematic design – are excluded)

• To, and including, maintenance and operation of the built asset, but excluding

core or front-line activities that are not directly related to the physical asset of

the project (for example, downstream activities, such as clinical services, are

excluded; as are activities around design, supply, installation or positioning of

loose and plug-in equipment; along with ancillary services that do not interact

with the building or asset such as, linen and porterage services in hospitals)


• From the period of time starting with the conclusion of the schematic design, to

the point the asset is deemed obsolete and requires rehabilitation or major and

extensive refurbishment (Hence, rehabilitation and refurbishment works are

also excluded)

3.2.3 Stage 1/Task B: Make-or-buy analysis (first-order)

a) Applying the integrative framework

Next, the model determines the make-or-buy decision for each activity; that is,

the model determines whether to locate an activity within or outside the firm. In the

context of this research, the focal firm equates to the government and, therefore, this

decision determines the vertical boundaries between the public and private sectors in

infrastructure projects. More specifically, the make decision, or internalisation, is a

mode of operation in which the government is able to exert direct control over

resources, and is wholly responsible for an activity. Thus, this make definition can

include a contract of employment and a government agency or a government

subsidiary where government owns more than 50 percent of the subsidiary; the buy

decision, or externalisation, on the other hand, comprises all other modes of

operation.

Based on the integration of TCE and RBT, Bridge and Tisdell (2004) develop

the concept of a capability and competence spectrum between the firm (or

government) and market, which classifies seven levels of capability and competence.

In Bridge (2014a), the integrative framework of vertical integration (henceforth

referred to as ‘the framework’) has been updated for calibration purposes (as shown

in Table 3.1). However, there has been no material change to the previous versions in

Bridge and Tisdell (2004) and Bridge (2008).

The classification levels are on a continuous scale of 1 to 7 in terms of relative

competitive advantage; however, these are overlayed by a categorical scale of the

reasons associated with each level. To elaborate, the firm (or government) and

market have varying capabilities in relation to certain activities, so that a capability

or competence (RBT) logic dominates (Levels 1, 2, 6, and 7), and reflects the

minimisation of production costs and maximisation of production benefits. On the

other hand, the firm (or government) and market might display similar levels of

capability and competence relative to an activity, in which case, a transaction cost


(TCE) logic (including bureaucracy costs and hold-up) is dominant in terms of

assigning the activity to either government or private sector to minimise transaction

costs (Levels 4a and 4b). At Levels 3 and 5, the frequency variable from TCE and

value variable from RBT dominate in explaining the relative capability or

competence; and these levels illustrate Coase’s TC theory.

Each level is represented by a distinctive pattern, comprising six variables or

measurements – three variables from RBT, and three from TCE. There are eight

theoretical patterns, with Level 4 having two different patterns in the framework. The

RBT measurements are concerned with the relative capability and competence of

government and private sectors with respect to each activity, and the TCE

measurements are concerned with the physical and technological attributes with

respect to each activity. The eight theoretical patterns reflect eight logical reasons to

make or internalise (Patterns 1 to 4a), or buy or externalise (Patterns 4b to 7) an

activity.

By applying the TCE and RBT measurements to each activity in the project, an

empirical pattern for each activity is generated; this is then matched with the closest

of the eight theoretical patterns. This matching then indicates whether the activity

should be internalised or externalised to achieve greatest effectiveness and

efficiency, including the most efficient allocation of risks.

Table 3.1 Detailed measurement of activity levels Adapted from “T he determinants of the vertical boundaries of the construction firm,” Construction Management and Economics, 22(8), 807-25, by A. Bridge and C. Tisdell. Reprinted with permission from Taylor & Francis, Copyright © 2004 Routledge, and from Bridge (2008, 2014a).

Level Logic Value (Capacity)

(RBT)

Rarity

(RBT)

Costly to Imitate

(RBT)

Asset Specificity

(TCE)

Uncertainty

(TCE)

Frequency

(TCE)

Mode of Governance

1 Capability (RBT)

+++ Score: 7

+ to +++ Score: 5 to 7



0 to +++ Score: 1 to 7


Internal

2 Competence (RBT)

++ Score: 6


0 Score: 1 to 4

0 to ++ Score: 1 to 6



Internal

3 Transaction Costs/ Competence

(Coase)

+ Score: 5

0 Score: 1 to 4

0 Score: 1 to 4

0 to + Score: 1 to 5

0 Score: 1 to 4


Internal

4a TCE/Hold-up (Williamson)

-/+ Score: 4

0 Score: 1 to 4

0 Score: 1 to 4




Internal

4b TCE/Hold-up (Williamson)

-/+ Score: 4

0 Score: 1 to 4

0 Score: 1 to 4



0/+ Score: 1 to 4

External

5 Transaction Costs/ Competence

(Coase)

- Score: 3

0 Score: 1 to 4

0 Score: 1 to 4


0 Score: 1 to 4

0 Score: 1 to 4

External

6 Competence (RBT)

-- Score: 2


0 Score: 1 to 4



0 Score: 1 to 4

External

7 Capability (RBT)

--- Score: 1





0 Score: 1 to 4

External


The following translates the levels in the integrative table into the context of

this research.

• Levels 1 and 7 are at the extreme ends of the spectrum, where Level 1 refers to

activities in which government is technically and/or organisationally superior to

market (that is, not able to be matched by the market in the short term, or at the

time when the procurement decision is made). Conversely, Level 7 represents

activities when the market is technically and/or organisationally superior to

government (that is, not able to be matched by government at the decision

date). At Levels 1 and 7, RBT variables (that is, valuable, rare and costly-to-

imitate variables) are dominant. At Level 1, the focal firm can exploit hold-up

(if asset specificity and uncertainty are high), and its buyers have no choice but

to do business with the firm. At Level 7, on the other hand, the focal firm has to

buy Level 7 activities from the market firm and can be vulnerable to hold-up.

The only way to mitigate this problem is via a inefficient discrete external

exchange; that is, via incorporation of a great deal of threats relating to non-

performance, as the focal firm has no capability to understand and manage the

transaction relationship. At Levels 1 and 7, using the threat of not allocating

future work to the contracting firm, might not be effective as a deterrent to

hold-up problems because of a small supply of firms at these levels. For this

reason, TCE does not explain make-or-buy at Levels 1 and 7.

• Level 2 represents the situation where government is technically better than the

market in terms of delivering less costly and/or higher quality products or

services, arising out of access to resources or tacit knowledge in resources.

Level 6 is a mirror image of Level 2 where the market is technically superior to

government. Similar to Levels 1 and 7, RBT variables are the dominant

variables, and the competitive advantage in Levels 2 or 6 is derived from

technical systems or procedures and firm-specific tacit technical knowledge

that cannot be readily sourced or imitated. At Levels 2 and 6, there can be

some possibility of threat of hold-up, given significant firm versus market

differences around access to rare resources. As the market is an oligopoly at

Levels 2 and 6, the threat of denying future work can be a more effective check

on hold-up problems than at Levels 1 and 7.


• In Levels 3 and 5, resources can be more readily sourced; thus, these levels are

differentiated from Levels 2 and 6. Level 3 represents government being

organisationally superior to the market in terms of providing less costly and/or

quicker response (but not superior quality) in the delivery of the activity, based

on managing resources within the activity. For example, if the government

employs a market firm to be available for work of unpredictable nature, this

market firm might be idle and not contributing directly to the activity and/or be

elsewhere while, at the same time, adding overheads to the government. In

addition, the government is not able to directly control the market firm’s

operatives, and this leads to inefficient implementation of the activity. On the

other hand, if the government attempts to tender this work in an ad-hoc

fashion, it will incur higher external transaction costs associated with finding

and appointing contractors, delays in implementing the activity (due to minor

variations), and costs related to contract management. Hence, for activities that

require a less costly and quicker response, the government is better positioned

than the market to carry out the activity internally, as represented by Level 3 of

the spectrum.

• Level 5, in contrast to Level 3, represent activities where the market is

organisationally superior to the government in terms of providing less costly

and/or quicker response, but not superior quality in the delivery of the activity.

Again, this is based on managing resources within the activity. This type of

work is more predictable and, more importantly, market firms have the

competitive advantage over the government in terms of accessing or

aggregating higher levels of work to generate flow, which includes: planning,

coordinating, procuring and management of supply chain, and addressing

potential negative opportunistic behaviour of suppliers. Therefore, market firms

could fundamentally display more organisational competence and make fewer

mistakes.

• In summary, Levels 3 and 5 are concerned with the economic efficiencies of

management and/or organisation of the resources within the activity, and speak

to the matter of capacity. If the firm chooses to internalise or increase capacity

surrounding the activity and demand drops (lower frequency), then government

or firm may incur the cost inefficiencies of additional overheads and

operational costs. Conversely, if the demand increases and there is a lack of


capacity, then more mistakes can occur as a result of overstretched supervision

or organisation. The decision to internalise or externalise depends on the

critical threshold of work, so as to justify the investment required in developing

the capabilities. At Levels 3 and 5, there is very little hold-up potential given

the nature of the transaction where there is a plentiful supply of firms, and

hold-up can be effectively checked by the threat of withholding future work

(Coase, 2000, p. 30).

• Levels 4a and 4b represent a potential negative opportunistic situation arising

where the market holds up the government by seeking to secure more reward

and/or costly variations (time and/or money) in, and/or arising from, the

activity; for example, in a case where the activity is critical, and an emergency

response is required.

b) Developing the combination of the value variable in RBT and the frequency variable in TCE

The procurement model is able to explain either internalisation or

externalisation based on the in-house capacity pertaining to the activity which

prevailed at the date on which the actual approach to procurement was decided;

however, at the same time, the model also indicates whether this level of capacity is

likely to have been economical (that is, right level of capacity) or not economical

(that is, too much capacity or too little capacity). More specifically, the model is able

to determine this by examining the answers to, and resultant symbols for, the value

(or capacity) variable and the frequency variable. In other words, there is a close

relationship between these variables.

The value (or capacity) variable has been accepted by Barney (2001a) as the

weakest among the predictors around the RBT variables, and is charged as being

tautological by Priem and Butler (2001). For example, a firm might internalise an

activity because this activity improves the firm’s profits (value), and to improve

profits (value) the firm internalises the activity. Moreover, the value variable from

RBT corresponds, or co-varies, exactly in proportion to each of the levels in the

framework and, thus, could be the only variable needed to predict make-or-buy.

However, this case is unlikely, given the very different logics behind the three

theories being used, and the fact that each theory has something important to

contribute.


The value variable is then similar to the capacity argument, such that a firm –

or, in this research, state government – might internalise an activity because of its

current capacity, or externalise an activity because of lack of current capacity.

Capacity can then be misleading in pursuance of VfM, and be wielded in a non-

economical way influenced by political or financial concerns. However, given the

explanatory merit of capacity, the procurement model makes the value variable

capture capacity but is mindful of its economic prediction weakness. It is important,

therefore, to read the result for the value or capacity variable in conjunction with the

result for the frequency variable, which gives a more objective economic indicator.

If the charge of inherent tautological weaknesses with the value or capacity

variable is accepted, then the frequency variable represents the next best or most

powerful variable in terms of being able to superficially and singularly (that is,

without reference to the other variables) explain internalisation or externalisation.

However, it is not this straightforward, and is complicated by another issue. If the

firm (or state government) simply keeps increasing its staff for activities that are

large and recurrent (elements of the frequency variable), and/or keeps adding

activities that are large and recurrent to its portfolio of internalised activities, then it

increases its bureaucracy costs, and costs associated with low power incentives. On

the whole, this amounts to increasing internal transaction costs. In other words,

Coase (1937) explains that, where internal transaction costs become too great, then

externalisation of all and/or part of large and recurrent activities (make and/or buy),

or not internalising the next large and recurrent activity at the margin, might be

efficient. This also explains the reason for not having one big firm or government in

market economies (Coase, 1988c).

In summary, a high frequency score on an activity (large and recurrent) begins

to suggest that this activity could be efficiently internalised; however, this is not the

entire story. To help complete the picture, a high frequency score on a particular

activity needs to be read in conjunction with (at least) a high score on the asset

specificity variable; furthermore, if high scores are also read on the other variables

on an entire pattern of scores in one of the rows in Table 3.1 (above), then the

economic case for internalisation becomes compelling.

Ideally, the value or capacity and frequency variables will positively correlate

or match, thus indicating approximately the right level of internalisation or


externalisation; if not, and they negatively correlate or mismatch, then this indicates

too much or too little capacity (or internalisation or externalisation). However, two

exceptions to this can arise:

1. Firstly, the issue of proximity: If a state government agency or subsidiary is a

subcontractor to a main contractor, the project or contract can be regarded as de

facto internalised. This could happen when the overall project is too large for

the state agency or subsidiary, yet the key activity is within range of the state

government or its subsidiary, and is physically in the middle of the other key

activities. The latter situation makes it difficult to organise this activity in the

midst of other activities which are being organised by an external firm; this, in

turn, can lead to the contractor claiming that it is being held-up by the state

agency or subsidiary.

2. Secondly, the issue of dynamic efficiency: There is also the possibility of the

state government giving work to the private sector to keep business turning

over. This rationale can help to explain a situation where an activity is

externalised, even though it appears a better option for the government to

internalise that activity. The model is not able to support such instances

because it is a static approach; however, in this scenario, neither the model, nor

theory, nor practice can be criticised as being categorically incorrect on

economic grounds, as the model promotes short-term economic efficiency,

while the practice is (possibly) promoting longer-term efficiency.

c) Key steps

In order to determine the level of key activities and whether these are

internalised or externalised, the model guides the user through the following steps:

1. Answer the questions relating to the RBT and TCE variables for each activity,

which are slightly different for internalised and externalised activity, in order to

generate a pattern of symbols for the variables or economic dimensions

representing the three theories being used; and

2. Match this empirical pattern or actual pattern to the closest theoretical pattern

in Table 3.1 (above), in order to establish an internalisation or externalisation

level for the activity.


In plain terms, this framework in the model explains the rationale for

internalisation or externalisation in terms of each of the above levels. The responses

to the questions are measured using a 7-point scale (numbers 1-7 represent scores on

a 7-point scale, and should not be confused with the 7 rationales or patterns that

explain internalisation or externalisation), and are then matched to a theoretical

pattern in the framework. As this research is testing the model in its explanatory

mode, the responses need to reflect the circumstances at approximately the point in

time at which the project procurement decision was made.

On the issue of the identification of discrete activities in the previous activity

analysis, if the user finds that there is conflict in their answers in make-or-buy

analysis – for example, if part of the activity generates different answers to those for

another part of the activity – then the model is prompting the user to split the activity

into two different activities. On the other hand, if the user finds that the same

answers to all ten questions occur for two or more activities, then the model is

prompting the user to consider combining the activities concerned. In other words,

there may be some iteration between this task and the previous task activity analysis.

This approach of identifying the party best able to manage risks associated with

an activity is a key departure from current practice. Instead of identifying and

assessing operational risks at the early stages of the project when information is

missing, the emphasis here is on resources owned by government and private sectors

relative to each project activity. Hence, this approach is able to more reliably

anticipate which party is better able to manage the risks associated with each activity.

3.2.4 Stage 1/Task C: Market analysis (first-order)

In this task, the model prompts the user to perform a check or Structure-

Conduct-Performance (SCP) analysis of each activity to corroborate the levels

identified through the framework. Level 7 activity represents the industry or market

as: having an oligopoly, duopoly, or monopoly market structure with a small number

of competitive firms (structure); possessing the market power to set prices and

(possibly) engage in tacit collusion (conduct); and sustaining above normal economic

returns (performance) through sustainable competitive advantage. Level 6 activity

represents the industry or market as: being of monopolistic competition with many

competitors (structure); predominantly focusing competition on technical

competence, such as cost leadership or product differentiation (conduct); and


achieving above normal economic returns (performance) through temporary

competitive advantage. Level 5 activity represents market at perfect competition –

that is, having a large number of competitors (structure). Because the resources are

easily accessible, most firms at this level are price-takers (conduct) and obtain

normal economic returns (performance). A summary of the SCP in conjunction with

the levels in the framework is given in Table 3.2.

Table 3.2 SCP analysis in relation to activity levels Adapted from “The determinants of the governance of air conditioning maintenance in Australian retail centres,” p. 249, by Bridge (2008), Queensland University of Technology. Reprinted with permission.

Level Structure Conduct Performance Competitive advantage

Level 7 • Monopoly/duopoly/oligopoly • Few or limited competitors • Costly entry and exit

• Market power to set prices

• Tacit collusion

Above normal economic profits

Sustainable competitive advantage

Level 6 • Monopolistic competition • Many competitors • Low cost entry and exit

• Cost leadership

• Product differentiation


Temporary competitive advantage

Level 5 • Perfect competition • Large number of competitors • Low cost entry and exit

• Price-taker Normal economic profits

Competitive parity

Level 4b • Many competitors • Potential hold-up behaviour


Competitive/ organisational parity

The model then prompts the user to consider any Level 7 activities arising out

of the large scale of this activity’s work in the project. Level 7 activity arises when

the activity is of large scale or size, and/or requires a rare technology, which limits

the number of market firms that are capable of carrying out the activity, resulting in

limited competition. These level 7 firms can exhibit opportunistic behaviour due to

their monopolistic, duopolistic or oligopolistic position with respect to the activity,

leading to high market power to set prices ex ante (or pre-contract) and cause hold-

up problems ex post (or post-contract).

If any Level 7 activities are identified, the model then guides the user to

attempt to break down the activity concerned into two or more, smaller or sub-

activities to reach the next tier of firms specialising in this activity, subject to further

iterations, until there are no Level 7 activities arising out of scale. This avoids lack of

competition due to size or possibly a more powerful sub-market firm (such as a

subcontractor) as part of the supply chain managed by a different upstream market

firm (for example, a main contractor). This iterative process of identifying and


breaking down Level 7 activities arising out of scale, addresses the ‘size’ issue,

which is noted by the House of Lords Select Committee on Economic Affairs (2010)

as one of the two key determinants of whether or not a project is suitable as a PPP.

3.2.5 Stage 2: Bundling analysis (second-order)

Having partitioned the project into internalised and externalised activities, the

focus shifts to the procurement of externalised activities. Here, given that the model

is being deployed in an explanatory mode, it could include an activity in this stage

that was actually internalised by treating it as externalised (that is, a mismatch

between the capacity and frequency scores, where capacity is positive but frequency

is low); conversely, it could exclude an activity that was actually externalised and

treat this as internalised (again a mismatch between capacity and frequency but, in

this case, capacity is negative and frequency is high).

The purpose of this task is to determine the level of bundling of activities

within a project, in pursuance of minimising bureaucracy costs and minimising the

potential for hold-up (and, therefore, also improving certainty with contracted time,

cost and specification, and maximising opportunities for improvements in terms of

whole-of-life production costs and benefits, through the alignment of incentives in

design, construction, operations and maintenance). The challenge in terms of

achieving optimal bundling is that, on the one hand, minimising bureaucracy costs

involves greater use of single-point contact with a private sector firm across a wide

range of activities (transfer of control to private sector firm), while on the other hand,

minimising hold-up involves the greater use of internalised management (or use of

agents) and control over private sector firms. Thus, there is tension or trade-off in the

extent to which both types of transaction costs can be minimised.

To address this issue, the model prompts the user to consider any potentially

troublesome activities that will lead to government becoming dependent or

vulnerable, and market firms becoming more powerful; that is, Level 4b activities

that can potentially cause hold-up ex post, arising out of high asset specificity and

high uncertainty. The model also guides the user to consider any residual Level 7

activities arising from size that cannot be further broken down because of physical or

proximity issues, and any Level 7 activities arising out of rare technology which can

lead to a lack of competition, and market power to set prices ex ante; this, again, can

cause hold-up problems ex post. The identification and exclusion of activities with a


very high potential for hold-up arising from a very high level of asset specificity and

very high level of uncertainty or unpredictability, addresses the second issue of

‘flexibility ’ or predictability, also nominated by the House of Lords Select

Committee on Economic Affairs (2010) as a key determinant of the suitability as a

PPP.

More specifically, the model prompts the user to consider if any of the

potential negative opportunistic behaviours by market firms associated with these

troublesome activities can be checked by the firm being engaged at the head of

supply chain or by the contract that includes these activities. That is, the firm at the

head of the supply chain, such as the main contractor who is coordinating or

subcontracting these troublesome activities, uses the threat of withholding future

work to address any potential negative opportunistic behaviour. If this approach is

not taken, then the model suggests that government either: 1) craft a special

relationship with the firms providing these troublesome activities through the firm

being engaged at the head of the supply chain that includes these troublesome

activities (for example, use of Prime Cost Sum and a nominated subcontractor or

supplier arrangement) which then drops down into an exchange relationship

consideration; or 2) a separate contract between the firm or state government and the

market firm providing the troublesome activity (in other words, excluding the

troublesome activities from any subsequent bundle).

The model then guides the user to bundle the remaining key activities into

main activities of design (D), construction (C), operations (O) and maintenance (M).

The user is now effectively undertaking second-order analysis (the point at which

current practice often begins) in assessing the activities involved in organising or

managing each of these main activities; for example, main contractor on

construction. Again, the model prompts the user to attempt to break down each main

activity if the size and/or complexity of a main activity leads to Level 7 (again,

leading to lack of competition and market power to set prices ex ante and hold-up

problems ex post).

From here, the user is able to identify any potentially viable major activities;

that is, bundle(s) of DCO, DCM, or DCOM main activities that can be market-

sounded as suitable for either PPP(s), or contract(s). If these cannot be identified,

then the default is separate D, separate C, separate O and separate M main activities


organised by government (either directly and/or using market firm, such as project

manager or consultant). In other words – and in terms of the key mechanism being

deployed to realise the benefits of the integration of Operations and Maintenance in

Design and Construction in pursuance of increased benefits and lower whole-life

costs – the model considers that private finance holds the greatest high power

incentive, followed by contract. If the market is not responding positively to using

these mechanisms, then bureaucracy is used to perform the integrating role.

An alternative to seeking to integrate operations, and/or maintenance, with

design and construction occurs when neither the buyer (government) nor supplier(s)

(private sector) has appreciably differential advantages with respect to operations

and/or maintenance, and when a conventional design (without significant

innovations) – which is focused on minimising capital cost – would not significantly

undermine whole-life costs. In this case, the model prompts the user to consider a

Design and Construct approach (or variation of this approach, including Managing

Contractor or Early Contractor Involvement).

In summary, there are three levels of bundling of production activities:

1. Key activities = about highest level of market specialisation at the nature of the

key activity

2. Main activities = Set of D, or C, or O, or M key activities (different degrees of

bundling of D, C, O & M key activities)

3. Major activities = Set of DCO, or DCM, or DCOM main and/or key activities

Each bundle includes the organisation and management of the activities; that is,

the planning and coordination of: tasks within activities, key activities within main

activities, and main activities within major activities. Hence, in one project, there

might be many project managers or contractors operating at different levels. Project

Managers or contractors are in the position in a supply chain to manage supply and

demand and amalgamate construction work volume; hence, they have high market

power to organise the supply chain and obtain economies of scale, and efficiencies of

time and costs. The Project Manager or contractor co-ordinates material flow

bottom-up from subcontractors and suppliers, and co-ordinates information flow

from client top-down. Organisation and management can also include procurement

and supply chain coordination or control; for example, Quantity Surveyors and

Project Managers.


In a contract, organisation and management activity occurs in terms of:

• Planning, procurement, coordination and control of timing and budgeting of tasks within key activities

• Planning, procurement, coordination and control of timing and budgeting of key activities within a main activity

• Planning, procurement, coordination and control of timing and budgeting of main activities within a major activity

Hence, there are multiple layers, and boundaries within these layers, in the

management of any one project. The outcome of this stage might comprise the

delineation of multiple bundles of key activities, and/or main activities, and/or major

activities to form the basis of one or more contracts. This task highlights a key

difference between this and the current procurement approach. The latter begins with

the organisation (or management) or second-order level, whereas the first-order

approach in the proposed procurement model considers management and

organisation only after the bundling of activities.

3.2.6 Stage 3: Exchange relationship analysis (third-order)

Finally, the model guides the user to consider crafting an efficient exchange

between the buyer or government and each firm or counterparty in each contract. To

do this, the model combines the strengths of both TCE and PAT. Indeed, Eisenhardt

(1989) urges that PAT be adopted to investigate the principal-agent relationship in a

governance structure, including contractual payment terms. In other words, efficient

governance in the exchange includes both seeking to pre-empt vulnerability and

negative opportunistic behaviour in the exchange (or hold-up), in this case by the

supplier and, at the same time, seeking to encourage positive behaviour through

incentives that help align interest between principal (the government in this research)

and the agent (the counterparty in each contract). Both approaches reduce external

transaction costs – hold-up costs, monitoring costs, and shirking costs. Additionally,

PAT is concerned with the development of an exchange that increases benefits

beyond reducing costs.

In the first step of this task, TCE as a theory of governance is deployed by the

model by re-measuring the TCE variables in respect of each contract to broadly steer

the exchange towards either an arms-length (discrete exchange) or a relational

exchange. The approach to measure the variables is based on the TCR trinity


developed by Bridge (2008), as shown in Figure 3.2. This measurement now includes

the task of planning and coordination across key activities. Thus, this measure of

main and major activities can produce a different result on the TCE variables than

that recorded across the TCE results measured in the constituent key activities.

Key: +++ = Extremely high incidence of variable ++ = Very high incidence of variable + = High incidence of variable 0 = Low high incidence of variable

Figure 3.2: Exchange relationship Adapted from “The determinants of the governance of air conditioning maintenance in Australian

retail centres,” p. 113, by Bridge (2008) Queensland University of Technology. Reprinted with permission, and Bridge (2014b).

In Box 1, activities have a very high potential for hold-up due to high asset

specificity and high uncertainty, that is Level 4b activity, and government can more

efficiently seek to control these activities using more relational exchange, such as

collaborative approaches or share risks in an alliance fashion.

Main or major activities dominated by Level 5 and 6 activities are reflected in

Box 2, and these activities can yield an efficient transfer of risk and control to the

contractor or consortium concerned using discrete efficient exchange. Standard

neoclassical contracts can be deployed to obtain a fixed price for construction only,

and/or DC, and/or DCOM package. If a DCOM contract with private finance is being

used in the case of Pattern 5 and 6 activities, then a conventional PPP arrangement is

appropriate.

Box 3 reflects activities with a very high potential for hold-up; that is, Level 7

activities. However, unlike Level 4b, government lacks in-house capability and lacks

access to agents to effectively collaborate with the contractor or consortium in terms

of design, construction, operations and maintenance. In transferring control to the

contractor or consortium, government can seek assurances through writing contracts

Efficient Exchange (naturally occurring)

+++ Relational +

Inefficient Exchange

+ Discrete +++

Efficient Exchange

+ Discrete +++ External Exchange

TCE Asset specificity +/+++ Uncertainty +/+++ Frequency +/0

0 0 0 +++ +/0 0

+ +++ + +++ +/0 0

a cb

1 3 2


with credible threats concerning compliance with contracted performance; hence, a

very arms-length and discrete exchange with the contractor or consortium.

In the continuum between relational exchange and discrete exchange in Figure

3.2 (above), connections can be made with three broad categories of contract in PAT,

as shown in Table 3.3 (below).

Table 3.3 Summary of outcome, hybrid and behaviour-based contracts in PAT

Behaviour-based Hybrid (outcome and behaviour- based)

Outcome-based

• Cost-plus • Low power incentive • Principal's risk to

completion • Agency cost

o specification of behaviour

o monitoring of behaviour o outcome uncertainty o high project complexity o less goal alignment

• Target outturn costs or guaranteed construction sum linked with gain-share or pain-share regime

• Risks balanced between agent and principal

• Suitable for outcome uncertainty


o specification of outcomes

o verification of outcomes o risk premium o suitability for

information asymmetry o outcome certainty o better goal alignment

To explain this further, there is a point in relational exchange that is envisaged

to be best governed by internalisation and exchange within the firm; this is depicted

by the dotted part of the line to the left in Figure 3.2 (above). This part of the line can

be connected to behaviour-based contracts in PAT. The solid part of the line starting

from the left in this figure can be connected to contracts that are somewhere between

behaviour-based contracts and outcome-based contracts (or hybrid contracts) in PAT.

The remaining part of the solid line in the figure can be connected to outcome-based

contracts in PAT. By degree, the further to the right of the solid line, then these

contracts incorporate greater information asymmetry in favour of the supplier (due to

a lack of knowledge held by government to collaborate in the exchange); therefore,

the power is held by the supplier. In this case, a greater level of credible threats

becomes necessary to pre-empt hold-up. At the same time, however, these are costly

to write and are priced by the supplier; consequently, this is not an efficient exchange

at one level. Nonetheless, it is, overall, the most efficacious approach to governance

in terms of addressing the risks in the exchange. Therefore, in addition to bundling

and excluding troublesome activities from the main bundles, and connecting TCE to

the three types of contracts in PAT, the model further addresses the second issue of

‘flexibility’ raised by the House of Lords Select Committee on Economic Affairs


(2010), and informs the buyer or government on the degree of flexibility that can be

exercised on these troublesome activities.

Finally, having carried out the above procedures, the model predicts a

procurement strategy, in terms of: 1. size of project; 2. extent of bundling; and 3. the

nature of the exchange relationship (including payment terms) between government

and private sector firm at the head of each bundle or contract within the project.

3.2.7 Outcomes and merits of first-order procurement decision-making model

In summary, the model starts with a focus on the key production activities of

design, construction, operations and maintenance and is, therefore, a first-order

analysis. This is in contrast to current practice that starts at an organisational level;

for instance, a second-order analysis where a main contractor or managing

contractor manages across key activities. The first-order approach responds to the

critique of inappropriate risk analysis associated with current practice by avoiding

being too focused on individual risks. Unlike current risk analysis, the first-order

approach does not attempt to identify or guess at individual risks at the early

schematic stages of a project when information is missing, and later second-guess

when government or its agents attempt to assess the nature and value of risks that are

better understood by other players, such as contractors. In other words, the

procurement model, which indirectly performs risk allocation at the activity level, is

an appropriate level of analysis in terms of accessing who is better at dealing with

inherent risks that might occur within the activity, and is more reliable than a model

that prematurely guesses the nature and extent of potential risk events.

At the same time, this approach provides a transparent framework that is based

on economic principles. In contrast, in MAUA-based approaches, risk analysis can

be overshadowed by the main concern coming the other way; that is, the matching of

key opening day outcomes to the procurement mode. Furthermore, this concern is

expressed at the organisational or second-order level, and undermines the appropriate

level of resources being allocated in the risk analysis.

In addition, the first-order analysis answers Chang’s call for an appropriate unit

of analysis that is mediated by governance (Chang, 2013) and, in so doing, leads into

a make-or-buy analysis around the key activities. This is the conventional TCE

approach that views transaction costs across technologically distinct boundaries;


thus, the unit of analysis is mediated or affected by governance structures (markets,

hybrids and hierarchies), and lends itself to make-or-buy analysis. This is effectively

an indirect approach to risk analysis; that is, it does not identify individual risks, but

rather – mindful of the information gap inherent in a schematic design – adopts a

more appropriate focus on who is best placed to deal with potential risks arising out

of an activity).

More specifically, the model deploys three theories in Stage 1/Task B: Make-

or-buy analysis. One of these theories is the dominant theory of procurement in the

field of Strategic Management (RBT), and two are Nobel Prize-winning theories

(TCE and TC). The three theories are explained as follows:

1. RBT addresses technical or production risks by determining which party is

better equipped to deliver the key activity and, therefore, better able to manage

potential risks arising out of the key activity. And which party is better able to

secure sufficient demand or workload around the activity to generate

economies of scale and/or learning economies to more efficiently deliver the

activity. (A dominant Strategic Management theory)

2. Coase’s transaction costs (TC) thesis focuses on internal transaction costs and

reveals which party is better placed to manage an activity and deliver the

activity more efficiently. (Nobel Prize for Economics, 1991)

3. TCE focuses on external transaction costs and surfaces potential exchange risk

arising out of hold-up as a function of a specific investment by one party (in

this case, state government) in the key activity. This creates a dependency of

state government on the supplying party, and means that the non-investing,

supplying party is the less vulnerable party and, therefore, the more powerful

party to be able to behave potentially in a negative opportunistic way on the

occurrence of a change or variation. Hence, the key variables in this theory are

asset specificity and uncertainty. (Nobel Prize for Economics, 2009)

In summary, the first-order procurement decision-making model has, at its

core, the integration of TCE and RBT to address key matters concerning risk

allocation, size, and flexibility as part of identifying opportunities to bundle design,

construction, operations and maintenance, and to address the nature of the

contractual exchange with each private sector party or entity engaged to deliver the

project or parts of the project. The procurement model is not predisposed to any

102

particular procurement mode.

within projects that can be investigated as potential PPPs.

Overall, the procurement

VfM and in relative terms (

MAUA, which creates a narrow approach to

Figure 3.3: First-order procurement decision

The strengths of the procurement

Table 3.4 Strengths of first-order procurement decision

More meaningful and reliable analysis and allocation of risk

Reduction in external transaction costs

Reductions in internal transaction costs

On the other hand, the strengths of the f

model become its weakness under the circumstances in which there

project outcome on the opening day of

example, time to opening day and/or cost certainty to opening day

case, however, the procurement

Chapter 3: Theoretical development

particular procurement mode. However, it uncovers projects or bundles of activities

be investigated as potential PPPs.

the procurement model opens up the opportunity to deliver superior

(as shown in Figure 3.3, below), and contrast

which creates a narrow approach to VfM in nominal terms.

procurement decision-making model and VfM in nominal terms

The strengths of the procurement model are summarised in Table 3.4

procurement decision-making model

Thorough analysis of relative capability or competence of parties with respect to activities (Production cost theory: RBT

Analysis of transactional attributes of project (TCE) Reduction in the likelihood of hold-up arising from the need to make sequential adaptations or variations (TCE)

Improvement in the efficiency of external exchangegreater alignment of objectives between parties (TCE & PATReductions in measurement costs to ensure compliance and low power incentives through transfer of ownership rights

A greater single point contact with an external or which leverages and opens up production benefits in terms of certainty (for example, time and cost certainty for opening day)well as production benefits in terms of improvements in(such as, lower operational and maintenance costs and higher levels of user utility) (RBT)

On the other hand, the strengths of the first-order procurement decision

model become its weakness under the circumstances in which there is

the opening day of the asset, and which must be achieved (for

example, time to opening day and/or cost certainty to opening day). Even in this

procurement model has an important role to play in terms of

Theoretical development

bundles of activities

up the opportunity to deliver superior

contrasts with

model and VfM in nominal terms

4 (below).

competence of parties BT)

up arising from the need to

of external exchange, including TCE & PAT)

Reductions in measurement costs to ensure compliance and low of ownership rights (PRT)

point contact with an external or private party, fits in terms of opening day), as

on benefits in terms of improvements in operations lower operational and maintenance costs and higher

procurement decision-making

is a visible

be achieved (for

). Even in this

model has an important role to play in terms of


creating a benchmark to predict procurement mode. Deviations from this predicted

procurement mode can then be justified in VfM nominal terms. The other weakness

in the procurement model concerns its static orientation. That said, it is able to cope

with whatever resources balance held by the government and the market, prevails at

the time the procurement decision is made with the model deployed.

The next section discusses and describes the approach to testing the

procurement model. It does this by first developing competition as a dependent

variable or proxy for VfM and, secondly, by developing a hypothesis to test the

procurement model.

3.3 Approach to testing the first-order procurement decision-making model

3.3.1 Developing the measurement of competition and flexibility as a dependent variable and indicator of market failure and proxy for value-for-money

While there are various definitions of VfM in the extant literature, a lack of

quantitative and comparable data in operations and maintenance creates difficulties

in measuring VfM using an absolute monetary value. In cases where the concept is

not able to be measured directly, Remler and Van Ryzin (2011, p. 102) recommend

developing a proxy, that is a “measure that substitutes for another unavailable

measure”.

Competition and flexibility have been widely recognised and accepted as a key

determinant of VfM; this is particularly the case in determining the suitability of

PPPs. The UK House of Lords Select Committee on Economic Affairs (2010), with

more experience in PPPs from which to draw experiential conclusions, considers

competition and flexibility to be fundamental drivers of the suitability or otherwise of

PPPs. More specifically, the House of Lords (2010, p. 31) finds that:

The projects most suitable for private finance are those where the requirements can be clearly specified at the outset and which are of a size that consortia of private sector companies can take on their balance sheets.

In other words, size affects the level of competition and is also instrumental in

avoiding potential market failure pre-contract in terms of seeking to crystallise not

only lower prices, but also innovations that can increase utility and reduce whole-life

costs. At the same time, a balance needs to be struck, as too much competition

becomes counterproductive and can be an indicator of market failure post-contract.

This is because the market can be signalling the prospect of gains arising from poorly


specified requirements or a lack of predictable requirements, which lead to hold-up

arising from variations in the PPP long-term contract. Thus, if there is a lack of

predictability and a need for a high level of flexibility that cannot be managed

effectively by the contract – and which creates the prospect of costly variations (that

have the effect of increasing the attractiveness of the project to the market) – PPP as

an efficient mode of procurement is undermined. In PPPs then, EoIs are reflective of

outcomes and efficacy of procurement in terms of how much the approach to

procurement has affected the prospect of sufficient competition (and market failure

pre-contract), and how well the approach to procurement has attenuated the

likelihood of hold-up associated with any lack of predictability (and market failure

post-contract).

On this basis, the procurement decision-making model addresses key issues in

determining the suitability of PPP by delineating size, bundling and exchange

relationship dimensions that collectively address competition and flexibility

concerns. The model, derived from NIE and RBT, offers the potential to identify

whether projects are suited or unsuited to PPPs. In so doing, it increases the chance

of selecting the procurement approach – either PPP or non-PPP – that can deliver

superior utility and whole-life cost outcomes; in other words, VfM relative to

alternative competing modes.

Other PPP-related literature regarding surveys on critical success factors for

PPP have been conducted in many countries, including Australia, the UK and Hong

Kong. The results of these surveys also identify competition as one of the key VfM

drivers (Arthur Andersen & LSE Enterprise, 2000; Cheung, Chan, & Kajewski,

2009; Dixon, Pottinger, & Jordan, 2005; Fitzgerald, 2004).

Competition drives down marginal costs and creates downward pressure on

prices. Not only does the absolute level of competitive tension creates downward

pressure on prices, it also facilitates the crystallisation of innovations in design,

based on outcome-based specifications that impinge on the time, cost and quality of

the project, and which influence the overall performance of the project across its life-

cycle. More importantly, empirical studies on tendering in the construction industry

show a correlation between a greater number of bidders and a reduction in the value

of the lowest bid (Domberger & Rimmer, 1994; McCaffer, 1979; Ngai, Drew, Lo, &

Skitmore, 2002; Skitmore, 2002).


In addition, auction theory scholars (Milgrom, 1979; Wilson, 1977) also

mathematically and empirically demonstrate that in a sealed tender auction, perfect

competition or intense competition leads to the convergence of the true value of the

object, even when bidders have incomplete information about its value (Brannman,

Klein, & Weiss, 1987).

In particular, Gupta (2002) examined 1740 highway construction projects in

the US over a five year period, and the empirical results indicate that the absolute

level of competition creates downward pressure on price. That is, the value of

winning bids decreases as the number of bidders increases, and the effect on value

becomes insignificant when the absolute number of bidders reaches a maximum

number; in other words, a competitive threshold is reached. Gupta determines this

competitive threshold to be near and up to 8 bidders in an open tender. Thus, Gupta

(2002) has made the important contribution of an optimal level of competition that is

dependent on size and bundling of the project, and the market at the time of tender.

This implies that when the maximum number of bidders in a market is reached, there

is optimum competition, and the bids reflect the best value of risk level that the

market is willing to undertake, given the economic conditions at that time and the

level of information provided to them. Any number greater than the optimum

competition threshold can result in unnecessary tendering costs.

Significantly, Skitmore (2002) analysed ten data sets (representing 1234

projects), this time mainly from the building industries in various countries,

including US, UK and Belgium. His findings are similar to Gupta’s (2002, p. 16),

where the regression curves show that the value of the lowest bid decreases until 8

bidders, and remains constant as the number of bidders or competition increases.

Skitmore (2002) concludes that there is a tendency for the value of bid to be

significantly higher than the pretender estimate when the competition is below eight

bidders; conversely, the value of bid comes close to the pretender estimate or, in

some instances, is lower than the pretender estimate when the competition is greater

than eight bidders (as shown in Figure 3.4 below).

Both Skitmore’s (2002) and Gupta’s (2002) empirical research establishes the

validity of using competition as an indicator of testing the goodness of a model in

terms of setting it on a path towards delivering superior VfM, and suitability as a

dependent variable and proxy for VfM. In other words, empirical research provides


evidences that an optimal level of competition can indicate VfM in terms of the best

value that the market is willing to bid given factors such as market conditions and

information given. Therefore, competition is established as a valid indicator of VfM

and serves as a suitable test for whether the predicted procurement mode is setting

the project towards delivering superior VfM.

Figure 3.4: VfM against competition

Adapted from “Raftery curve construction for tender price forecasts. Construction Management and Economics,” by M. Skitmore, Construction Management and Economics, 20(1), 83-9. Reprinted with

permission from Taylor & Francis, Copyright © 2002 Routledge.

A comprehensive literature review carried out by Flanagan, Lu, Shen and

Jewel (2007) shows that there is no consensus on how to measure the concept of

competition and competitiveness, either at the project or industry level, in the

construction industry. The review shows that several studies measure the level of

competition in terms of the number of competitors in the market, and Ngai et al.

(2002) measure competition in terms of the number of potential firms in the

construction market. Studies also show that the number of firms has an effect on

competitive behaviour among firms and affects market price; that is, there is a direct

relationship between number of firms and competition (Gupta, 2002). In addition,

based on Porter’s (1980) Five Forces model, competitiveness increases with the entry

of a new firm into the market. Furthermore, the use of a quantitative measure (such

as the number of firms to measure competition level) represents an objective

evaluation.

For the purposes of this research, actual competition is measured by the

attractiveness of the project, as indicated by the number of firms demonstrating their

willingness to bid, either through open tendering or Expressions of Interest (EoI).

This avoids measuring competition based on selective tendering (which does not

(�� )

��%

Optimal competition at 8 bidders

Number of bidders (Level of competition)


reflect the actual level of interest in the market), and EoI can serve as a key indicator

of avoiding both pre and post-contract market failure that is consistent with VfM.

3.3.2 Market failure ex ante, and pre-contract market failure

With regards to neoclassical economics, one major driver for market failure is

non-competitive markets or markets with limited competitors. This leads to an

inequilibrium in allocation of goods and services (or Pareto inefficiency) in a free

market, hence resulting in market failure. In TCE terms, transaction costs in general

impede the formation of the market (Arrow, 1969; Williamson, 1985). For example,

in a monopoly or duopoly where there is only one or two competitors in the market,

market equilibrium will not be considered Pareto efficient, and indicates the potential

for market failure ex ante or pre-contract due to a lack of competition, or imperfect

competition. With lower competitiveness in the market, there is a lack of incentives

for contractors to be innovative in order to be price competitive, and this can impinge

on performance across the whole-of-life of the asset. Results from the empirical

study in Figure 3.4 (above) also indicate a lower level of VfM at lower competition

level.

As discussed in Stage 1/Task B: Make-or-buy analysis and Stage 1/Task C:

Market analysis, the model deals with lack of competition arising out of scale or size

of project, by dividing the project or activity into bundles of activities, and mindful

of the market or supply at the same time. For example, major projects which are

bundled in a large-sized package might attract lower levels of competition. Empirical

results show a limited number of contractors that are capable of AUD 800 million

and above in the Australian major infrastructure market (Teo, Bridge, Gray, &

Rowlinson, 2012). However, if the activity cannot be broken down further due to

highly specialised technological asset specificity, and there is only one firm that can

provide the service (Level 7 activity in the procurement model), there can be

considerations to enhance its integration with the firm; for instance, by considering

the possibility of a bilateral monopoly such as a merger, or developing a relational

exchange with a preferred supplier or subcontractor (Williamson 1985; Winch 1989).

In the determination of low levels of competition, Selten (1973) shows that 5

competitors represent the dividing line between few and many, when modelled as

moves in a non-cooperative game and pertaining to a bidding scenario. In other

words, low levels of competition or low EoI – that is, 4 or less firms demonstrating


their willingness to bid for a project creates tight oligopoly conditions and associated

pricing constraints, along with ineffective competition (Beattie, Goodacre, &

Fearnley, 2003; Shepherd, 1982). As such, 4 or less EoI can be used as an early

indication of market failure ex ante arising from a lack of competition associated

with scale or size of the project.

3.3.3 Measure of competition and post-contract market failure

A study of the major infrastructure market in Australia shows that current

contracting practice could have resulted in market failure ex post (Sweeney 2009).

Market failure is not an absolute theory (Arrow, 1969), and in addition to the lack of

competition as a factor, TCE also envisages market failure occurring ex post due to

post-contractual opportunistic behaviour. For instance, asset specificity in

construction projects tends to be higher post-contract when the investment has been

made, and the costs of replacing the main-contractor, subcontractor, or supplier are

higher both financially and in terms of the implications for project progress. This is

in contrast to pre-contract when there are plenty of alternative main-contractors,

subcontractors, or suppliers in the market (Winch, 2001).

This change to a form of bilateral monopoly post-contract is known as

‘fundamental transformation’ (Williamson 1985). Opportunistic behaviour is not

uncommon in the construction industry, especially for complex projects associated

with high uncertainty, and the market can perceive the project to be lucrative in terms

of a potential for variation claims, and the opportunity to make superior profits post

contract (Sweeney 2009). This can open up opportunities for hold-up ex post. For

example, Sweeney examines the Australian construction market and observes that

contractors can use low bid tendering strategies to win contracts in an extremely

competitive market (that is, high EoI), and adopt a claim strategy resulting in high

variation claims ex post (Sweeney, 2009, p. 141).

As discussed in Stage 2: Bundling analysis and Stage 3: Exchange relationship

analysis, the procurement model deals with opportunistic behaviour in transactions

which are susceptible to changes when the environment is too complex or uncertain

to be fully specified. This is the case for a Level 4 activity, or when the market is

limited to a small number of players, such as in a Level 7 activity. When the activity

is on the critical path and susceptible to time and cost delays, main contractors or


subcontractors can hold-up clients including government in terms of seeking to

appropriate gains from variations to the works.

In other words, high levels of competition, or high EoI (over 8), can be an early

indication of market failure ex post arising from the prospect of lack of flexibility; for

example, projects that are subjected to the dynamics of change and are locked into

long-term contracts, such as an extremely complex PPP, create opportunities for

post-contractual opportunistic behaviour.

In summary, high EoI (over 8) is empirically shown (Gupta, 2002; Skitmore,

2002) to yield little production benefits in terms of lower prices and lower incentives

for design innovations and, at the same time high EoI can indicate the prospect of

market failure ex post or post-contract, due to potential negative opportunistic

behaviour as a result of lack of flexibility (Sweeney, 2009; Williamson, 1985). On

the other hand, low EoI (4 or less) are not sufficient to avoid oligopoly pricing

constraints and resultant ineffective competition (Beattie, Goodacre, & Fearnley,

2003; Selten, 1973; Shepherd, 1982) and can indicate market failure ex ante or pre-

contract, including as a result of the issue of size or high level of bundling. For these

reasons, 5-8 (inclusive) EoI is derived as an optimal level of competition, and

consistent with VfM for the purposes of this research.

3.3.4 Development of hypothesis and approach to testing

Having established and justified EoI as a valid indicator of VfM in terms of

avoiding market failure pre and post-contract in the previous Sections 3.3.2 and 3.3.3

(and, as shown in Points 2 and 3, Figure 3.5, below), the next step involves statistical

analysis of empirical data to justify the reliability of the effects of the procurement

strategy (size, bundling and exchange relationship) on EoI. This analysis informs the

development of the hypothesis to test the model (Point 1, Figure 3.5, below). The

results of the statistical analyses are given in Chapter Five.


Figure 3.5: Summary of theoretical development

On the basis of having established the validity and reliability between the key

aspects of procurement strategy on EoI and VfM, the corresponding refined

hypothesis (illustrated in Figure 3.6), is given as:

Actual competition (Box D) is expected to be in the optimum range of

competition (5 to 8 EoI) (Box C) in cases in which actual procurement (Box A)

matches the theoretical procurement (Box B – as informed by the first-order

procurement decision-making model) and outside the optimum range in cases where

there is an appreciable mismatch between actual procurement and the theoretical

procurement.

Figure 3.6: Procurement-competition/flexibility hypothesis

(Source: Teo, Bridge, Gray, & Jefferies, 2011)

As part of developing the hypothesis, data was collected on actual procurement

(Box A), and actual competition (Box D) in both road and health sectors in Australia

via a questionnaire survey of major road and health infrastructure projects. Data was

also collected on optimum competition (Box C) via a nationwide survey of civil and

Box A

Box B

Box D

Box C

Actual Procurement

Predicted Procurement First-Order Decision-

making Model (New integrated theory)

Actual Competition

Optimum Competition

Nobel Prize winning theories

First-order procurement decision-making model

Key Procurement Dimensions

Size Bundling Exchange

Procurement strategy

EoI Avoidance of market failure (Pre-contract and post-contract)

VfM (starting on path toward superior ratio of costs and benefits in whole-life terms)

Point 1

Point 2

Point 3

Transaction characteristics

Government and market

capabilities and competence

Key activities

Project schematic


building contractors. The hypothesis – in which Box A matches Box B, and at the

same time in the project concerned, Box C matches Box D – shows that theory and

practice are mutually supportive and that procurement is contributing to optimising

VfM.

More fundamentally, the greater the parity between the actual observed

competition on a project and the optimum competition (5-8 EoI) for that project –

based on the actual prevailing capacity surrounding the project vis-à-vis the

theoretical procurement mode that represents a particular and efficient configuration

of risk allocation; bundle(s) of externalised activities; and the nature of the external

exchange relationship with each externalised contract(s) for the project concerned –

the greater the success of the actual procurement mode in terms of its likelihood of

delivering superior VfM in relative terms, and a key proxy for VfM. Specifically, if

actual competition is between 5-8 EoI, then actual procurement is expected to be a

closer match to the theoretical procurement (informed by the first-order procurement

decision-making model). If actual competition is of a lower or higher number and

outside of the 5-8 EoI range, then an appreciable mismatch between actual

procurement and theoretical procurement is expected.

To test the hypothesis, and in terms of Box B, a multiple case studies approach

(Yin, 2009) was preferred, as it allows comparison between and within cases, and

enables predictions of similar or contrasting results across cases on a theoretical basis

(Baxter & Jack, 2008). A multiple case studies approach can also test theory in

various sectors and attest to reliability in various projects. If two or more cases are

shown to support the theory, then replication has been achieved.

Based on theoretical replication and the spread of EoI in the sample, an

approach was taken where sub-optimal (low and high) and optimum EoI cases were

replicated, with the expectation of different results across optimal and sub-optimal

cases. In contrast, and in terms of literal replication, the same results were expected

within each case in the two categories of optimal and sub-optimal EoIs.

In summary, based on the case study design illustrated in the 2x2 matrix in

Figure 3.7 (below), four case studies were selected from the sample of 87 projects.

This selection met Yin’s (2009, p. 54) recommendation concerning sufficiency of the

number of cases studies for in-depth analytical generalisation. The model was

expected to predict a different procurement approach to current practice at low or


high level EoI, and a similar procurement approach at optimal EoI. It is noted in

Figure 3.7 (below) that cases were selected from two sectors so as to further

strengthen the validity and reliability of results.

Figure 3.7: Multiple case studies’ replication design

(Source: Teo, Bridge, & Gray, 2013)

3.3.5 Refutability and hypothesis testing

Based on the above discussion of using multiple case studies to test the model,

the following four scenarios were generated to either support or refute the theory:

• Scenario 1: If actual competition is optimal between 5-8 EoI, and actual

procurement approach matches closely to the predicted procurement in terms

of key aspects of procurement – that is, make-or-buy, level of bundling in key

or main or major bundles of activities, and the nature of exchange relationship

– then empirical research supports theory.

• Scenario 2: If actual competition is optimal between 5-8 EoI, and actual

procurement approach is contrary or appreciably dissimilar to the predicted

procurement, then the predicted procurement can be refuted, and empirical

research informs and further develops and tests the theory.

• Scenario 3: If actual competition is sub-optimal (that is, of relatively high or

relatively low and outside 5-8 EoI), and actual procurement is similar to the

predicted procurement, then the predicted procurement proposed by the

procurement model can be refuted and empirical research informs and further

develops theory.

• Scenario 4: If actual competition is sub-optimal (that is, of relatively high or

relatively low and outside 5-8 EoI), and actual procurement is appreciably

different from the predicted procurement, then the predicted procurement is

Lite

ral r

eplic

atio

n

Theoretical replication

Sub-optimal/ High or low EoI

Expect different from model

Road case study #R2

Optimal EoI

Expect similar to model

Road case study #R1

Sub-optimal/ High or low EoI

Expect different from model

Hospital case study #H2

Optimal EoI

Expect similar to model

Hospital case study #H1


expected to generate an EoI that is in the optimal EoI of 5 to 8, then empirical

research supports theory.

The outcomes of Scenarios 1 and 4 support the hypothesis, while Scenarios 2

and 3 refute the hypothesis.

3.4 Summary

This model represents a change from the current practice of selecting

procurement mode which may begin with a dominant and visible project

performance outcome that is to be observed on the opening day of the piece of

infrastructure – the outcome can focus on either time, cost, or quality in either

absolute terms, or in terms of certainty – to an approach which begins with an

analysis of the project conditions.

The model outlined in this paper reflects this change and differs significantly

from Infrastructure Partnerships Australia’s (2008) five step process for selecting a

delivery model, and from Grimsey and Lewis' (2009) procurement framework. That

is, the model in this paper goes beyond the project’s scale or complexity and includes

(at the outset) an analysis of the capability and competencies of government and

private sectors relative to the project. And, in so doing, the model explicitly involves

the integrative application of Nobel Prize-winning and empirically tested transaction

cost theory, and production cost or benefit theory, to measure the project conditions.

From this comprehensive basis, the model then indicates a procurement approach.

When deploying the practice of procurement selection based on a key opening

day project performance outcome, VfM is most accurately measured on a nominal

scale. That is, the procurement mode either delivers, or does not deliver, on the key

project performance outcome. If it does deliver on this outcome, then the

procurement has been successful, even though it may under-deliver on other project

performance outcomes observed on the opening day, and might well fail to minimise

transactions costs. This is because no planned attempt has been made to address the

full range of these transaction costs in the procurement selection. Thus, the risk of

this procurement mode not delivering, even on the key project performance outcome,

is real.

Even when procurement selection does start with conditions related to project

characteristics and government or market capability, it is likely to stumble if it


attempts conventional risk analysis, given incomplete contracting and the long-term

nature of infrastructure contracts. Thus, the default position again becomes a focus

on a dominant and more visible project performance outcome to be observed on the

opening day of the piece of infrastructure.

The first-order procurement decision-making model in this research could be

complementary to the PSC, insofar as it can cross-validate the PSC. Moreover, given

that the transaction costs and production theory measurements are all indirect and do

not require estimates in monetary terms, the first-order procurement decision-making

model then lends itself to public dissemination – when part or all of the PSC might or

might not be made widely available. As an alternative to the PSC, however, the first-

order procurement decision-making model can save time and cost in not going into

the full extent of detail as is required in the reference project design associated with

the PSC. Also, selection of a PPP by the first-order procurement decision-making

model (and in the absence of a reference project), can send a stronger signal to the

private sector. This, in turn, could encourage more innovation and competition.

In summary, the first-order procurement decision-making model addresses the

weaknesses of MAUA and offers a potential way to procure a more reliable method

of risk allocation, based on Nobel Prize-winning economic theories. The model is not

predisposed to any particular procurement mode; rather, it identifies an efficient

procurement approach, including size, bundling, and exchange relationship

(including payment terms) by deploying TCE and RBT at the first-order project

level.

The model also addresses the two key determinants of the suitability of PPP,

that is, competition and flexibility, and can uncover projects (or bundles of activities)

within projects that can be investigated as potential PPPs. Furthermore, the model is

designed to cope with changing market conditions, and with the comparison of

government and private sector resources at the time of the project, and in any

location.

Chapter 4: Research design and methods 115

Chapter 4: Research design and methods

4.1 Introduction

This chapter justifies the use of multiple research methods to collect data, and

multiple sources of evidence and techniques to test the theoretical model developed

in Chapter 3. The chapter begins with the discussion of the research design, and links

this to the components of the hypothesis comprising actual competition, potential

competition, actual procurement, and predicted procurement. A rationale for the

research design (based on survey and case study methods) is given, followed by

details of the design and administration of the survey of major road and health

infrastructure projects, and a survey of civil and building contractors. Next, the

overall approach to, and procedures for testing the procurement-

competition/flexibility hypothesis and the first-order procurement decision-making

model (using the case study method) is detailed. Lastly, ethical matters are

considered.

4.2 Research design

4.2.1 Generally

‘Research design’ represents a structured approach to research inquiry, and

comprises a logical and systematic plan of connecting empirical data and the

hypothesis. ‘Research methods’, on the other hand, refer to the actual procedures and

techniques used to obtain the data. These are determined by the nature of the

hypothesis which indicates the type of data to be collected, and the strategies for data

analysis that will link the data to the proposition (Yin, 2009, p. 35). That is, the

research methods, including the data collection instruments, need to address the

components of the hypothesis.

The research methods were carried out in two stages. Stage 1 involved the

design and administration of a questionnaire survey of major road and health projects

to collect data on actual EoI or competition (Box D of the hypothesis), and a number

of key procurement dimensions within actual procurement (Box A of the hypothesis)

corresponding to the outcomes of the procurement model. The data and information

collected, and the statistical analysis carried out, was designed to test the reliability

116 Chapter 4: Research design and methods

of the effect of the outcomes of the procurement model and EoI. Additionally, a

questionnaire survey of civil and building contractors was designed to collect data on

potential competition (Box C of the hypothesis). Overall, Stage 1 constituted a major

portion of the research process. This was a result of the fact that a great deal of care

and consideration was taken in designing and finalising the decision-making model,

before proceeding to Stage 2.

Stage 2 involved testing the first-order procurement decision-making model by

analysing selected case studies (from the survey of major road and health projects)

through the mechanisms of the model which generate a theoretical procurement

strategy (represented by Box B). In cases where there was an observed mismatch

between actual procurement and theoretical procurement (informed by the model),

then the data from the survey of civil and building contractors was used to estimate

the theoretical EoI that would have arisen from the theoretical procurement. Table

4.1 (below) summarises the research methods employed and the corresponding data

collection instruments.

Table 4.1 Research methods and data collection instruments

Components of hypothesis Research method Data collection instrument

Potential competition Survey method Questionnaire survey of civil and building contractors and interviews with contractors

Actual competition Survey method Questionnaire survey of major road and health projects and interviews with road and health agencies

Actual procurement Survey method Questionnaire survey of major road and health projects and interviews with road and health agencies

Predicted procurement Case study method Case study questionnaire, interviews and documents

Figure 4.1 illustrates the overall design of the research, and the rationale for

employing survey and case study methods is given in Sections 4.2.2 and 4.2.3. The

research methods, steps and procedures used to test the procurement-

competition/flexibility hypothesis are fully detailed in Section 4.3 further on in this

chapter.


Figure 4.1: Research design

4.2.2 Survey method rationale

The survey method is a widely used research method in various fields and

disciplines. The method is an efficient and relatively inexpensive technique to

administer, and can have a wider population reach. The method is particularly useful

in obtaining quantitative and qualitative data, and is efficient in reaching out to the

intended respondents.

In addition to collecting useful data from a representative sample, the findings

can be statistically generalised from a sample to a population. The responses can be

organised in a standardised manner, and analysed using statistical software packages

such as SPSS or Microsoft Excel. The responses obtained are preliminarily analysed

through data reduction. This enables the data to be simplified (through coding and

clustering), relationships to be explored, and significance to be gauged (Walliman,

2011).

Theoretical development

Components of Procurement-competition /flexibility

Research methods

Data collection instruments

Data sources

Stage 1 Stage 2

Literature review

Actual competition

Actual procurement

Optimal competition

Predicted procurement

Development, testing and validation of first-order procurement decision-making model

Survey method Case study method

• Questionnaire survey of major road and health projects

• Interviews with road and health agencies

• Questionnaire survey of civil and building contractors

• Interviews with contractors

4 Case studies (Case study questionnaire and interviews and documents) • Road case study #R1 • Road case study #R2 • Health case study #H1 • Health case study #H2

Government agencies of Health and Main Roads in WA, SA, VIC, NSW and QLD

Prequalified contractors in the civil and building sector in WA, SA, VIC, NSW and QLD

analysis analysis


4.2.3 Case study method rationale

In addition to the survey method, the case study method has been an important

technique in qualitative or social science research and, especially, in “investigating a

contemporary phenomenon in-depth and within its real-life context” (Yin, 2009, p.

18). Its unique ability to “cope with a technical distinctive situation, with many more

variables of interest than data points” (Yin, 2009, p. 18) ensures that the rich depth of

information is not lost in statistical calculation and mathematical formulae.

Furthermore, the reliability of the data is ensured by triangulating and corroborating

evidences from multiple sources of information, such as survey instruments,

documentation and interviews. In addition, interviews generate greater respondent

trust, and more in-depth information. This major strength is particularly useful in

studying the complex nature of both the construction industry and major

infrastructure procurement in the public sector. More importantly, the case study

method is a comprehensive research strategy that includes the testing of the

theoretical model through data collection and analysis (Yin, 2009, p. 36).

More specifically, the case study method provides a complete framework to

test a theoretical model through analytic generalisation (Yin, 2009, p. 15). This is

consistent with the research aim of testing the first-order procurement decision-

making model in the procurement of major infrastructure. Analytic generalisation is

similar to conducting an experiment where a particular phenomenon is replicated and

generalised through a series of tests to determine causal relationship; that is, to link

data (effects) collected to the proposed theory (cause) through a process of pattern-

matching. This is in contrast to statistical generalisation where generalisation is

based on statistical frequencies.

In terms of the number of case studies, there are no fixed rules, and Yin

considers four to six case studies as an ideal number (Yin, 2009, p. 54). If the results

are as predicted, they will provide strong support for the initial set of propositions.

4.2.4 Merits of multiple methods

Multiple methods, which comprise of survey method and multiple case study

method, were the preferred methods for this research. The analysis of quantitative

data (positivism) obtained from the survey method can shed light on the relationships

between the variables, and case study method enables a deeper understanding of


complex structures which cannot be obtained from the survey method alone.

Therefore, the adoption of a combination of methodologies can achieve a

corroboration or triangulation of quantitative and qualitative results, and result in an

improved understanding of theory and phenomena, especially in the inherently

complex construction industry.

Construction management scholars encourage researchers to embrace

methodological pluralism in order to gain greater understandings of the industry

(Dainty, 2008; Runeson, 1997; Seymour, Crook, & Rooke, 1997). Hence, this

research adopts both the case study and survey approach, in order to accomplish

superior outcomes both in analytic and statistical generalisation.

4.2.5 Reliability and validity

The survey and case study methods are designed to ensure the reliability and

validity of the measurements. ‘Reliability’ refers to the consistency of the measure,

which is the extent to which the same results can be reproduced following the same

procedures; ‘validity’, on the other hand, is concerned with how well the measure

represents the true construct of interest and reduces measurement error (Remler &

Van Ryzin, 2011, pp. 106, 118). Specifically, reliability is concerned with

minimising the occurrence of random errors and bias, while validity refers to how

well the research is capable of measuring the construct. Four types of validity are

considered while designing the items in the questionnaire: face validity, content

validity, criterion-related validity, and construct validity. These are defined as

follows:

• ‘Face validity’ refers to how well a measure, comprising of a range of items,

captures what it is supposed to measure ‘on the face of it’. In other words, it is

concerned with how the measure appears.

• ‘Content validity’ refers to how well the measure captures all the important

dimensions of the construct. More specifically, the measure should reflect the

content of the intended domain of content.

• ‘Criterion-related validity’ refers to the extent or level that a measure relates to

other criteria that demonstrate its validity.

• ‘Construct validity’ refers to the extent to which the measure corresponds with

other items that are logically or theoretically related in the hypothesis.


The following sections give the details of the steps and procedures of the

survey and case study methods.

4.3 Questionnaire design and administration

4.3.1 Actual procurement and actual competition (survey of road and health projects)

a) Purpose and content

It was necessary to design a data collection instrument relating to the

procurement and competition (EoI) of major road and health projects. It was

determined that a questionnaire could be designed to obtain essential information on

one project to help test the procurement-competition/flexibility hypothesis (as shown

in Figure 3.6) in terms of establishing whether a relationship exists between EoI and

the key dimensions of procurement informed by the procurement model. The

questionnaire survey was directed at government road and health agencies.

Specifically, this questionnaire survey – the ‘Information Required Schedule’ – was

designed to obtain qualitative and quantitative information on major construction

projects in road and heath sectors that is not readily available in the public domain,

including the details of EoI. The schedule comprises three sections – namely, project

details, project scope and procurement, and tendering approach – and includes a total

of 19 questions. (Please see Appendix B for the final draft version of the

questionnaire instrument for this survey.)

Section A of the instrument requests information to identify relevant projects

that fall within the delimitations or parameters of the research. Projects could either

be a road or hospital (or healthcare-related) project with a capital value over $50

million. The construction contract, or major construction contract, needed to have

been executed in the period between July 2005 and June 2010 (inclusive), or

proposed – that is, it needed to be current, deferred or discontinued, or (at least) at

the EoI stage – during this period of time.

Section B establishes the procurement approach for each project in relation to

size or value, procurement mode and payment terms. Specifically, the questions in

this section were designed to gain information regarding the project delivery method,

nature and scope of design, construct, operations and maintenance, and (possibly)

private finance in the project. In addition, information was requested for the type of


contract being utilised in terms of the nature of payment terms, the scope and nature

of works in each contractual package, including the largest construction contract.

Section C seeks to understand the tendering procedure and policies concerning

the largest civil or building construction component of the contract, and also

estimates the level of attractiveness generated by public sector road and health

projects to the public sector, in the period between July 2005 and June 2010

(inclusive). Specifically, the names and number of contractors who expressed interest

or tendered during the expression of interest, open tender, invitation to bid, and final

bid stages, were requested. Further details were also requested concerning the

tendering procedures and outcomes of the project in terms of time, cost and quality.

b) Questionnaire design and pilot meetings

The first draft of this questionnaire was created on 18 March 2010, and the nine

page final draft was completed on 26 July 2010 after a further three drafts. Table 4.2

(below) is a summary of the design of the final questionnaire. Comments from

government respondents on the initial draft were considered, and some terminology

was simplified and amended for better industry understanding. For instance,

‘integrating theory’ was changed to ‘theory’ only, and ‘work packages’ to ‘method of

delivery’. Complicated tables were deleted and re-designed into easy to follow

questions. The on-line version of the schedule was designed and developed

concurrently, using on-line survey software ‘Keysurvey’. The on-line version of the

questionnaire was developed to be as similar to the hardcopy version as possible.

These alternative versions were designed to give respondents the option to complete

to complete the hardcopy version or password protected on-line questionnaire..

Table 4.2 Design of questionnaire survey of major road and health projects

Date Questionnaire version 18/03/2010 Draft of hardcopy version 08/06/2010 Final draft of hardcopy version 12/07/2010 Draft of on-line version 26/07/2011 Final draft of on-line version

A total of five preliminary meetings were held with government representatives

(at the directorate level) from Queensland Public Works, Queensland Main Roads,

and Queensland Project Services. These meetings were important as they helped to

harness support and commitment from the government agencies, and to resolve any

issues or queries on the survey of road and health projects. Issues were raised


regarding the confidentiality of sensitive information, and the level and quantity of

information that needed to be provided.

The purpose and objectives of the research, as well as the theoretical

underpinnings and schematic of the first-order procurement decision-making model,

were presented and explained to the government representatives to convey the

importance of the research and the significance of its contributions to infrastructure

procurement. The feedback and comments on the procurement model and

questionnaire were noted and considered to further improve the model. The details of

the pilot meetings held are summarised in Table 4.3 and the outcomes of the

meetings are summarised as follows:

• A mutual confidentiality understanding was achieved, and QUT’s ethics policy

was explained.

• Access to data to meet the research rigour was secured.

• It was agreed that the information collected would be related to three stages of a

project’s development (in line with the model’s information requirements): 1) at

the stage prior to the procurement decision; 2) at the expressions of interest stage;

and, 3) at the tender stage (initial tender stage if more than one stage).

• It was agreed to only request documents or information relating to the above

three stages during interviews with respondents on selected projects.

• The relevant agency and personnel to complete the schedule were identified.

• The issue of the scope of the research was resolved; specifically, health-care

social infrastructure (rather than educational) was chosen as the focus of the

research.

• The questions in the survey of road and health projects were streamlined to

obtain essential information, and to reduce confusion concerning terminology.

Table 4.3 Pilot meetings for design of questionnaire in survey of major road and health projects

No. Date Duration Departments Interview, designation 1 25/8/2010 1 hour Department of Public Works Executive Director 2 30/03/2010 1 hour Department of Public Works Director 3 05/07/2010 1 hour Project Services Director, Project Manager 4 24/08/2010 1 hour Department of Main Roads Deputy Chief Engineer 5 25/08/2010 1 hour Department of Public Works Executive Director


c) Population and sampling frame

The population of projects which corresponds with the scope of the research

(as described in Section 1.10) is defined as all road and hospital (or health-care

related) infrastructure construction projects in which expressions of interests were

established in the period between July 2005 and June 2010 (inclusive), and which

had a capital value over $50 million in five mainland states in Australia: New South

Wales (NSW), Victoria (VIC), Western Australia (WA), Queensland (QLD), and

South Australia (SA). The sample was obtained by using the stratified sampling

technique. That is, project information was obtained separately from each stratum or

mainland state in Australia. The aim was to reduce the effect of coverage bias, and to

ensure that projects in the targeted states were included. This was advantageous in

achieving an even geographical distribution of projects. For example, it prevented

oversampling in a smaller state with a smaller number of projects in the overall

infrastructure portfolio.

An initial sampling frame was constructed by compiling the list of capital

assets published in the budget papers by the state treasuries for each financial year

from 2006-07 to 2009-10. This period also encompassed the expression of interest

stage beginning 2005-06. A sampling frame is a subset of the population. The aim of

sampling is to attain a sample from the target population representative of the target

population (described above). Road and health construction projects greater than $50

million, and new works proposed in the period 2006-07 to 2009-10, were compiled

in the sampling frame. The information was extracted from the budget papers, as

shown in Table 4.4 (below).

Table 4.4 List of budget paper documents

State Budget Paper document New South Wales Infrastructure statement: Budget Paper No. 4 (Budget 2006-07, 2007-08,

2008-09, 2009-10) Queensland Capital statement: Budget Paper No. 3 (Budget 2006-07, 2007-08, 2008-09,

2009-10) South Australia Capital investment statement: Budget Paper No. 5 (Budget 2006-07, 2007-

08, 2008-09, 2009-10) Victoria Service delivery: Budget Paper No. 3 (Budget 2006-07, 2007-08, 2008-09,

2009-10) Western Australia Budget statements: Budget Paper No. 2 volume 2 (Budget 2006-07, 2007-

08, 2008-09, 2009-10)


d) Interviews and administration

Subsequent to finalising the design of the questionnaire and the successful pre-

testing of the on-line version, it was formally launched on 30 August 2010. The

respondents were emailed both the printable version and the hyperlink to the on-line

version. The reliability or quality of the analysis of the questionnaire results

depended significantly on the number of respondents and the sample size. Through a

series of project meetings with participating agencies, approvals from the relevant

health and road department agencies in the five mainland states at the directorate

level were obtained. The list of relevant projects that satisfied the parameters of the

research in both health and road sectors were established, and the relevant

government respondents were identified. This was subsequently followed-up with

phone calls to liaise with the person in the relevant state department who was

coordinating the completion of the questionnaires.

Finally, over a period of ten to twelve months, the data collection for this

survey was completed. During the entire data collection process, constant reminders

and liaison was the key to ensuring responses. The pool of projects submitted

represented the database for case study selection in Stage 2 of the research. The

government agencies were supportive of the research and submitted 87 projects in

total across road and health sectors.

After receiving the completed questionnaires, face-to-face meetings were

organised with each project manager or contract manager who submitted

questionnaire responses. The objective of these meetings was to clarify any

anomalies or misunderstandings indicated in, and to fill in any gaps in, the responses.

These responses were then amended, where necessary. At the same time, further

information pertaining to risk analysis and procurement selection guidelines and

policies was gathered for each project. This combination of face-to-face interviews

and questionnaire data helped to increase the comprehensiveness and reliability of

the data submitted. Moreover, this interaction with the respondents added a deeper

and richer insight into the projects from their inception to their tendering stages. The

meetings with project managers who submitted the schedules are recorded in

chronological order in Table 4.5.


Table 4.5 List of interviews following submission of questionnaire survey of major road and health projects

Date Interview times Government agency Interviewee Western Australia 2/9/2010 11:00 – 12:00 Building Management and Works Senior policy officer 8/9/2010 14:00 – 15:00 Building Management and Works Project Manager 9/9/2010 16:00 – 17:15 Main Roads, Major Projects Director 13/9/2010 09:30 – 10:30 Department of Health Project Manager

15:00 – 16:00 Department of Health Project Manager 13/9/2010 09:30 – 10:30 Building Management and Works Project Manager South Australia 8/10/2010 09:30 – 12:30 Department of Transport, Energy and

Infrastructure Manager, Contracting Services

14:00 – 17:00 Department of Transport, Energy and Infrastructure

Manager, Contracting Services

11/10/2010 09:30 – 10:30 Department of Treasury & Finance Executive Director 12/10/2010 09:30 – 12:30 Department of Health Director, Asset Services

16:00 – 17:00 Department of Health Project Manager New South Wales 14/12/2010 09:00 – 11:00 Department of Treasury & Finance Project Manager 16/12/2010 09:00 – 10:00 Department of Treasury & Finance General Manager 16/12/2010 10:00 – 11:00 Department of Treasury & Finance Project Manager 16/12/2010 11:00 – 12:00 Department of Treasury & Finance Project Manager 15/04/2011 10:30 – 11:30 Roads and Traffic Authority Project Manager

14:00 – 15:00 Roads and Traffic Authority Project Director 16/04/2011 11:00 – 11:30 Department of Treasury & Finance Director 19/04/2011 10:30 – 1130 Roads and Traffic Authority Project Manager

14:00 – 15:00 Roads and Traffic Authority Project Manager Queensland 16/09/2010 10:30 – 1130 Project Services Project Manager 17/09/2010 09:00 – 11:00 Department of Public Works Director Victoria 16/05/2011 09:30 – 10:30 Partnerships Victoria Director

11:00 – 11:30 Client Advisory Services Assistant Director 17/05/2011 13:15 – 14:00 VicRoads Director Contract Services

14:00 – 15:00 VicRoads Project Director 18/05/2011 10:00 – 10:45 Department of Health Project Director

11:00 – 11:30 Client Advisory Services, Transport Team Advisor

e) Data entry and coding procedures

The data submitted was analysed using the statistical software, Statistical

Package for the Social Sciences (SPSS). References were made to SPSS (1998) in

terms of coding and analysis. All the responses were computed into SPSS Statistics

Version 19. Reliability and consistency of the data was ensured by developing a set

of codes and guidelines to facilitate data entry; this helped to minimise any errors in

the data entry. The responses were coded into alphanumeric codes, except where

free-text responses were required (The list of coding guidelines of responses is given

in Appendix C). Consistent codes for similar responses helped in reducing entry

errors: a missing answer was coded as 999; and ‘Not applicable’ and ‘0’ responses


were coded as 777, and treated as missing data. Long open-ended responses were not

coded, but left as text for subsequent analysis.

After computing all the data, it was rechecked again to minimise entry errors.

Project values were coded into value categories to provide more meaningful data to

the analysis, and to identify patterns that might exist. Furthermore, to ensure internal

validity of responses, the responses submitted were checked for consistency, and any

anomalies and inconsistent responses were immediately clarified with respondents.

f) Response rate and representation of data

Representativeness of data is highly dependent on the level of response rate

that is necessary for generalising a sample to a population. A total of 87 projects,

worth $32 billion in total, across both road and health sectors, were submitted. In

order to establish the extent to which the sample of the submitted projects was

representative of the population of budgeted projects within the research parameters,

a review of statewide budget papers (for 2006-07, 2007-08, 2008-09, 2009-10

budgets) was carried out to derive an indicative population for projects falling within

the research parameters; this was then reconciled with projects submitted, to

determine the percentage response. Table 4.6 (below) shows the list of budget papers

that were reviewed across the five mainland Australian states.

Table 4.6 List of statewide budget papers reviewed

State Budget papers NSW Infrastructure Statement 2006-07: Budget Paper No. 4 (NSW Treasury, 2006)

Infrastructure Statement 2007-08: Budget Paper No. 4 (NSW Treasury, 2007) Infrastructure Statement 2008-09: Budget Paper No. 4 (NSW Treasury, 2008) Infrastructure Statement 2009-10: Budget Paper No. 4 (NSW Treasury, 2009)

QLD State Budget 2006-07: Capital Statement, Budget Paper No. 3 (Queensland Government, 2006) State Budget 2007-08: Capital Statement, Budget Paper No. 3 (Queensland Government, 2007) State Budget 2008-09: Capital Statement, Budget Paper No. 3 (Queensland Government, 2008) State Budget 2009-10: Capital Statement, Budget Paper No. 3 (Queensland Government, 2009)

SA 2006-07 Capital Investment Statement: Budget Paper 5 (Government of South Australia, 2006) 2007-08 Capital Investment Statement: Budget Paper 5 (Government of South Australia, 2007) 2008-09 Capital Investment Statement: Budget Paper 5 (Government of South Australia, 2008) 2009-10 Capital Investment Statement: Budget Paper 5 (Government of South Australia, 2009)

VIC Public Sector Asset Investment Program 2006–07: Budget Information Paper No. 1 (Department of Treasury and Finance Victoria, 2006)

Public Sector Asset Investment Program 2007–08: Budget Information Paper No. 1 (Department of Treasury and Finance Victoria, 2007)



WA 2006-07 Budget: Budget Paper No. 2 Volume 2 (Government of Western Australia, 2006) 2007-08 Budget: Budget Paper No. 2 Volume 2 (Government of Western Australia, 2007) 2008-09 Budget: Budget Paper No. 2 Volume 2 (Government of Western Australia, 2008) 2009-10 Budget: Budget Paper No. 2 Volume 2 (Government of Western Australia, 2009)


The review indicates a total of 146 projects budgeted in the five states, and

within the parameters of the research. Thus, the 87 projects with submitted responses

represented more than half of the 146 projects budgeted between 2005 and 2010. The

sample was analysed further in terms of the representativeness of the sample within

the research parameters, namely: sector, project location, timing (commencement

date) and capital value (size) of the projects. Of the 146 budgeted projects, 99 were

road and 47 were healthcare-related projects. Sixty-one road and 26 healthcare-

related projects across the five mainland states, each falling within the research

parameters were submitted. In summary, then, the sample represents more than 50

percent of budgeted projects in both sectors, with 62 percent in roads and 55 percent

in the health sector, as summarised in Table 4.7 (below).

Table 4.7 Response to survey of major road and health projects categorised by sector

Sector Projects budgeted in five mainland states and within research parameters (population)

Projects submitted (sample)

Response

Roads 99 61 62% Health 47 26 55% Total 146 87 60%

In terms of representativeness of project locations, Table 4.8 (below) shows the

number of budgeted projects and projects submitted (and percentage response) in

each of the five mainland states between 2005 and 2010. At least near or above 50

percent in each sector and across all states was obtained, with the exception of the

health sector in Victoria. Queensland had the highest total number of completed

questionnaires, at 33 (67 percent); this was followed by NSW, with 28 (64 percent)

projects or schedules.

Table 4.8 Response to survey of major road and health projects categorised by project location

State Road budgeted

Road submitted

Health budgeted

Health submitted

Total budgeted

Total submitted

NSW 39 22 (56%) 5 6 (100%) 44 28 (64%) QLD 32 25 (78%) 17 8 (47%) 49 33 (67%) SA 6 3 (50%) 4 3 (75%) 10 6 (60%) VIC 12 6 (50%) 6 2 (33%) 18 8 (44%) WA 10 5 (50%) 15 7 (47%) 25 12 (48%) Total 99 61 (62%) 47 26 (55%) 146 87 (60%)

In terms of timing or commencement date, Table 4.9 gives an indication of the

comparability of the timing of the budgeted 146 projects (using the year a project

first appeared in budget papers) with the submitted 87 projects (using a project’s

actual commencement date). The submitted projects capture at least 45 percent of


budgeted projects in each of the periods, as shown in Table 4.8 (above). A higher

proportion of the projects submitted belong to the 2008-10 period.

Table 4.9 Response to survey of major road and health projects categorised by project timing

Year Road budgeted

Road submitted

Health budgeted

Health submitted

Total budgeted

Total submitted

2005-07 34 17 15 5 49 22 (45%) 2007-08 25 11 16 9 41 20 (48%) 2008-09 21 16 10 6 31 25 (80%) 2009-10 19 17 6 6 25 20 (80%) Total 99 61 47 26 146 87

In relation to size or capital value of the projects, Table 4.10 (below) shows the

spread of the value of the projects across five value categories in terms of the

numbers and total value of projects within each of the five value categories, in both

road and health sectors.

Table 4.10 Response to survey of major road and health projects categorised by project size

Project size in value category

Roads budget #

Roads budget $b

Roads submit #

Roads submit $b

Health budget #

Health budget $b

Health submit #

Health submit $b

Totals budget # ($b)

Totals submit # ($b)

Approx. between $50-100milion

38 2.576 20 1.219 16 1.058 7 0.481 54 (3.634)

27 (1.700)

Between $100-250million

36 5.358 20 2.945 19 3.235 10 1.543 55 (8.593)

30 (4.488)

Between $250-500million

14 4.848 11 4.132 5 1.969 4 1.537 19 (6.817)

15 (5.669)

Between $500million and $1billion

9 6.627 6 4.007 2 1.384 1 0.722 11 (8.011)

7 (4.729)

More than $1billion

2 3.400 4 9.840 5 8.054 4 5.871 7 (11.454)

8 (15.711)

Totals # and ($billions)

99 22.809 61 22.143 47 15.700 26 10.154 146 (38.509)

87 (32.297)

Across both budgeted projects and submitted projects, and in both sectors, a

similar pattern occurs with respect to the frequency of the number of projects within

the value categories and the total value of projects. That is, a low number of higher

value projects account for an appreciably higher proportion of the value.

In terms of road projects, 74 (75 percent) of the budgeted road projects belong

to the two lower value (or most frequently occurring) categories (between $50-100

million category and $100-250 million category), which account for $7.934 billion


(35 percent) of the total value of budgeted road projects. At the same time, 11 (11

percent) of the budgeted road projects fall in the two higher value categories

(between $500 million-$1 billion category and greater than $1 billion category),

which account for $10.067 billion (44 percent) of the total value of the road projects.

Meanwhile, 40 (66 percent) of the submitted road projects belong to the two

lower value (or most frequently occurring) categories (between $50-100 million and

$100-250 million), which account for $4.164 billion (19 percent) of the total value of

the submitted road projects. At the same time, 10 (16 percent) of the submitted road

projects fall in the two higher value categories (between $500 million-$1 billion

category and greater than $1 billion category), which account for $13.847 billion (63

percent) of the total value of the submitted road projects.

A low number of higher value projects accounts for an appreciably higher

proportion of the value; however, this pattern is less pronounced when excluding the

single PPP road project. More specifically, 40 out of 60 (67 percent) of the submitted

road projects (excluding the PPP project) belong to the two lower value (or most

frequently occurring) categories (between $50-100 million and $100-250 million),

which account for $4.164 billion (24 percent) of the total value of the submitted road

projects (excluding the PPP project). At the same time, 9 (15 percent) of the

submitted roads projects (excluding the PPP project) fall in the two higher value

categories (between $500 million-$1billion and more than $1 billion), which account

for $9.047 billion (52 percent) of the total value of the submitted road projects

(excluding the PPP road project).

With respect to health projects, 35 (75 percent) of the budgeted health projects

fall in the two lower value (or most frequently occurring) categories (between $50-

100 million and $100-250 million), which account for $4.293 billion (27 percent) of

the total value of the health projects. At the same time, 7 (15 percent) of the budgeted

health projects fall in the two higher value categories (between $500 million-$1

billion and more than $1 billion), which account for $9.438 billion (60 percent) of

the total value of the health projects.

Meanwhile, 17 (65 percent) of the submitted health projects belong to the two

lower value or most frequently occurring categories (between $50-100 million and

$100-250 million), which account for $2.024 billion (20 percent) of the total value of

the submitted health projects. At the same time, 5 (19 percent) of the health projects


fall in the two higher value categories (between $500 million-$1 billion and more

than $1 billion), which account for $6.593 billion (65 percent) of the total value of

the submitted health projects.

A low number of higher value projects account for an appreciably higher

proportion of the value; however, this pattern is less pronounced when excluding the

three PPP health projects. That is, 16 out of 23 (70 percent) of the submitted health

projects (excluding the three PPPs) fall in the two lower value (or most frequently

occurring) categories (between $50-100 million and $100-250 million), which

account for $1.81 billion (22 percent) of the total value of the submitted health

projects (excluding the three PPPs). At the same time, 3 (13 percent) of the submitted

health projects (excluding the three PPPs) fall in the two higher value categories

(between $500 million-$1 billion and more than $1 billion), which account for

$4.807 billion (59 percent) of the total value of the submitted health projects

(excluding the three PPPs).

g) Summary of sample representation in survey of major road and health projects

In summary, the data obtained is representative of road and health projects in

the five mainland states, with a capital value (approximately) in excess of $50

million, and in which expressions of interest was established in the period July 2005

and June 2010. On the basis of location, timing and size, and with the exception of

health projects in Victoria, the sample appears to be representative of budgeted

projects. It is, at least, sufficient for the purpose of generating distinct patterns

relating to aspects of the procurement decision, competition, and the tendering

approach, which can be used to identify contrasting case studies for hypothesis

testing.

4.3.2 Theoretical competition (survey of civil and building contractors)

a) Purpose and content

The purpose of the questionnaire survey of civil and building contractors was

to estimate the theoretical competition (EoI) arising from the procurement strategy

suggested by the model, in case studies where there is a mismatch between actual

procurement and the theoretical procurement suggested by the model. This estimate

is based on delineating or mapping sectors of contractors with the capacity and


capability to express interest in the theoretical approach to procurement suggested by

the procurement model.

The questionnaire was targeted at top-tier main contractors capable of

delivering projects in excess of $50 million in the road and health sectors nationwide.

The questions were largely designed to map the actual prevailing capacity

surrounding the project vis-à-vis the theoretical or predicted procurement mode that

represents a particular and efficient configuration of: risk allocation; bundle(s) of

externalised activities; and the nature of the external exchange relationship with each

externalised contract(s) for the project concerned. The information allowed for the

identification of broad sectors in major infrastructure, and an understanding of the

Structure-Conduct-Performance of firms in each sector. The last section of the

questionnaire concerned the effects of tendering on bidding. The final version of the

questionnaire comprised of four sections. (A sample questionnaire for the building

contractors is given in Appendix D.)

Section A required general information regarding the participating contractor,

and information to identify the respondent (such as name, designation, email address

and contact number). Section B sought to generate data that would divide the civil

and building construction market in each state into sectors (or pools of firms), whose

pricing decisions could affect each other. Four parameters were used to sectorise a

firm’s construction activities within the focal state: 1) type and size of project

(hospitals and roads over $50 million in value); 2) two time periods (that is, pre-GFC

and mid- or post-GFC); 3) the scope of the firm’s building or civil construction

services and/or procurement modes offered [that is, construct only (CO) (lump-sum

or fixed-price contract); design and construct (DC); design, construct and maintain

(DCM); design, construct, operate and maintain (DCOM); construction management

(CM); managing contractor (MC); early contractor involvement (ECI); alliance

contracting (AC); any procurement mode that includes private finance, such as

public-private partnership (PPP)]; and 4) the geographical scope of the firm’s

building or civil construction operations within the state.

Section C analysed the structure-conduct-performance of each firm in each

sector. It generated data concerning the firm’s actual and projected capacity and

actual performance, which enabled a more accurate assessment of the potential


competitiveness in each sector in the focal state, mindful of the size of the project

and the procurement type.

Section D sought feedback on the tendering practice and procedure used by the

state government department(s) or agencies which are responsible for delivering

major public sector building and civil construction projects, including hospital (or

healthcare) and road projects (over $50 million capital value). It provided general

practical knowledge, such as the effects of tendering and other related factors on

bidding and construction capacity, and provided additional information that helped to

test the procurement-competition/flexibility hypothesis.

b) Questionnaire design and pilot meetings

The first draft of this questionnaire was created on 18 March 2010, and was

completed on 28 September 2010 for piloting. The draft version of the questionnaire

was subjected to a further nine pilot meetings with civil and building contractors and

infrastructure-related organisations before it was finalised on 11 April 2011. Table

4.11 (below) is a summary of the nine pilot meetings with contractors.

Table 4.11 Survey of civil and building contractors – Pilot meetings

Date Duration Contractor Interview, designation 20/04/2010 1 hour Seymour Whyte Chief Executive Officer 22/07/2010 1 hour Fulton Hogan General Manager – Northern Region 3/08/2010 1 hour Vision Stream (Leighton

subsidiary) General Manager – Northern QLD

17/08/2010 1 hour Wiley & Co. Director and Cost Planning Manager 26/08/2010 1 hour Baulderstone Building General Manager 8/09/2010 1 hour Laing O’Rourke SEQ State Manager (Building) 15/12/2010 1 hour Fulton Hogan General Manager – Northern Operations 20/01/2011 2 hours Baulderstone Building General Manager 25/02/2011 1 hour AI Group Associate Director – Construction and

Infrastructure

Feedback from these pilot meetings recommended separately tailoring the

questions to suit the civil and building sectors in each state to prevent ambiguity in

the questions and respondent misinterpretation. That is, a target questionnaire was

designed for each sector in each state. A firm could operate in more than one state,

specialise in more than one sector, or a respondent might hold positions in multiple

states, and this could lead to misinterpretation of the questions. The benefits of

creating multiple targeted questionnaires also include: 1) increasing the validity and

reliability of the data obtained using clearly designed targeted questions, and 2)


minimising the need for clarifying answers. As competition is analysed on a

geographical basis, this helps with clarity and consistency of the data collected.

Hence, ten versions of the questionnaire were created, including on-line

versions: one for the civil sector, and one for building sector in each of the five

targeted states. Each version had specific instructions for respondents to provide

information pertaining to either of the two infrastructure sectors and the targeted

states. All versions contained the same questions, except for the wording that varied

in terms of state and sector. Each respondent was required to provide responses

relating to their experiences in each particular state, and to their respective sector

only. Table 4.12 (below) summarises the design of the questionnaire survey of civil

and building contractors.

Table 4.12 Development of survey questionnaire for civil and building contractors

Date Questionnaire Version 18/03/2010 Contractors’ questionnaire Draft 1 08/06/2010 As above Draft 2 12/07/2010 As above Draft 3 03/08/2010 As above Draft 4 20/09/2010 Contractors’ questionnaire (Civil) Final draft (Civil) 28/09/2010 Contractors’ questionnaire

(Building and Civil) Final draft (Civil) Final draft (Building)

22/03/2011 Contractors’ questionnaire (Building and Civil – QLD)

Final version (Building – QLD) Final version (Civil – QLD)

11/04/2011 Contractors’ questionnaire (Building and Civil – NSW, QLD, SA, VIC, WA)

Issued version (Building – NSW, QLD, SA, VIC, WA) Issued version (Civil – NSW, QLD, SA, VIC, WA)

16/05/2011 Contractors’ questionnaire (Building and Civil Contractors – NSW, QLD, SA, VIC, WA)

Launch of ten on-line versions of questionnaire (Building – NSW, QLD, SA, VIC, WA; Civil – NSW, QLD, SA, VIC, WA)

The on-line version of the questionnaire was developed subsequent to the

issued version of the questionnaire using the same on-line survey software

(‘Keysurvey’) for the design of the questionnaire survey of major road and health

projects. Similarly, 10 separate versions for each sector in each state were created.

After the successful pre-testing of the civil contractor and the building contractor on-

line versions, the on-line questionnaires were formally launched on their own

website on 16 May 2011.

c) Population and sampling frame

The initial sampling frame for the survey of civil and building contractors was

obtained from the prequalification registers of contractors maintained separately by

the participating authority of each jurisdiction. As this research is concerned with the


interactions among firms, the geographical nature of competition was taken into

consideration. Different branches of the same parent firm were treated as separate

firms, as each branch could have had a different business focus and human resource

organisation based on the local market. However, transferability of resources – that

is, exchange of expertise, resources and capabilities interstate – and its effects on

competition were also considered.

The population of contractors was obtained from the prequalification website

of each mainland state in relation to each year in the period between 2005 and 2010.

This is differentiated from the current national prequalification scheme (that

commenced on 1 January 2011) and is outside the parameters of this research. The

prequalification scheme requires contractors to apply to be prequalified in the various

categories of contractors, in order to be able to bid for different types and size of

infrastructure projects. Before the national prequalification scheme, each jurisdiction

had its own prequalification register of contractors and categories of infrastructure.

This meant that two branches of the same company in different states might not

qualify to bid for the same sector or value category.

The authority in each jurisdiction is responsible for maintaining its own list of

contractors. As the lists are updated yearly, the list of contractors prequalified in the

years between 2005 and 2009 is not publicly obtainable. To achieve a reliable

database for the targeted population of contractors, the relevant authorities across the

five states were contacted to source the respective lists of contractors who

prequalified between 2005 and 2010. This was an effort to reduce sampling error and

non-coverage error, and to address data reliability. Table 4.13 shows the consolidated

number of prequalified contractors capable of construction to the value of $50

million and above, in both road and health sectors in the years between 2005 and

2010. The contractors were mainly tier one and tier two contractors. The analysis in

the next chapter treats each questionnaire as representative of each company for the

specific sector, and as a separate respondent.


Table 4.13 Breakdown of mailed questionnaires

Jurisdiction Building Contractors

Roads Contractors

Roads and Building Contractors

Total number of contractors

Total number of questionnaires sent

NSW 10 8 8 26 34 QLD 19 13 10 42 52 SA 11 11 6 28 34 Vic 12 14 8 34 42 WA 27 20 4 51 55 Total 79 66 36 181 217

The questionnaires were posted to the entire population of contractors; thus, the

survey took on the characteristics of a census. Those contractors who were

prequalified in both civil and building sectors were sent two questionnaires. A total

of 217 questionnaires were posted; this included emails containing links to the on-

line survey.

d) Improving response rates (reducing non-response bias)

Non-response is considered to be a serious problem with the survey method

(Dillman, 1991; Remler & Van Ryzin, 2011). Longitudinal behavioural studies of

response rates have shown a decrease in many countries over the past twenty years,

as indicated by statistics conducted by national census institutes in various countries

(Bethlehem, 2009, p. 217). In a research examining 175 different academic studies

covering approximately 200 000 respondents, Baruch (1999) arrived at similar

findings on the notable decline in response rates – from an average of 64 to 48

percent in the period 1975 to 1995.

As confidence in the results of the research relies significantly on response

rates, it is important to consider ways to improve these rates without compromising

research rigour. Total Design Method (TDM), pioneered by Don Dillman in 1978, is

an approach aimed at improving response rates for data collection of self-

administered mail surveys. TDM was successfully implemented (from the 1970s to

the 1990s) in various surveys, including surveys by the United States government

and various educational and private organisations (Dillman, 2008).

Don Dillman has since adapted the original TDM method into the Tailored

Design Method (Dillman, 2000), which now includes alternative communication

modes, such as web or internet surveys, telephone respondent contact, and mixed

modes of data collection. Each type of questionnaire – web-based and mail-based –

can compensate for the other’s weaknesses. The dual distribution can also increase


response rates, and is useful where the target population has access to the web in the

workplace (Bethlehem, 2009).

The key components of TDM were applied to the design and administration of

the questionnaire survey of civil and building contractors, with the aim of increasing

response rates. The applications and importance of each TDM component – namely,

notification letter, cover letter, sponsorship, personalisation, incentives, respondent

contact, questionnaire design, incentives and questionnaire design – are now

discussed in detail, and summarised in Table 4.14 (below).

With regards to the notification letter, informing the respondents of an

important upcoming survey is an important step, as it motivates the respondents to

pay more attention to the questionnaire when it arrives. Meta-analyses of empirical

research regarding survey response, indicates that pre-notification and follow-ups

tend to increase the response rates (Kaplowitz, Hadlock, & Levine, 2004). In this

survey, therefore, notification letters were mailed out one week prior to the posting

of the actual questionnaire to notify respondents that an important survey regarding

procurement would be arriving in a few days, and that the respondent’s contribution

to this survey would be greatly appreciated. (A sample of this notification letter is

given in Appendix E).

A well-designed questionnaire includes a cover letter to explain what the

survey is about, and why a response is important (Dillman, 2000). The cover letter

for the survey in this research summarised the grounds and motives for the survey,

gave a concise explanation of what was needed to complete the survey, and indicated

who was/were the most suitable person/s to respond. The title of the cover letter,

Reforming the procurement of construction and financing of Australian

infrastructure, is a topical issue in the construction industry and was, therefore, likely

to draw the respondent’s attention to the questionnaire. It is also important at this

point to clearly state the anonymity of the research; in this case, that only the

research team could access the responses; that no government agencies or funding

body would have access to any of the data obtained in the questionnaire; and that

only aggregated results containing non-identifiable data would be published or made

available to funding bodies. The timeframe of 20 working days from receipt of the

questionnaire to submission of the completed questionnaire was also indicated in the

cover letter. Contact details of the research team and the funding government agency


were also included at the bottom of the letter so that the respondents could contact

the research team for any assistance; more importantly, the inclusion of these contact

details also fosters trust with respect to the survey’s legitimacy. (Appendix F

provides a sample of the cover letter addressed to a civil and building contractor).

Empirical research has also shown that university endorsement or sponsorship

can help to improve response rates (Kaplowitz, Hadlock, & Levine, 2004). The

notification and cover letters were printed using original letterheads from

Queensland University of Technology (QUT) to signify that the research was

legitimate and originated from a respectable organisation, rather than from an

individual. In addition, given the level of detail required in the questionnaire,

personalised letters of support encouraging contractors to participate in the survey

were sourced from the respective government agencies in both the health and road

sectors in each state where possible, and mailed out with the survey packages. This

added another level of authenticity that the research had been endorsed by the agency

that was working with the contractor. The names of other funding organisations

supporting the research were also included in the questionnaire package.

However, respondents were not meant to feel coerced into responding as a

result of this high level of support. Rather it was clearly stated in the questionnaire

that participation in the survey was entirely voluntary; that respondents could

withdraw from participation at any time during the project without comment or

penalty; and that the decision to participate (or not) would not impact upon any

current or future relationship with QUT or any of the funding bodies or organisations

supporting the research.

Personalising questionnaires, rather than directing them to a general

respondent, is a crucial step in encouraging respondents to complete a survey,

especially given the level of sensitivity and information required in the questionnaire.

Information regarding each firm’s up-to-date postal address, name, email, phone

numbers, and designation of suitable participants and personal assistants, were

obtained by contacting individual firms. A total of 181 contractors were contacted to

obtain the respective respondent’s contact information, and this initial contact took

more than two months to complete. Up-to-date contact information, such as postal

address and names of respondents, were updated again prior to the second mail-out

of questionnaires. This is important because subsequent follow-up calls and


reminders continued over a period of eight months and there could have been cases

of personnel changes or structural changes in the company. Personalisation of the

notification and cover letters was carried out using the mail merge function in

Microsoft Word.

Monetary incentive is not considered appropriate for an organisational survey.

Instead, the firm was encouraged to complete the survey as an incentive because it

provided a significant opportunity to give feedback to governments on statewide

issues of procurement and tendering policy and practice. The other incentive was the

option of being sent a copy of the summary of the analysis of aggregated and non-

identifiable data from the questionnaire survey of civil and building contractors. In

this way, respondents could determine their own interpretations and draw

independent conclusions – beyond the conclusions of any QUT report.

With a targeted population, it is possible to reduce the problem of non-contact

bias with a series of follow-up calls and emails. Speaking directly to the respondent

or the personal assistant helps to gain their trust and co-operation. Sometimes, there

is the need to get past a gatekeeper in an organisation (Dillman, 2000, p. 327), for

example, a receptionist and/or personal assistant. Follow-up with a telephone call is

an excellent way of attempting to progress beyond gatekeepers, who are the first

people to access the mail.

Respondents were contacted within the first two weeks of mailing out the

questionnaires to follow-up and ascertain that the intended respondent had received

the questionnaire package, and to speak with that respondent to obtain an

approximate submission date. It was also important to confirm whether the intended

respondent would be completing the questionnaire, because the questionnaire could

have been passed on to another colleague within the company to be completed. It

might even have been overlooked due to personnel changes.

The chances of losing contact can be high if there is no follow-up procedure in

place. Hence, a schedule of reminder emails and follow-up calls to respondents was

established before the submission due date. As the timeliness and timing (for

example, the need to contact each respondent or personal assistant at times that

suited them) of these follow-ups was crucial, a diary of all contact and notes on

follow-up progress for each respondent was kept. These included a list of the

contractors who had submitted, confirmed submitting, not confirmed submitting, and


not submitting the questionnaire was kept and updated. This was used to help keep

track of which contractor to contact and follow-up.

Follow-ups should be carried out using a friendly and positive tone to

encourage respondents to contribute to the research; and be flexible in extending the

deadlines for submission. After two rounds of reminder emails and follow-up emails,

a last round of final call emails and follow-up emails was sent as the absolute cut-off

for respondents who had not confirmed or submitted the questionnaire. The entire

exercise was completed in approximately eight months.

Questions were kept short and were unbiased. The clarity of some questions

was improved by designing them using a tabular format. Web and mail versions were

designed to be as similar as possible, so as to reduce ambiguity and to increase the

ease of completing the on-line version. The online questionnaire had a progress bar

that allowed respondents to know the percentage of questions remaining. It also

included a ‘save’ facility that allowed them the option of saving their responses and

completing the questionnaire at another time. There were also no compulsory

questions, and respondents had the flexibility of submitting the questionnaire

whenever it was completed. Table 4.14 (below) summarises the TDM approaches

that were applied.

Table 4.14 Summary of TDM approaches applied in the survey of civil and building contractors

TDM approach Application

Notification letter Pre-notice, personalisation, importance of survey Cover letter Personalisation, cut-off date, statement of confidentiality or anonymity,

information about sponsorship Contact respondent Get past gatekeepers, gain co-operation of respondents, conduct follow-ups,

email reminders, obtain submission date, resend questionnaires via email Anonymity QUT ethics policy Incentives Summary of results available to respondents; opportunity to provide

feedback to government on current tendering and procurement practice Mixed mode strategy Mail survey, on-line survey, telephone follow-ups

e) Administration and data collection

Commencing 4 May 2011, the final versions of the 217 questionnaires were

mailed out in batches state by state, to 181 contractors, capable of delivering projects

in excess of $50 million across five mainland states in Australia. The following

summarises the steps in the administration of the survey for each state:

• Step 1: A notification letter addressed to the intended respondent was posted to

all respondents.


• Step 2: A week after sending the notification letter, the initial questionnaire

package was then mailed to all respondents with instructions on how to access

the on-line questionnaire. It consisted of the cover letter, letter of support from

the relevant agency and the civil or building contractor questionnaire, or both.

A follow-up call was then made a few days later to make contact, and to

ascertain if the respondent had received the questionnaire package.

• Step 3: A few days before the due date, a reminder questionnaire package was

emailed to all non-respondents. Shortly after, this was followed-up with calls to

determine if they would be responding to the questionnaire, and/or if it was

necessary to extend the due date.

• Step 4: A second round of reminder questionnaire packages and follow-up calls

was conducted a few days before the new deadline.

• Step 5: A final round of survey package were emailed to the remaining non-

respondents for their final chance to participate in the survey, and followed-up

with calls.

After eight months of consistent follow-up calls and email reminders to respondents,

the on-line questionnaire was finally closed towards the end of February 2012.

Reasons for eligible contractors not completing the questionnaire included: 1) their

being too busy with tendering a number of jobs in the peak season, and 2) it being

part of company policy not to disclose sensitive information. Table 4.15 (below) is a

summary of the timeline for the administration of the survey.

Table 4.15 Administration of questionnaire in survey of civil and building contractors

State Notification letter

Questionnaire mail out

Follow-up email

Reminder email 1

Phone call follow up

Reminder email 2

Phone call follow up

Final call email

Final call follow-up call

SA 4/05/2011 11/05/2011 25/05/2011 12/08/2011 12/08/2011 24/08/2011 31/08/2011 31/10/2011 1/11/2011

VIC 4/05/2011 11/05/2011 25/05/2011 24/08/2011 26/08/2011 1/09/2011 8/09/2011 31/10/2011 1/11/2011

NSW 6/06/2011 14/06/2011 07/07/2011 25/07/2011 27/07/2011 2/08/2011 9/08/2011 26/10/2011 27/10/2011

QLD 14/06/2011 29/06/2011 21/07/2011 1/08/2011 03/08/2011 26/08/2011 26/10/2011 26/10/2011 27/10/2011

WA 29/06/2011 04/07/2011 27/07/2011 5/08/2011 09/08/2011 26/08/2011 25/10/2011 25/10/2011 26/10/2011

f) Non-response bias analysis

The fact that the survey of civil and building contractors was targeted at the

senior management level, such as the CEO and managing director, could have

contributed to a high non-response rate. However, this effect was alleviated with a

cover letter which explained that the respondent could forward the questionnaire to

other appropriate personnel for their response. Feedback from respondents indicated


that the main reasons for non-response were ineligibility factors; this was despite

prior efforts to ensure the eligibility of contractors.

Some non-responding contractors clarified that they did not feel that the firm

was eligible to participate in the survey as it did not apply to their operations, even

though they were pre-qualified. Other feedback indicated that they had not yet

participated in a higher value project, even though they were pre-qualified to do so;

hence, they were unable to respond to the questionnaire. Some of the firms were too

small to tender for projects greater than $50 million. Another reason was that some

of the firms had gone into involuntary liquidation, or had ceased operations in that

particular state. These firms would not be considered eligible contractors and were

not part of the target population. As a result of these clarifications, therefore, the

sample of contractors reduced from 181 to 163; this helped to reduce the effects of

non-response bias. Table 4.16 is a summary of the number of eligible contractors in

each state.

Table 4.16 Summary of eligible contractors in each state

Contractors mailed out Non-eligibility Eligible contractors Jurisdiction Building Civil Building and Civil Total Less Total NSW 10 8 8 26 1 Building 25 QLD 19 13 10 42 2 Building 40 SA 11 11 6 28 1 Civil; 3 Building 24 Vic 12 14 8 34 1 civil 33 WA 27 20 4 51 4 civil; 6 building 41

Total 79 66 36 181 Total 163

g) Response rates and representation of data

Overall, a total of 60 completed questionnaires comprising 26 civil

questionnaires and 34 building questionnaires, were received out of a possible 199

distributed questionnaires – or from a possible 163 eligible contractors – after an

eight month period. Table 4.17 shows that the response rates were 28 percent for the

civil sector and 32 percent for the building sector; these rates are comparable and not

biased towards either sector. Response rates were calculated as a percentage of

eligible contractors or questionnaires, as recommended by Bethlehem (2009). The

overall response rate was 30 percent.

Responserate(%) = ��!��

��"��!��#�� x 100%


Table 4.17 Response to survey of civil and building contractors by sector

Sent Received % Response Civil 94 26 28 Building 105 34 32 Total 199 60 30

All responses were authored by senior executives (ranging from Managing

Director, Chief Executive Officers (CEO), Director, State Manager, Construction

Manager, Business Development Manager, Contracts Manager), as illustrated in

Figure 4.2 (below).

Figure 4.2: Respondent profile in survey of civil and building contractors

In his research involving 200 000 respondents, Baruch (1999) suggests that a

lower response rate is more acceptable for a survey targeted at populations of

Managing Directors or CEO than for one directed at a general sample of individuals.

Similarly, Henderson (1990) argues that a 20 to 30 percent response rate is typical

for a mail-out survey targeted at CEO-level respondents (cite in Baruch, 1999), and

that there is no consensus on an acceptable and reasonable response rate in academic

studies (Baruch, 1999, p. 422). Similarly, Bethlehem (2009) notes that an acceptable

response rate is largely dependent on a number of factors, such as, topic of survey,

target population, time period, location, and organisation of the fieldwork

(Bethlehem, 2009). Furthermore, in construction management research, a 30 percent

response rate is considered to be an acceptable and representative rate. Akintoye and


MacLeod (1997) assert that a 30 to 40 percent response rate is typical of a

construction industry and cannot be regarded as biased; this is in reference to Moser

and Kalton’s (1971) assertion that the results would be biased if the return rate of a

postal survey was below 30 percent. Overall, therefore, the aim is to achieve (at

least) a minimum response rate of 30 percent.

Considering that the respondents were mainly at a senior executive level, and

given the sensitivity of the information required, a response of 60 returned

questionnaires (30 percent response rate) is considered to be a reasonable outcome in

comparison to other (similar) studies targeting directorate level respondents in

Australia. For instance, Roads Australia obtained 39 responses from national

contractors in 2012 and, in another survey in Queensland, 44 responses (23 percent

response rate) were obtained from owners, property developers, consultants and

contractors (Lyons & Skitmore, 2004). Table 4.18 (below) analyses the

representativeness of the received data, with a detailed breakdown of the response

rates in each state.

Table 4.18 Response rate for surveys of civil and building contractors

State/ Sector

Number of questionnaires Number of contractors Sent Submitted Not confirmed Not submitting Total Sent Received

NSW Building 17 10 59% 2 12% 5 29% 9 8 8 2 Civil 16 7 44% 2 13% 7 44% 8 5 Total 33 17 52% 4 12% 12 36% 100% 25 15 60% QLD Building 27 11 41% 4 15% 12 44% 17 1

0 8 3

Civil 23 7 30% 4 17% 12 52% 13 4 Total 50 18 36% 8 16% 24 48% 100% 40 15 38% SA Building 14 5 36% 2 14% 7 50% 8 7 4 1 Civil 16 2 13% 7 44% 7 44% 9 1 Total 30 7 23% 9 30% 14 47% 100% 24 6 25% VIC Building 22 4 18% 12 55% 6 27% 14 8 3 1 Civil 19 2 11% 12 63% 5 26% 11 1 Total 41 6 15% 24 59% 11 27% 100% 33 5 15% WA Building 25 4 16% 10 40% 11 44% 21 4 2 2 Civil 20 8 40% 4 20% 8 40% 16 6 Total 45 12 27% 14 31% 19 42% 100% 41 10 24% Grand Total

199 60 30% 59 36% 80 49% 164 51 31%

In terms of the number of questionnaires returned, NSW had the highest

response rate at 52%, followed by QLD 36% and WA 30%; VIC and SA had the

lowest response rates of 23% and 15% respectively. NSW had the most


representative response with more than half the eligible contractors in the NSW

market responding to the questionnaire.

h) Data entry and coding procedures

The procedure for data entry, coding, and checking of responses to the

questionnaire of civil and building contractor, using SPSS software was similar to the

procedure outlined earlier in this chapter (Section 4.3.1e), and Appendix G supplies

the coding guidelines for the questionnaire survey of civil and building contractors.

In ensuring internal consistency of the responses, corroborating questions were

designed. For instance, three direct questions about turnover or performance of the

firm, and indirect questions on the number of employees in the firm, were asked.

These questions helped to add to the evaluation of the performance or size of the

firm. Any missing responses were also clarified with respondents over the phone or

through emails to see if they had inadvertently missed the question, or to seek the

reason for their non-response. This helped to reduce response and non-response

errors, and to improve the overall reliability of the data obtained.

Having discussed the survey method, the following section describes the

approach to the conduct of the multiple case study research to test and validate the

first-order procurement decision-making model.

4.4 Components of case study design (theoretical procurement using the procurement model)

The five main components of case study design, as developed by Yin (2009, p.

27) – research question, research proposition, unit of analysis, logic linking the data

to the proposition, and criteria for interpreting findings – are explained in this

section. Case study protocol is then also addressed.

4.4.1 Research question

The research question is formulated in the research aim (Section 1.6) and the

research objectives (Section 1.7).

4.4.2 Proposition

The theoretical proposition sets the direction for the type of data to be collected

and the strategies for analysing the data linking to the proposition (Yin, 2009, p. 35),

and is vital in developing the proposition at the design stage of the research. The


proposition in this research is synonymous with the procurement-

competition/flexibility hypothesis developed in Section 3.3.4.

4.4.3 Unit of analysis

The unit of analysis for both TCE and RBT is the transaction. The ‘transaction’

is equivalent to the make-or-buy of an activity in a project. A ‘project’ can be viewed

as a coordinated series or bundles of similar transactions, activities, or resources that

are not separable. In the first-order decision-making model, a project is analysed at

the activity or first-order level, and the make-or-buy analysis employed in the case

study method is fundamentally at the activity level, and that is ‘transaction’ in TCE

or RBT terms.

As discussed in Section 3.2.3c, the model seeks to identify an activity by

looking for specialisation in the market, through the process of answering the vertical

framework questions in make-or-buy. More specifically, the answers will indicate

that the user needs to combine activities when the same answers are found across two

or more activities, or to split activities when one part of the activity generates

different answers to another part of the activity.

4.4.4 Logic linking case study data to hypothesis in the approach to selecting and testing case studies

In order to test the hypothesis so that the results were generalisable beyond the

cases studied, a multiple case study approach was selected in pursuance of analytical

generalisation, as discussed in Section 3.3.4. This is in contrast to the high response

rates and statistical generalisation required in the survey method. Analytic

generalisation is employed in such a way that the developed theory acts as the basis

for comparison with the empirical data obtained from case studies (Yin, 2009). If two

or more cases are shown to support the theoretical proposition, it provides

affirmation for the theory, and replication has been achieved (Yin, 2009, p. 38). This

replication logic provides the basis for the design of the multiple case study

approach.

Each case study can be considered as conducting an experiment, and literal

replication is analogous to conducting multiple experiments where similar results are

replicated. Theoretical replication, on the other hand, is the replication of contrasting

results. Literal replication refers to the selection of cases with similar predicted

results, whereas theoretical replication produces contrasting results based on


anticipatable reasons or conditions from theory. Thus, based on theoretical

replication, the model is tested at the widest extremes on the dependent variable:

optimal competition, which supports the hypothesis; and sub-optimal competition,

which does not support the hypothesis. This testing increases the opportunity to

develop and enhance the validity of this model.

Thus, the hypothesis and theoretical replication suggests two types of case

studies, as discussed in Section 3.3.4: one study with optimal EoI (between 5 to 8

EoI), and the other with suboptimal EoI (either high or low EoI outside of 5-8). In

addition, based on literal replication, similar predicted results were duplicated; that

is, optimal and sub-optimal EoI were repeated, with one study for the road sector and

one study for the health sector. These two very different sectors were chosen so as to

demonstrate the flexibility and transferability of the procurement model.

The case study design is illustrated in Figure 4.3 (below), which summarises

analytical generalisation in a 2 x 2 matrix. Further details of the selection of case

studies are discussed further in Section 4.4.8a of this chapter.

Figure 4.3: Case study design

4.4.5 Criteria for judging quality of research design

To ensure the consistency and quality of the data collected from each case

study, the tests of validity of case study design proposed by Yin (2009) – namely,

construct validity, internal validity, external validity, and reliability have been

addressed in the design of the case study questionnaire instrument or protocol.

‘Construct validity’ is concerned with developing operational measures related

to the theoretical concepts being researched, and aims to reduce subjectivity in case

study research. In this research, multiple sources of evidence were deployed to

improve the construct validity of findings, such as survey instruments, interviews,

and documents.

Lite

ral r

eplic

atio

n

Theoretical replication

Sub-optimal/Low EoI

2 EoI

Road case study #R2

Optimal EoI

8 EoI

Road case study #R1

Sub-optimal/High EoI

15 EoI

Health case study #H2

Optimal EoI

5 EoI

Health case study #H1


‘Internal validity’ refers to the extent to which the theory about cause and

effect is supported by the study (Walliman, 2011). This is a major issue in

explanatory case study research, where the researcher can risk making incorrect

inferences, and incorrectly concluding causal relationship without the knowledge of

other causal factors. To improve internal validity, a combination of pattern-matching

was employed in the analysis of case study data. Pattern-matching is one of the most

suitable technique because internal validity is enhanced when the observed pattern

matches with the expected pattern (Trochim, 1989). For example, ‘internal validity’

is enhanced if the case studies which do not support the actual procurement

approach, were based on rival-theory, such as MAUA-based approach.

‘External validity’ refers to the extent to which findings can be generalised to

populations (Walliman, 2011). This test is to establish whether the empirical patterns

are generalisable beyond the case studies. Multiple case studies, as opposed to only a

single case, were designed to enhance external validity. The model was tested in two

different sectors and markets, in the civil and building construction sectors, and at

different competitive levels.

‘Reliability’ of the case study design refers to the minimisation of errors and

bias in the study, such that another researcher following the procedures and protocols

is able to repeat the same study and achieve similar results. As recommended by Yin

(2009), in order to enhance the reliability of the case study, a formal case study

protocol needs to be developed, and is essential for multiple case study research to

ensure consistency of data collected.

The next section describes the entire process of the design of the case study

protocol and the questionnaire instrument.

4.4.6 Design of case study protocol and questionnaire instrument

a) Unstructured and exploratory interviews

On the basis of having established the validity and reliability of the effects of

key aspects of procurement strategy on EoI, and having developed the procurement-

competition/flexibility hypothesis, the approach to testing the model using

competition as a dependent variable or proxy for VfM was determined in

consultation with an Associate Supervisor and an Associate Professor with expertise

in procurement. Their comments have partly contributed to the idea of linking VfM


with avoiding market failure in terms of high and low EoI, and this forms an

important aspect of testing the model. The sample of major road and health projects

was then checked for external validity, and the sample obtained from the

questionnaire survey attracted EoIs ranging from low to high EoI levels. The range

of the sample’s EoI indicated the viability of using low, high and optimal EoI, as a

method of testing the model.

In the next phase of the exploratory process, meetings were organised in the

Department of Transport and Main Roads and Department of Public Health in

Queensland. The aim was to ascertain the extent of vertical integration within

government, by estimating the activity levels in relation to D, C, O and M activities.

The discussions contributed significantly to the articulation of the activity levels

relating to the differential capabilities of government and market in the context of

road and health infrastructure procurement. The interviews also involved exploring

the capability and capacity of government in broad terms based on RBT variables –

value, rarity and imitability; and TCE variables – asset specificity, uncertainty and

frequency. An initial version of the protocol and questionnaire instrument was used

in these interviews. This helped to ensure the external validity and construct validity

required in the development of the case study questionnaire instrument, and depicted

a much clearer view of the market structure in both road and health sectors.

In terms of the road sector, the former Deputy Chief Engineer, and former

Director of Research and Development (Contracts, Technical Research and

Capability) were interviewed. In the health sector, the Acting Assistant Director,

Health, Health Law and Order Portfolio, Project Services, Department of Public

Works, was interviewed. Table 4.19 (below) is a summary of the exploratory

interviews conducted.

Table 4.19 Unstructured and exploratory interview register No Date Duration Time Interviewee Aim of interview 1 03/04/12 1 hour 0900-

1000hr Associate Professor Method of testing the model using

competition as a proxy for VfM 2 17/04/12 1 hour 1440-

1540hr Associate Supervisor Method of testing the model using


1300hr Associate Supervisor Method of testing the model using


1400hr Former Deputy Chief Engineer, and former Director of Research and Development in Department of Main Roads, QLD

Exploratory meetings to explore the extent of vertical integration in government in the road sector

5 06/07/12 1 hour 1500-1600hr

Acting Assistant Director of Department of Public Works, QLD

Exploratory meetings to explore the extent of vertical integration in government in the health sector


The interviews gave greater insight into the government’s internal capability

and capacity, such as its human resources, plant and equipment. The discussion

around capabilities within government revealed the nature and size of activities that

are normally internalised and externalised in state governments and agencies. These

internal capabilities are state and sector-specific and may vary across different states

and sectors. For example, in Queensland, the Department of Transport and Main

Roads has a greater in-house capability than the Department of Public Works

(Health) across design and construction activities. Despite having a certain level of

capability, both departments lack the internal capacity to carry out design and

construction activities. For both road and health sectors, government has internalised

some aspects of operations and maintenance activities, and has established

government agencies, such as RoadTek (road) and Q-Build (health). The interviews

indicate that the research is pointed in the right direction in terms of the predictive

strengths of the RBT and TCE theories in minimising costs and maximising benefits.

Finally, potential specialists in both road and health sectors, with experiences

not limited to any particular state, were recommended to assist in the design of the

scope of the analysis in Stage 1/Task A: Activity analysis, and in the design of the

questionnaire instrument, which measures RBT and TCE variables in Stage 1/Task

B: Make-or-buy analysis. The experts were interviewed using an improved version

of the questionnaire, in a semi-structured format; this is further explained in the next

section.

b) Semi-structured interviews

In these semi-structured interviews, expert knowledge was accessed to ensure

construct validity, internal validity, external validity, and reliability in the design of

the protocol and questionnaire instrument. The protocol serves as a working template

for documenting the results of each task in the model, and for maintaining internal

consistency of the data collected. From these exploratory interviews, the protocol and

questionnaire were further modified, and the next version included improvements to

the taxonomy of explanations of the activity levels, and to the intelligibility of the

questions relating to make-or-buy activity. The semi-structured interviews were

conducted on the basis of two hypothetical projects – the construction of a typical

new road, and a typical new hospital facility – to assist the design of the


questionnaire instrument. (This initial version of the questionnaire is included in

Appendix H.)

Table 4.20 (below) is a summary of the suite of meetings which spanned a

period of four to five months. A total of approximately 55 meeting hours were spent

on designing the protocol and questionnaire, and on refining the approach in the

model.

Table 4.20 Case study questionnaire design and semi-structured interview register

No Date Duration Time Interviewee Aim of interview 1 05/09/12 4 hours 1300-1700hr Former contract superintendent,

Department of Public Works, QLD Initial development of activity analysis for health projects

2 12/09/12 4 hour 0845-0945hr Former contract superintendent, Department of Public Works, QLD

Initial development of activity analysis for health projects

3 05/10/12 2 hours 1530-1730hr Former Deputy Chief Engineer of Research and Development in Department of Transport and Main Roads, QLD

Presenting preliminary model

12 11/10/12 3 hours 0800-1100hr Former Principal Project Manager, Department of Housing and Public Works, QLD

Presenting preliminary model

4 23/10/12 2 hours 1430-1630hr Principal Advisor (Engineering Program) in Department of Transport and Main Roads, QLD

Activity analysis for road


Make-or-buy analysis for road

6 08/11/12 5-6 hours 1200-1700hr Principal Advisor (Engineering Program) in Department of Transport and Main Roads, QLD







Make-or-buy analysis road (final)

10 17/12/12 3 hours 0900-1200hr Former contract superintendent, Department of Public Works, QLD

Activity analysis for health


Make-or-buy analysis for health


Make-or-buy analysis health (final)

Meetings were first organised with the former Principal Advisor (Engineering

Program) in the Department of Transport and Main Roads, for advice on the road

sector, and then with the former Contract Superintendent, Department of Housing

and Public Works, for advice regarding the health sector. A great deal of time was

spent, and a great deal of effort was made prior to the actual administration of the

case studies to ensure that the protocol and questionnaire instrument were designed

without ambiguity.


Firstly, typical key activities across D, C, O and M of road and health projects

were identified, and the corresponding activity levels assigned. This step helped to

explicitly define ‘key activity’, and to delimit the scope of activity analysis. It

became clear from the discussion that the boundaries for downstream activities of D,

C, O and M refer to anything physical that forms part of the new built asset or

facility, and which can be moved or operated in a controlled way, but not by the core

users; that is, not by cars, doctors, or nurses. For the road sector, services such as

emergency response services are not included; for hospitals, clinical services such as

those provided by doctors and nurses, and administrative services such as payroll and

human resources are not included. The discussion also clarified the issue of

internalisation and externalisation when exceptions arise. Through this process of

identifying typical activities, there was a greater understanding of the different types

of maintenance activities undertaken in road and health projects; these are discussed

further in Chapter 5.

The experts were then asked to list the most relevant non-redeployable asset or

investment made by the firm (asset specificity), the most potential source of variation

in time or delay (potential hold-up), and any specific investment on the value

variable from the government perspective. This specification addresses construct

validity and ensures that appropriate dimensions are used to measure TCE and RBT

variables associated with each key activity. The feedback was incorporated in the

design of the questions in Version 3 of the questionnaire instrument. The experts

subsequently assisted in scoring Version 3 questions, and attempted to match these

with the levels assigned, using the technique of pattern-matching. This process

resulted in fine tuning the calibration of the scale and further polished the questions;

this, in turn, enhanced the construct validity of the operational measures employed in

the questionnaire.

Through the above process, the strengths of the activity levels (based on the

integration of TCE and RBT) in explaining the reasons for internalisation and

externalisation were greatly improved (as discussed in Section 3.2.3a), and

contributed significantly to the development of Levels 3 and 5 in explaining

differential competencies between government and market. This led to the

development of capacity as a measure of value variable, and to the closer relationship

between value and frequency variables (as discussed in Section 3.2.3b). The


procedural steps of the model and the design of the questionnaire instrument were

finally formalised. In total, four versions of the protocol and questionnaire for road

projects, and two versions for health projects were developed before the design

process was complete after a period of four to five months. These iterations are listed

in Table 4.21 (below).

Table 4.21 Case study protocol and questionnaire design

No Case study approach design 1 Developing methods for testing the model 2 Case study Design 1 3 Case study Design 2 4 Designing protocol approach 5 Case study protocol Version 1 (Initial version) 6 Case study protocol Version 2 7 Case study protocol Version 3 8 Case study protocol Version 4 9 Case study protocol Version 5 10 Case study protocol Version 6 (Final version)

The first part of the protocol gives an overview of the case study research

project and includes: the project title, research objectives, and statements of

confidentiality and voluntary participation, as required by ethics. The second section

describes and explains the steps of applying each task or analysis in the procurement

model, including the main aim of each task and the parameters for implementing the

procedures. The framework and the plain language explanation for the seven levels

of internalisation or externalisation are also included in the protocol. Finally, the

standard questions relating to the RBT and TCE variables are given. (The template is

given in Appendix I.)

The design of the questions for each variable is further discussed in detail in

the next section.

c) Design of questionnaire instrument for TCE and RBT variables

The protocol presents a template for carrying out and recording the results of

the tasks in the model. The bulk of the protocol comprises of the case study

questionnaire. The main purpose of the questions is to create constructs that can

capture the dimensions of the variables underlying RBT and TCE, thus

operationalising the variables. Based on the critical realist ontological perspective

discussed in Sections 1.9.2 and 1.9.3, and in line with Godfrey and Hill’s (1995)

recommendations, this research combines the use of observable measures to explain

the unobservable phenomena. Even though the key variables of opportunism,


competitive advantage, tacit knowledge and skills, uncertainty, asset specificity,

value, rarity and imitability are intangible or unobservable in positivist terms,

Godfrey and Hill (1995) encourage researchers to identify constructs that are

observable to measure the extent of the unobservability of the variables.

The questions for RBT variables are concerned with competitive advantage,

and are measured in terms of value (capacity), rarity and non-imitability. In regards

to TCE, the questions for asset specificity and uncertainty variables are designed to

determine the potential for hold-up ex post, while frequency questions aim to

determine the economies of internalising or externalising the activity. The scores of

the questions generate an empirical pattern which is then matched with the pattern in

the framework, and assigned a corresponding activity level. Two questions were

designed for each variable, with the exception of the frequency and value variables,

in order to corroborate the results.

The next section discusses the design of the questions for RBT, followed by

TCE variables, which follow the order of the variables in the framework and the

questionnaire. Reviews of empirical research in RBT criticises RBT scholars for

using data such as industry measures or performance data, which are theoretically

disconnected from theory, and for a lack of theoretical justification of the data for

hard-to-measure or intangible resources (Armstrong & Shimizu, 2007; Molloy,

Chadwick, Ployhart, & Golden, 2011; Newbert, 2007). The following aims to

address construct validity by explaining the link between the measure and RBT

variables, including TCE variables, and to clarify the construct of the measurement.

1. Value or capacity

Having developed capacity as an alternative means of explaining value – that

is, viewing capacity as the organisation of valuable resources – and having

strengthened the relationship between value and frequency variables, this construct

helps to reduce the charge of tautology on value, and incorporates a dimension of

organisation or process to create the competitive advantage. Based on the discussion

in Section 3.2.3b, ‘capacity’ refers to the state of capacity of government at the time

of procurement decision, and to the extent of increasing the capacity and internal

costs (that is, value) in order to carry out the activity either internally or externally at

that time. Based on interviews from field experts, ‘capacity’ can be defined in terms

of the number of staff employed by government. More specifically, the question


reflects the constraints on the existing number of full-time staff (in terms of budget

and/or numbers) measured using a 7-point scale, ranging from ‘beyond capacity’ to

‘within capacity’. The question is worded differently for internalised and externalised

activity, as follows:

Internalised activity

1. Capacity a. In terms of staff employed by state department (including within a sub-

agency/subsidiary) to work full-time on the activity in the case study, how much would this have been resisted by capacity constraints (would have exceeded relevant staff salary scales/budget and/or overstretched supervisory staff)?

Respondents requested to circle: Significantly beyond capacity 1 2 3 4 5 6 7 Comfortably within capacity

Externalised activity

1. Capacity a. If state department had employed additional/new staff (including within a sub-

agency/subsidiary) to work full-time hours on this activity in the case study, then how much would this have been resisted by capacity constraints (would have exceeded relevant staff salary scales/budget and/or would have overstretched supervisory staff)?

Respondents requested to circle: Significantly beyond capacity 1 2 3 4 5 6 7 Comfortably within capacity

2. Rarity

In terms of rarity, Barney (2001a) disagrees with Priem and Butler’s (2001)

critique, given that rarity has been fully dimensionalised in his 1991 paper, and is

adequate in generating testable assertions. Barney (1995, p. 52) dimensionalises the

rarity of a resource in terms of the average level of the resource possessed by

competing firms, and most empirical research defines the variable based on this

conceptualisation (Newbert, 2007). Thus, rarity can be defined in terms of: 1. The

extent or number of market firms that possess the resources required, with tacit

knowledge and skills being the most valuable resources across D, C, O, M activities;

and 2. The level of supply of competing market firms capable of supplying the

activities. Specifically, the questions for both internalised and externalised activities

are as follows:


2. Rarity a. How much was the knowledge and skills required in this activity in the case

study possessed by all top-tier specialist local firms capable of delivering the activity?

Respondents requested to circle: possessed by all firms 1 2 3 4 5 6 7 possessed by few firms

b. How much was there a sufficient supply of firms capable of delivering the activity to the case study?

Respondents requested to circle: plentiful supply of firms 1 2 3 4 5 6 7 scarce supply of firms

3. Costly-to-imitate

Again, Barney (2001a) refutes the critique of the non-imitability variable,

arguing that it is not eventually contested by Priem and Butler (2001), and that the

variable is well parameterised. Barney (1991, p. 110) states that physical technology

that can be purchased, such as complex machineries or tools, cannot be a source of

competitive advantage and is, therefore, excluded. It follows that the term ‘non-

imitable resources’ refers to the tacit skills and knowledge or proprietary information

of new technologies.

Several empirical studies have employed a variety of tangible proxies to

measure the non-imitability of tacit skills and knowledge. For example, Henderson

and Cockburn (1994) measure the level of productive research effort within

pharmaceutical firms. Newbert (2007) lists a number of measures for the non-

imitable variable based on an extensive review of RBT empirical research; for

example: number of patent citations; perceived level of difficulty of imitation by

competitors; degree to which a firm acquires knowledge about its products or

components by tailoring its products to specific customers or applications

(customisation); and location of knowledge regarding resources (to name but a few).

These measures relate to the work of Zander and Kogut (1995) who develop

five constructs to measure the extent to which capability or knowledge and skills can

be easily communicated and understood; namely, codifiability, teachability,

complexity, system dependence, and product observability. The results of empirical

testing on 44 types of innovation across 20 manufacturing firms indicate that

codifiablity and teachability are the most significant factors for affecting speed of

transfer (or imitability) of capability or knowledge and skills, and give the most

direct insight into the extent to which knowledge is tacit and difficult to


communicate. Codifiability reflects the extent to which knowledge can be articulated

in documents and manuals, while teachability is related to the ease with which

knowledge can be communicated and learnt by new employees (or, in this research,

applied in new projects or customisation).

Thus drawing inspirations from Zander and Kogut’s (1995) questionnaire, the

questions for measuring costly-to-imitate are designed to determine the level of

difficulty of writing and delivering the activity according to the performance brief or

manual, and to determine the level of difficulty involved in all top-tier specialist

local firms (capable of delivering the activity) to develop the same knowledge and

skills required to deliver the activity. After further comments from field experts, the

final questions were framed as follows:

3. Costly-to-imitate a. How difficult would it have been for a performance brief or manual to be written

(that reflects knowledge, policies and procedures) for this activity in the case study and followed in order to deliver the activity in the case study?

Respondents requested to circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult

b. How difficult would it have been for all top-tier specialist local firms capable of delivering the activity to develop the same knowledge and skills required to deliver the activity in the case study?


Having discussed the fundamental principles of operationalising the RBT

variables in the literature, the next sections focus on the design of measures to

capture the dimensions of TCE variables.

4. Asset specificity

As discussed previously, asset specificity is the most important variable in

assessing the potential for hold-up. Given that asset specificity refers to the extent to

which an investment is firm-specific and cannot be redeployed easily for alternative

purposes, the main purpose of the ‘asset specificity’ question was to measure the

extent to which an investment is the most acute source of hold-up. Information from

the meetings with industry experts determined the most important type of asset

specificity creating potential hold-up. Furthermore, evidence was sourced from the

internet to corroborate the most significant form of investments or assets related to

each activity, and human resource and temporal asset specificities were found to be


the most critical. Although physical or site asset specificities in construction are

significant in terms of sunk costs, the construction-related equipment and assets are

most likely adaptable, and can be redeployed to other locations or types of projects,

(Masten, Meehan, & Snyder, 1991, p. 8). They are not, therefore, considered to be a

vital source of potential hold-up.

Human resource asset specificity is related to the time and effort needed to

reach full performance or level of customisation (or adaptation) required by a

supplier or firm (Heide & John, 1988; John & Weitz, 1988; Poppo & Zenger, 1998).

Anecdotal evidence from interviews reveals that the establishment of tacit

knowledge and expertise (human resource asset specificity) in the construction

industry developed in carrying out the activities requires a significant amount of time

and investment. Poppo and Zenger (1998) measure human asset specificity in terms

of the level of skill assets required to perform the activity, including the level of

customisation in terms of a set of procedures or functions. Similarly, the question is

designed to measure the level of investment and/or time needed to acquire new

knowledge and/or adapt existing knowledge to deliver the activity in the case study,

and on a 7-point scale of minimum level at 1 to substantial level at 7. The question is

given as follows:

4. Asset specificity a. How much investment and/or time would be needed to acquire new knowledge

(including software/hardware) and/or adapt existing knowledge (including software/hardware) to deliver the activity in the case study; that is, beyond knowledge already possessed by a top-tier specialist local firm with capability to deliver the activity?

Respondents requested to circle: minimal investment and/or adaption time 1 2 3 4 5 6 7 substantial investment and/or adaption time

Temporal asset specificity is associated with the difficulty or costliness of

replacing the supplier or firm in the midst of a project (Masten, Meehan, & Snyder,

1991, p. 9). Given the importance of timing and coordination in the delivery of

projects, the default of a firm (activity) can significantly impact on the progress of

the project and budget, especially those activities on the critical path of the program,

and potentially hold-up government due to the cost of switching firms (Eccles,

1981b). Although the skills and knowledge required to carry out the activities might

be sourced fairly easily in the construction market, it is the difficulty of sourcing an


alternative supplier at short notice that creates the potential for hold-up. For example,

Poppo and Zenger (1998) measure temporal asset specificity in terms of how costly it

is, in terms of time and resources, to switch to the function of outsourcing (if

internalised), or to switch suppliers or firms (if externalised). Thus, the temporal

asset specificity question measures the extent of impact on project timeline and/or

budget if government decides to externalise the activity (in the case study) and

replace the firm providing the activity; it measures this on a 7-point scale of minimal

impact at 1, to substantial impact at 7. The question is as follows:

4. Asset specificity a. At any stage in the delivery of the activity in the case study project, how much

negative impact (in terms of effect on the project’s timeline and/or budget) would have been experienced by the state department, if the state department had decided to externalise the activity and replace the firm providing the activity to the case study?

Respondents requested to circle: minimal impact 1 2 3 4 5 6 7 substantial impact

5. Uncertainty

In the process of determining the most suitable types of uncertainty,

endogenous or behavioural uncertainty was discounted because the focus was on

identifying potential sources of costly variation in timeline and/or budget and,

therefore, is more concerned with exogenous uncertainty. Exogenous or primary

uncertainty can arise due to: 1) environmental uncertainty, that is, random acts of

nature (Williamson 1985); 2) volume uncertainty (Walker & Weber, 1984), that is,

unpredictable changes in scope or demand or specifications (Bridge 2008, p.103); 3)

technological uncertainty, that is, the extent to which the skills required are rapidly

changing (Poppo and Zenger 1998); or 4) uncertainty arising from time that can

result in changes in knowledge and skills. However, for most of the activities,

technology is not rapidly changing, and uncertainty due to time in the long-term is

less relevant to the research.

For these reasons, the first question pertaining to uncertainty addresses volume

uncertainty by measuring the predictability of each activity – on a 7-point semantic

scale, from ‘extremely straightforward’ at 1, to ‘extremely difficult’ at 7 – in terms of

types and amount of time required by an expert in the field. The question is as

follows:


5. Uncertainty a. a. How much were the tasks (types and amount of time) in the activity required

in the case study straightforward to predict by someone with expertise in the activity in major projects?


The second question, then, addresses the uncertainties caused by environmental

factors (that is, by unknown factors relating to changes in physical conditions and/or

state department changes), by measuring the likelihood of the activities to be subject

to change in terms of task types, and the amount of time taken for these tasks, as

shown in the following question. The likelihood of changes due to environmental

factors are measured on a scale of ‘extremely small’ at 1, to ‘extremely large’ at 7.

5. Uncertainty b. How much were the tasks (types and amount of time) in the activity required in

the case study likely to be subject to changes caused by unknown factors relating changes in physical conditions and/or state department changes (e.g. scope) during the period allowed for the activity in the case study?

Respondents requested to circle: extremely small 1 2 3 4 5 6 7 extremely high

6. Frequency

TCE envisages the frequency dimension in terms of two elements; that is, the

extent of a portfolio of “large and recurrent transactions” (Williamson, 1985, p. 60)

including, but not exclusive to, the focal transaction or activity, such that more

frequency justifies costs of internalisation (internal transaction costs are spread).

Frequency is then deployed in a classical production cost way to capture division of

labour, which is limited by the extent of the market (Smith, 1776). That is, future

demand or work is needed to seek or justify benefits from investments in learning

economies and/or technology to achieve economies of scale. This means that

frequency is measured on the focal firm’s side only (not both sides and, in this case,

state government side only). It is measured in terms of whether there are large

transactions or activities (using the same resources as the focal transaction or

activity), and whether there are recurrent large transactions or activities (using the

same resources as the focal transaction or activity) being generated by state

government and into the feasible future.


The issue then becomes the definition or the methods of measuring ‘large’ and

‘recurrent’. In this research, ‘large’ is defined as a relative measure of the size of the

transaction or activity, and greater than the majority of all new similar kinds of

transaction or activity demanded, or in the market concerned. ‘Recurrent’ is defined

as being, how mainstream or typical the transaction or activity again in demand or in

the market concerned (mindful of a future period beyond the decision date of more

than one year duration up to the point that state government can reasonably estimate;

say, to the next state election).

In terms of measuring frequency, the model recognizes that there are different

ways to fully capture the potential to generate economies of scale and/or learning

economies, and to leverage natural or network advantages across road and health

projects.

a) Measuring the frequency variable in road projects

The issue being observed is the design and construction of public sector road

projects which comprise of resources and activities that are also present in other

types of public and private sector construction projects. Thus, the demand and

market surrounding design and construction activities is much bigger than the design

and construction of any one public sector road. In contrast, the demand or market for

the operations and maintenance activities of a road is largely represented by state and

suburban networks; hence, this demand is owned or generated by the government

and creates opportunity for economies of scale and/or learning curve economies

beyond the scope of the operations and maintenance of the focal road.

Consequently, frequency, which is designed to capture the potential for

economies of scale and/or learning curve economies and which, at the same time,

captures the potential to offset increased costs of internalisation, can be more

accurately measured at the level of project in terms of its size element, with respect

to the design and construction of new public roads. In contrast, in relation to

operations and maintenance activities of road projects, the size element of frequency

variable, can be better reflected by the level of aggregate of the roads in the network

and approximately, therefore, by the geographical scope. Given that larger projects

create larger size activities and vice versa, frequency need only be measured twice in

roads; namely, 1. at the project level to represent the frequency of all design and


construction activities and 2. at the network level, or in terms of geographical scope

of all the operations and maintenance activities in roads.

Given that the definition of ‘a project’ incorporates a discrete timeline and

budget, frequency in relation to design and construction activities is measured in

discrete project terms; that is, in terms of its capital value size, and the extent to

which the (case study) project is typical. Operations and maintenance of a road

project also falls within a recurrent annual budget and timeline; thus, frequency is

measured in terms of the geographical scope of the network, and in terms of the

extent to which the operations and maintenance of the project is typical.

To illustrate further: The fact that road maintenance activity is highly

contestable and that there are many firms that are capable of delivering road

maintenance, shows that the core technical knowledge and skills required in

maintenance are widely disseminated and, therefore, afford neither state government

nor market a competitive advantage in the possession of the inherent technical

knowledge and skills required. Thus, other economic drivers determine the relative

advantages of state government delivery versus market sector delivery of this activity.

Specifically, more time to plan before implementation, and the high unpredictability

and frequency of an activity, affords advantages to the party that can potentially

access the entire and uninterrupted knowledge of the road network. Because the state

government has the responsibility for and primary rights over access to knowledge of

the entire network, which means that through internally delivering road maintenance,

state government can create an entire and uninterrupted knowledge of the network.

The following notes briefly explain how government can leverage these potential

advantages with respect to three types of maintenance activities in road projects, and

create synergies from the interrelationships of these three types of activities:

• Activity A: This first type of activity is related to the planning for inspections

or data collection and planning conducting maintenance. The accumulated

knowledge of the entire road network is a substantial advantage in this activity,

in terms of control and strategic planning of the nature of maintenance to

undertake on individual roads, so as to allocate available funds most effectively

across the entire network. This activity is associated with consistency or quality

of the presentation and condition of roads for users across the entire network,

and is a key part of the logic behind Level 2. This planning activity is ongoing


or has high frequency, and there is sufficient time to allow consideration of the

kind of maintenance to undertake. Given that maintenance is dynamic – with

the exposure of assets to elements of weather, such as sun and rain – a key part

of the knowledge for this activity is tacit, and is generated by more general

implementation of maintenance; that is, by conducting inspections or data

collection and implementation of routine and programmed maintenance in (the

next) Activity B. These inspections and data collection activities on the

condition of network, are affected or restricted by the travel distances and

resources allowed. An exception arises with respect to any highly specialist

type of maintenance that is not typical of the network and which, therefore,

does not benefit from an entire and uninterrupted knowledge of the network.

• Activity B: A second type of maintenance activity refers to conducting

inspections or data collection and implementation of routine and programmed

maintenance. Inspections can be conducted formally and informally, or in tacit

mode; that is, en route or as part of maintenance, and fed back into the formal

and informal knowledgebase used in Activity A. Time is important and time

restrictions apply more so with routine than with ad hoc maintenance. The

greater the accumulated knowledge of the network, the more significant is the

advantage in terms of tactical planning; that is, with respect to the timing, the

location, the sequence or clusters of roads, so as to conduct maintenance in the

most efficient way. Opportunities for efficiencies increase as access to the

network increases. This factor is a key part of organising resources, and a key

component of the logic behind Level 3. However, a tension can arise as a result

of the unpredictable nature of maintenance, that can give rise to variations; this

tension also increases as access to the network increases. In turn, this can create

the potential for negative opportunistic behaviour or hold-up by the contractor,

and is a key part of the logic behind Level 4. A level 4 activity can be delivered

efficiently in a relational arrangement with a market firm, such as an alliance,

where the project or transaction is infrequent. However, this activity is ongoing

and frequent, and internal costs are divisible, and diminish on a unit basis as

more of the network is delivered by internalisation.

• Activity C: A third type of maintenance refers to conducting inspections or data

collection, and implementation of emergency maintenance. Response time is

critical, and this attribute mutes the advantages of the knowledge of the


network in terms of the role and benefits of planning. The key economic

imperative arises, therefore, from internal costs of organising a speedy response,

ensuring control over the response, and avoiding external costs arising from

hold-up, costly variations, and an inability to foresee all eventualities (a key

part of the logic behind Level 4). In this regard, internalisation has advantages

over externalisation. Moreover, if state government delivers Activity B

internally, then the cost of Activity C is marginal only, or the unit cost of

Activity B is further divisible by Activity C.

As a preamble note to the modelling of the road cases, Activity B (in particular,

preventive activities) and Activity C are treated as anticipated or planned

procurement (that is, in the future), in contrast to design, construction and operations

that have been procured (that is, in the past). Activities B and C are still relevant,

even though the decisions have yet to be made in advance of actual maintenance.

In summary, key network advantages lie naturally with state government with

respect to roads. Activities in which state government has an advantage over the

private sector arise from knowledge of the state’s network, such as non-specialised

maintenance activities. Given the dynamic condition of roads being fully exposed to

inclement weather, the state government is involved in updating knowledge

concerning the condition of the entire network; and to plan and spend monies on

maintenance to deliver consistency for road users across the entire state network.

b) Measuring frequency variable in health projects

In contrast to roads, key network advantages in health buildings arise within

the physical boundaries of the building, particularly in relation to the network or

integration of highly technological and highly specialised mechanical and electrical

services (including the Building Management System). Hence, in the case of health

infrastructure, the size element in the frequency variable across all design,

construction, operations and maintenance activities is more accurately measured at

the project level. As a result, to represent the frequency of all activities, the

frequency of each activity need only be measured once at the project level.

On measuring frequency, it is not tautological to ask what is the most

frequently occurring project, that is, in terms of typicality and size (or capital value),

as frequency relate to the entire project discernible by its physical boundaries, such

as geography, length of roads, or gross floor area (GFA) in the case of health


projects. The model does not necessarily accept the project as one contract; rather, it

determines whether it is best sized as one contract, or broken down into smaller

contracts. In summary, frequency is measured in terms of typicality and size at the

project level, when there are no natural network advantages (Version A), and in

terms of geographical scope at the network level when there are network advantages

favouring state government provision (Version B). This is determined as follows:

6. Frequency (Version A at the project level) a. In the state electoral period surrounding the case study (from commencement of

state government term to end of term incorporating the case study procurement decision date), how did the case study compare with the state department’s most frequently occurring or typical new project in terms of capital value and what is the capital value size category of the case study? Please circle: 1. = Case study is not typical, and a small project ($1 to $10 million) 2. = Case study is not typical, and a moderately-sized project ($10 to $100

million) 3. = Case study is not typical, and a large project ($100 to $250 million) 4. = Case study is not typical, and a very large/extremely large project (greater

than $250 million) 5. = Case study is typical, and a small to moderately-sized project ($1 to $100

million) 6. = Case study is typical, and a large project ($100 million to $250million) 7. = Case study is typical, and a very large/extremely large project (greater than

$250million) 6. Frequency (Version B at the network level) a. In the state electoral period surrounding the case study (from commencement of

state government term to end of term incorporating the case study procurement decision date), how did the operations and maintenance to case study compare with the state department’s most frequently occurring or typical operations maintenance and what is the geographical scope of the state department’s operations and maintenance responsibility? Please circle: 1. = Case study was not typical and a small scope (main roads in state capital

city) 2. = Case study was not typical and a moderately scope (main roads in state

metropolitan area) 3. = Case study was not typical and a large scope (main roads statewide) 4. = Case study was not typical and a very large/extremely large scope (main

roads statewide and suburban roads) 5. = Case study was typical and a small to moderate scope (main roads in state

capital city to metropolitan area) 6. = Case study was typical and a large scope (main roads statewide) 7. = Case study was typical and a very large/extremely large scope (main roads

statewide and suburban roads)


4.4.7 Measurement scale

The questions were designed to measure the perceptions of the respondents in

relation to RBT and TCE variables. Consistent with the approach to the measurement

of observation or perception taken by other resource-based researchers, such as

Anderson and Schmittlein (1984) and Poppo and Zenger (1998), a 7-point semantic

differential scale was used to represent the level of degree from low to high for all

questions. By answering the (above) RBT and TCE questions for each activity, an

empirical pattern was generated for each activity, and was then matched with the

theoretical pattern of closest fit or best-fit of the variables in the framework. The

corresponding activity level reflects the competence and capabilities surrounding the

government and the market during the time that the procurement decision was being

made.

4.4.8 Administration of case studies

a) Case study selection

Upon finalising the design of the case study approach and case study

questionnaire instrument, the next step was to identify case studies according to

literal and theoretical replication, as discussed in Section 4.4.4.

To select the cases study projects from the sample of 87 projects, all projects

were firstly rank ordered according to EoI numbers and categorised into road and

health sectors, and then further grouped into optimal EoI and sub-optimal EoI. Figure

4.4 shows the spread of EoI in the sample ranges from (largely) low levels of EoIs to

15 EoIs.

Figure 4.4: Sample of submitted road and health projects

0

2

4

6

8

10

12

14

16

2 3 4 5 6 7 8 9 10 11 12 13 14 15

Nu

mb

er

of

pro

ject

s

Number of contractors expressed interest at EoI

Roads

Hospitals


In the category of optimal EoI (between 5-8), there were 26 road projects and 8

health projects available for selection. Road Case Study #R1 with 8 EoI and Health

Case Study #H1 with 5 EoI were chosen randomly. With regards to sub-optimal

category of EoI (outside of 5-8), the approach taken and in pursuance of theoretical

replication (as explained in Section 4.4.4), was to select cases at the extreme ends of

the EoI levels. Hence, from the case studies given in Figure 4.4 (above), Health Case

Study #H2 with 15 EoI was selected and Road Case Study #R2 was selected

randomly from amongst the 8 road projects with 2 EoI. In selecting two cases in both

the optimal and sub-optimal EoI categories, literal replication was also achieved.

Having identified the case studies from the sample of projects obtained from

the survey of major road and health projects, initial meetings were arranged with the

directors of road and public works authorities, to obtain approvals. The research

objectives and the procedural steps of the model were explained in these meetings,

and the relevant contacts for each project were then identified. Table 4.22 (below)

summarises details of the selected case studies.

Table 4.22 Summary of selected case studies

Case study R1 R2 H1 H2 Sector Road Road Health Health EoI Optimal EoI

(8 EoI) Sub-optimal (low – 2 EoI)

Optimal (5 EoI)

Sub-optimal (high – 15EoI)

Capital value in categories

$50-100 million $250-500 million $250-500 million $250-500 million

Commencement 2009-10 2004-05 2006-07 2007-08 Actual procurement mode

Traditional construct only

Alliancing Public Private Partnership

Managing contractor

Actual payment terms

Fixed-price lump sum

Guaranteed construction sum with pain-share/gain-share

Fixed monthly payment to year 2035

Target outturn cost with pain-share/gain-share

b) Case study data collection

Documents and drawings were requested from each relevant government

department or agency. Site plans, construction drawings, contract documents and

program of works, helped to define the scope of the project and gave an indication of

activity analysis. For Road Case #R1, the hardcopies of the contract documents were

formally requested, and retrieved from the archives. The documents were perused,

and relevant documents photocopied on-site on a couple of occasions. For Road Case

#R2, the documents were stored in digital format, partly in a hard-disk and partly in

the state government project database. There was a great deal of effort involved in

the search for the required information and the assistance and support of the Senior

Program Support Officer was greatly appreciated. The documents for Health Case


#H2 were obtained from the department in digital format, while the documents for

Health Case #H1 were received via emails. To ensure construct validity and internal

consistency of the research, all information was documented in the case study

database (See Appendix J).

Table 4.23 Data collection register (in chronological order)

No Date Duration Data collection 1 17/07/2012 2 hours Collecting documents for #H2 2 18/07/2012 4 hours Printing retrieved archived documents for #R1 3 26/07/2012 4 hours Printing retrieved archived documents for #R1 4 13/12/2012 3 hours Searching and printing documents for #R2 5 10/01/2013 4 hours Searching and printing documents for #R2 6 31/05/2013 4 hours Printing retrieved documents for #H1

c) Structured interviews

The structured interviews were sequentially carried out with the project

manager of each case study, beginning from Road Case Study #R1. The main

contributions from the interviewees were: 1. input into the identification of activities

for each case study in Activity analysis; 2. answering the questions related to

measurement of TCE and RBT variables, and matching the levels in Make-or-buy

analysis; and 3. indicating the supply or competitiveness of the market surrounding

each activity. Overall, a total of approximately 110 hours were spent on the

administration of all four case studies. The entire process of case study method began

in July 2012 and took almost 13 months to complete, as shown in Table 4.24.

Table 4.24 Case study meeting register (in chronological order)

No Date Duration Time Aim of interview 1 03/12/2012 3-4 hours 1230-1500hr Activity and make-or-buy analysis for #R1 2 12/12/2012 7 hours 1000-1700hr Make-or-buy analysis for #R1 3 17/12/2012 3 hours 0900-1200hr #H2 first meeting 4 16/01/2013 5 hours 1000-1500hr Begin Activity analysis for #R2 5 24/01/2013 5-6 hours 1100-1500hr Activity analysis for #R2 6 07/02/2013 7 hours 1200-1900hr Activity analysis for #R2 7 15/02/2013 7-8 hours 0930-1700hr Activity analysis for #R2 8 19/02/2013 8 hours 0900-1700hr Make-or-buy analysis for #H2 9 21/02/2013 1-2 hours 1200-1330hr Make-or-buy analysis for #R2 10 25/02/2013 8 hours 0900-1700hr Make-or-buy analysis for #H2 11 05/03/2013 5-6 hours 0930-1500hr Make-or-buy analysis for #R2 12 08/03/2013 8 hours 0900-1700hr Make-or-buy analysis for #H2 13 12/03/2013 8 hours 0900-1700hr Make-or-buy analysis for #H2 14 26/03/2013 2 hours 1300-1500hr Finalise and approval of #H2 15 27/03/2013 8 hours 0900-1700hr Make-or-buy analysis for #R2 16 28/03/2013 8 hours 0900-1700hr Make-or-buy analysis for #R2 17 11/07/2013 8 hours 0900-1700hr Analysis for #H1 18 18/07/2013 8 hours 0900-1700hr Analysis for #H1 19 20/08/2013 2 hours 1400-1600hr Analysis for #H1


The next section provides more detail of the process of activity analysis and

administering the case study questionnaire in the structured interview in each case

study.

1. Administering Road Case #R1 in review mode

For Road Case Study #R1, a list of 53 activities were initially identified as

potential key activities, by using information obtained from road project

documentation before the interview. During the interview, the project manager

identified several activities which belonged to similar technological boundaries, or

which were trivial in size relative to the project. In this way, the initial 53 activities

were reduced to 23 key activities in the first iteration, and to 18 key activities in the

last iteration. The results of Activity analysis for Road #1 are given in Section 5.3.1.

Using the case study questionnaire, the project manager was required to

indicate internalisation or externalisation, and to match the level of integration using

the 7 levels of vertical integration by answering the questions (including

identification of the greatest potential source of costly variation in time and/or

money, and the specific investments in each activity). The results of these tasks are

summarised and discussed in Section 5.3.2.

2. Administering Road Case #R2 in review mode

Road Case Study #R2 was a significantly larger and more complicated project

than #R1, and a significant amount of effort was needed to understand the extent and

boundaries of this project. As for #R2, the number of key activities for #R2 was

scoped down by the project manager from an initial 144 to 61, after a number of

iterations. The activities are listed in full in Section 5.4.1.

The project manager’s input was important in explaining the TCE and RBT

dimensions surrounding each activity. This was particularly the case for: the asset

specificities and uncertainties related to the driven tunnel construction; its interaction

with cut-and-cover tunnel; coordination with rail operators in scheduling

uncertainties; and staging issues related to alignment with existing road

infrastructure. The TCE and RBT scores were then matched to a level in the

integrative framework. These results are summarised in Section 5.4.2. The project

manager identified a number of troublesome activities – Level 4b and Level 7

activities arising out of rare technology – and also highlighted the limited supply of


firms in the local market able to provide the design and construction of the driven

tunnels in Level 7 activities.

3. Administering Health Case #H2 in review mode

In relation to Health Case Study #H2, the project manager streamlined the key

activities from the initial 88 key activities to a final 56 key activities. This included

the identification of specialist items that are unique to a hospital, such as the Building

Management System, and specialist hydraulics (differentiated from the regular

hydraulics system), such as medical gases and pneumatic tube systems. The key

activities for Health Case #H2 are given in Section 5.5.1.

Subsequent time was mostly spent on answering the questions, given the large

number of activities generated by the nature of the project, and the need to

understand the nature of these activities in relation to TCE and RBT dimensions, and

the supply or market surrounding each activity; in particular, those activities that

required specialist knowledge. The project manager also clarified the extent of the

state government’s maintenance capability in the health sector, which is mainly

related to non-specialist activity. This process of examining the TCE and RBT

attributes of the maintenance activities identified some with mismatching levels with

actual practice. The results are given in Section 5.5.2.

4. Administering Health Case #H1 in review mode

Given the similarity in the nature of the health projects, the activities of #H1

were similar to those of #H2, and a total of 79 activities were identified by the

project manager. These are listed in Section 5.6.1. Equipped with a better

understanding of activities in health projects as a result of the analysis of #H2, the

process of understanding the TCE and RBT attributes in association with each

activity was more straightforward. The scores and matching levels are summarised in

Section 5.6.2.

The next section discusses the procedures for obtaining ethical clearance for

both the survey and case study research.

4.5 Ethics

Ethics approval is required for any data collection that involves human

participation. Even though this research involved low risk primary data collection


through human participation in surveys and interviews, ethics approval was required

for each data collection instrument before distribution to participants. Upon

finalisation of each data collection instrument, application for ethical clearance was

submitted as soon as possible, to prevent any delays in research progress. The details

of the research were submitted, together with the final version of the data collection

instrument, to the QUT Human Research Ethics Committee for review. Individual

ethical clearance was approved and obtained for the survey of major road and health

projects, the survey of civil and building contractors, and the case study

questionnaire at different stages of the research. This process ensured that the

research was in compliance with the requirements of the National Statement on

Ethical Conduct in Human Research, and the QUT Code of Conduct for Research.

4.6 Summary

This chapter sets out and justifies the research methods and techniques

employed in the research. Considerations of reliability and validity were employed in

both the survey and case study methods. The key components of the survey method

were employed in this research and discussed in detail; that is: design and

development of the data collection instrument, the population and sampling frame,

administration of the survey and/or interviews, data collection, data entry and

analysis, and the methods used to achieve a higher response rate in the self-

administered survey of civil and building contractors.

A preliminary analysis of the representativeness of and responses to the survey

of major road and health projects and the survey of civil and building contractors is

presented. The approach to selecting case studies and the underlying logic in testing

the hypothesis using the case study method are presented. Lastly, this research was

conducted in accordance with QUT ethics policy for human research.

The next chapter examines the data set obtained from the survey of major road

and health projects in terms of observing any patterns in procurement approaches in

the road and health sectors, the general relationship between aspects of procurement

with competition, and the results of testing the procurement model based on the

procurement-competition/flexibility hypothesis.

Chapter 5: Analysis of survey and case study data 171

Chapter 5: Analysis of survey and case study data

5.1 Introduction

With reference to Sections 3.3.2 and 3.3.3, the validity of EoI as the dependent

variable, and as a key indicator of both competition and flexibility and a proxy of

market failure, has been established. Specifically, relatively low EoI (4 EoIs or less)

indicates market failure pre-contract due to a lack of competition and creates

oligopolistic pricing conditions; whilst high EoI (9 EoIs or more) indicates market

failure post-contract due to a lack of flexibility in contractual arrangements and

creates potential hold-up arising from troublesome activities. In other words,

avoiding market failure pre-contract by optimising competition, can lead to lower

capital costs and greater incentives for innovations in design that reduces whole-life

costs and improve VfM; and at the same time, avoiding market failure post-contract

by excluding troublesome activities from the main activity bundle and contracting

separately using contractual arrangements that addresses hold-up, can minimize

external transaction costs related to hold-up and again improving VfM.

As a complement to establishing, consistent with VfM, this chapter begins by

presenting a test of the reliability of EoI as the dependent variable. That is, key

dimensions of actual procurement are assessed for their association with EoI (as part

of actual competition), using results from the questionnaire survey of major road and

health projects. The remainder of the chapter comprises an analysis of the case

studies and an assessment of the extent to which the results from each case study

(theoretical or predicted procurement) support (or otherwise), the procurement-

competition/flexibility hypothesis. In some cases, this also involves an analysis of

theoretical competition, using results from the questionnaire survey of civil and

building contractors. Finally, a discussion of the findings is given.

5.2 Questionnaire survey of major road and health projects

5.2.1 Actual procurement

The distribution of various procurement types in road and health sectors in the

sample of 87 projects are shown in Table 5.1 (below). This table provides an

172 Chapter 5: Analysis of survey and case study data

overview of the distribution of different types of procurement and includes, in each

procurement mode, the most frequently occurring value category and total value. As

shown in Table 5.1 (below), the majority of the road projects have been procured

using Construct only (CO) and Design and Construct (DC). These comprise 63

percent of the total number of road projects, representing 32 percent of their total

value. Alliance Contracting (AC) and PPP projects, on the other hand, comprises

only 25 percent of the total number of road projects and represent 52 percent of the

total value of all road projects. This is evidence that AC and PPP projects are mainly

higher value projects, while CO and DC are the main procurement approaches

associated with lower value projects. On the other hand, Managing Contractor (MC)

dominates the health projects submitted in terms of size (70 percent of the total value

of health projects) and number (50 percent of the total number of health projects).

In relation to the inclusion of operations and maintenance and the design and

construct of the project, there are only five projects across both road and health

sectors that include the operations and/or maintenance (O&M) component(s). These

account for the minority of the total value of the projects.

Table 5.1 Actual procurement types across road and health sectors

Procurement Type

No. of road projects

Roads value category mode (#)

Roads value $b

No. of health Projects

Health value category mode (#)

Health value $b

Total no. of projects

Total value $b

CO 24 50-100M (7) 100-250M (7)

3.392 6 50-100M (3) 0.514 30 3.906

DC 14 100-250M (5) 3.383 1 NA 0.145 15 3.528 DCM 1 NA 2.200 0 0 0 1 2.200

DCOM 0 0 0 0 0 0 0 0

CM 0 0 0 1 NA 0.230 1 0.230

MC 0 0 0 13 250-500M (4) 100-250M (4)

7.040 13 7.040

ECI 6 100-250M (3) 1.608 2 NA 0.225 8 1.833 AC and PF 16 100-250M (6) 11.560 3 250-500M (1)

500M-1B (1) > $1B (1)

2.000 19 13.560

Total 61 22.143 26 10.154 87 32.297 Key • CO = Construct only • DC = Design & Construct

• DCM = Design, Construct & Maintain

• DCOM = Design, Construct, Operate & Maintain • CM = Construction Management

• MC = Managing Contractor • ECI = Early Contractor Involvement

• AC = Alliance Contracting • PF = Any procurement mode that includes private finance (e.g. Public-Private Partnership)


Table 5.2 (below) shows the extent to which projects in the sample are

delivered as single or multiple contracts, the most frequently occurring procurement

type and value category, and their total value. The majority of road and health

projects are delivered as single contracts and, compared to projects delivered as

multiple contracts, tend to comprise the two lower value categories (that is, below

$250 million). Thus, the procurement approach is not appreciably affected by the

practice of multiple contracts, as now shown in Table 5.2.

Table 5.2 Single or multiple contracts across road and health sectors

Single or multiple contracts


Roads procurement type mode (#)


Roads value $b (%)

No. of health projects

Health procurement type mode (#)


Health value $b (%)


Total value $b (%)

Single contract

44 DC (15) 50-100M (20)

12.289

(56%)

18 MC (8) 250-500M (7)

7.117

(70%)

62 19.406 (60.1%)

Multiple contract

17 DC (10) 100M-250M (6)

9.854 (44%)

8 MC (6) 100-250M (3)

3.037

(30%)

25 12.891 (39.9%)

Total 61 22.143 26 10.154 87 32.297

Table 5.3 (below) gives the spread of various payment terms and, again,

includes details concerning the most frequently occurring procurement type, value

category, and total value. Sub-total A represents projects in the sample in which the

contractor takes the contractual risk for the upper limit in the price, whereas Sub-

total B shows projects in which government takes the contractual risk for upper limit

in the price.

Table 5.3 Payment terms across road and health sectors

Payment terms


Roads procurement type mode (#)


Roads value $b

No. of health projects

Health procurement type mode (#)


Health value $b


Total value $b

Lump sum (fixed price) including fixed periodic payments

32 Design and Construct (12)

0-100M (10)

13.054 11 Construct Only (6)

50-100M (4)

2.742 43 15.796

GCS/GMP (including share of savings)

1 Construct Only (1)

250-500M (1)

0.315 10 Managing Contractor (10)

250-500M (5)

6.665

11 6.980

Sub-total A 13.369 (60%)

9.407 (93%)

54 22.776 (71%)

Remeasurement (including schedule of rates and prices)

10 Construct Only (10)

0-100M (7)

0.772

Nil Nil Nil Nil 10 0.772

Target outturn cost and cost plus (including pain/gain share)

18 Alliance (14)

$100-250M (8)

8.002

5 Managing Contractor (2)

100-250M (4)

0.747

23 8.749

Sub-total B 8.774 (40%)

0.747 (7%)

33 9.521 (29%)

Total 61 22.143 26 10.154 87 32.297


A conspicuous pattern can be observed in terms of risk allocation; that is, for

the majority of road and health projects, the contractor takes the contractual risk for

the upper limit in the price. Health sector projects show an appreciably greater

reliance on the contractor taking the contractual risk for the upper limit in the price.

These cases represent 93 percent of the total value of health projects, whereas road

sector projects in which the contractor takes the contractual risk represent only 60

percent of the total value of road projects. This represents a point of difference

between road and health projects, where the contractor in the latter has a strong

tendency to undertake the contractual risks for the upper limit in the price, which

corresponds with the dominance of MC in health sector. In contrast, in the former,

there is a greater balance in terms of government and the contractor taking the

contractual risks for the upper limit in the price and this corresponds with CO, DC

and AC as the three most popular procurement modes in the road sector.

5.2.2 Actual competition

Table 5.4 and Table 5.5 (below) show the distribution of EoI in road and health

sectors respectively across the years studied. In addition, the tables also display the

number of contractors that were invited to tender (ItT) and submitted tenders (ST)

across these years. As discussed in Section 3.3.1, the attractiveness of the project to

the market can be indicated by the number of firms demonstrating their willingness

to bid either through EoI or the number of firms submitting bids in open tender by

eligible contractors. In both Table 5.4 and Table 5.5, it can be seen that the number

of contractors that bid during open tender (OT) are included in the EoI figures. EoI is

differentiated from ItT and ST, which are at the end of a three-stage filtering process,

and do not reflect the actual level of interest in the market as clearly as EoI and OT.

Typically, in the two government sectors studied, the approach to tendering

starts with EoI, or a request for proposals. These EoIs or proposals are initially

assessed in Stage 1, and around four contractors are typically invited to submit. In

Stage 2, two or more proponents are then selected for further interviews and

selection workshops to prepare and submit tenders. Finally, the selected proponent

progresses to Stage 3 to finalise the commercial arrangements and agreement.


Table 5.4 EoI, ItT, and ST in the road sector

Date Missing data

1 2 3 4 5 6 7 8 9 10+ Mean Total

2006 • EoI (#)

incl. 2 x OT (3; 4)

• ItT (#) • ST (#)

1 1 1

2 3 3

2 1 2

4 2 1

1 1 1

1 1 1

1 1 1

4.27 4.22 4.11

12 10 10

12

2007 • EoI (#)

Incl. 1 x OT (2) • ItT (#) • ST (#)

0 0 0

4 2 3

1 4 4

2 1 1

2 1 1

2 1 1

1 2 1

1 1 1

4.31 4.42 4.00

13 12 12

13 2008 • EoI (#)

Incl. 4 x OT (3; 4; 5; 7)

• ItT (#) • ST (#)

0 0 0

1 1

1 3

5 7 4

4

2

1 1 1

5.31 4.67 4.33

13 9 9

13 2009 • EoI (#)

Incl. 3 x OT (4; 5; 6)

• ItT (#) • ST (#)

1 1 1

1

1 3 2

2 3

2 1

3 2 2

2

1

1 2 1

3 2 2

5.93 5.91 4.83

15 12 12

15

2010 • EoI (#) • ItT (#) • ST (#)

0 0 0

1

1 2 1

1 1

1

4 2 1

1

2 2 1

1 1 1

6.00 5.38 5.00

8 8 8

8 Total projects 6 61

Table 5.5 EoI, ItT, and ST in the health sector

Date Missing data

1 2 3 4 5 6 7 8 9 10+ Mean Total

2007 • EoI (#)

Incl. 1x OT(2), 1xOT missing

• ItT (#) • ST (#)

5

1 2

7 5

2

1

3.75 3.00 2.71

9 7 7

9

2008 • EoI (#)

Incl. 1x OT(2) • ItT (#) • ST (#)

2 2 2

1 2 3

1 3 2

1

2 1 1

1

1

5.78 3.50 3.38

9 8 8

9 2009 • EoI (#) • ItT (#) • ST (#)

1 1 1

1 1

1 1

1

2 1 1

4.75 3.75 3.75

4 4 4

4

2010 • EoI (#) • ItT (#) • ST (#)

1

2

1

1 2 1

1

1 1

1

6.33 4.50 3.75

4 4 4

4 Total projects 6 26


Table 5.4 and Table 5.5 (above) show that the mean number of EoIs increased

appreciably from 2008, and post-GFC. Figure 5.1 (below) combines EoI of both road

and health projects, and illustrates the number of EoI in each year. The line of best fit

for average EoI begins to slope upwards from 2009.

Key • Linear = Line of best fit

• Mov. Avg.= Use two periods to smooth out fluctuations and highlight trends

Figure 5.1: Road and health EoIs

Figure 5.2 (below) illustrates combined EoIs in both road and health projects

against the procurement commencement date. The figure shows that the mean EoI


levels increase from approximately 4.27 EoIs in 2006, to approximately 6.09 EoIs in

2010. A significant increase in EoI occurred in 2008, at around the occurrence of the

GFC.

Figure 5.2: Road and health EoIs and commencement dates

Figure 5.3 (below) shows a stark decrease in EoIs at around $800 million, and

a range of EoIs between 2 to 15 contractors (with a mean of 5.16) for project values

below $800 million. For project values in excess of $800 million, the range of EoIs is

between 2 and 4 contractors, with a mean of 3.86.

Figure 5.3: Road and health EoIs and overall capital value


5.2.3 General relationship between procurement and competition

In order to determine if there are any significant relationships between EoI and

the key procurement dimensions in actual procurement, inferential statistical testing

was carried out with respect to EoI and capital value of project, procurement mode

and payment terms. The results show that there is a statistically significant

relationship between EoI (or potential competition) and each of the key dimensions

of procurement (which correspond to the nature of the procurement strategy

informed by the procurement model). These key dimensions comprise capital value

of the project (which is related to the size of the project); procurement mode (which

relates to bundling); and payment terms (which relates to the nature of exchange

relationship).

5.2.4 Key patterns from actual procurement, and indications of scope to improve VfM

Having established both the validity and reliability of EoI as an early visible

indicator that the procurement approach has set a project on a path towards superior

VfM, further analysis of the EoI in the sample indicates that there is appreciable

room for improvement in procurement. This is true for 57 percent of the projects

(allowing for missing EoI data) outside of optimal range which, accordingly, are

likely to be on an inferior path in terms of delivering VfM. Specifically, the sample,

comprises of: 34 (or 39 percent) projects in the sample with EoI in the optimal range

of EoI (5 to 8); 9 (or 10 percent) projects with high EoI (9 or more); 36 (or 42

percent) projects with low EoI (4 or fewer); and 8 (9 percent) projects with missing

EoI data.

The weaknesses of current procurement decision-making approaches (based on

MAUA, and as summarised in Section 1.3.1) undermine risk analysis, and are

expected to result in a lack of sophistication in the delivery of projects. Indeed, the

three key dimensions of actual procurement observed indicate a short-term

orientation; that is, a minimum time to opening day of the asset and start of

operations, and/or minimum capital cost. More specifically: 1. project size or value;

2. bundling; 3. exchange relationship (payment terms). These dimensions are detailed

below.


1. Project size or value

Across the sample of projects submitted in the survey of major road and health

projects, there are a low number of higher value projects that account for an

appreciably higher proportion of the value.

Forty of the submitted road projects (66 percent) comprise the two lower value

or most frequently occurring categories (between $50 to $100 million and $100 to

$250 million); these account for $4.164 billion (19 percent) of the total value of the

submitted road projects. At the same time, 10 of the submitted road projects (16

percent) fall in the two higher value categories (between $500 million to $1 billion,

and more than $1 billion), which account for $13.847 billion (63 percent) of the total

value of the submitted road projects.

In terms of health projects, 17 of the submitted health projects (65 percent)

comprise the two lower value or most frequently occurring categories (between $50

to $100 million and $100 to $250 million), which account for $2.024 billion (20

percent) of the total value of the submitted health projects. At the same time, five of

these projects (19 percent) fall in the two higher value categories (between $500

million to $1 billion, and more than $1 billion), which account for $6.593 billion (65

percent) of the total value of the submitted health projects.

In summary, the majority of road and health projects (62 projects, or 71 percent

of the submitted projects representing $19.406 billion, or 60 percent of the value of

submitted projects) are delivered as a single contract. In terms of projects delivered

as multiple contracts, these tend to comprise the two lower value categories below

$250 million.

2. Bundling

With regards to the 61 road projects, lower value projects ($50-100 million) are

dominated by CO (24 projects), while the larger value projects (over $100 million)

are dominated by: DC (15 projects); AC (14 projects); and ECI (6 projects). The

remainder are two projects that comprise DCOM (including a PPP). On the other

hand, MC in 13 projects (eight of which were greater than $100 million) dominates

the health projects submitted. Similar to the case with road projects, there is again a

small number of projects that comprise DCOM (namely, three PPPs).


3. Exchange relationship (payment terms)

Health projects show a much greater reliance on the contractor taking the

contractual risk for the upper limit in the price; that is, in 21 (or 81 percent) health

projects (representing $9.407 billion, or 93 percent of the value of submitted

projects). In contrast, 33 or 54 percent of road projects (representing $13.369 billion,

or 60 percent of the value of submitted road projects) rely on the contractor taking

the contractual risk for the upper limit in the price.

In summary, this analysis of actual procurement indicates scope for

improvement in procurement in pursuance of VfM. This improvement might well

relate to changes to the patterns observed in the sample, and concern the three key

procurement dimensions; for example, changes

1. From larger projects (comprised mainly of a single contract) to smaller projects

(comprised of more multiple contracts)

2. From very limited inclusion of operations and maintenance as part of the project

bundle, to a greater incidence of operations and maintenance bundled with

design and construction, and

3. From a limited use of multiple exchange mechanisms within a project (related to

dominance of single contracts) to more multiple exchange mechanisms within

the project

5.2.5 Summary and key patterns

The significant patterns in the sample are summarised as follows:

• A low number of higher value projects account for an appreciably higher

proportion of the overall value of the $32 billion sample studied. This pattern

continues even when PPPs are excluded from the sample (although this pattern

does become less pronounced).

• The majority of the road and health projects are delivered as single contracts.

• Road projects of a larger value (over $100 million) are dominated by DC, ECI,

and AC, with only a few comprising DCOM (including PPP). In health

projects, MC is a dominant approach and, again, only a few projects comprise

DCOM (namely three PPPs).

• Statistically significant relationships exist between: EoI and project size or

project value; EoI and degree of bundling; and EoI and exchange relationship


(payment terms). Furthermore, also related to project size, a stark decrease in

EoIs is observed at around $800 million.

In summary, given that the results demonstrate a significant relationship

between EoI and project value (or size of project), EoI and procurement modes (or

bundling), and EoI and payment terms (or nature of exchange relationship), there is

strong empirical evidence and support for the reliability of the procurement-

competition/flexibility hypothesis.

As indicated in Section 4.4, the next step after establishing the reliability of the

effects of key procurement mode (size, bundling and nature of relationship), is to

deploy the first-order procurement decision-making model, and to test the outcomes

of the model using the procurement-competition/flexibility hypothesis. In the

following sections, each case study analysis begins with an introduction to the

project in terms of nature of project, costs, time of construction and type of

procurement approach, followed by the outcomes of applying the procedural steps in

the model. Each case study analysis in the following sections includes activity

analysis, followed by make-or-buy analysis, market analysis, bundling analysis and,

finally, nature of exchange relationship analysis.

5.3 Road Case Study #R1 (Theoretical procurement)

Road Case Study #R1 had eight EoI and lies in the optimum range of 5-8 EoI.

Case Study #R1 comprised the widening of a six kilometre length of an existing

motorway from four to six lanes, including the installation of concrete medians and

noise barriers. The project was procured using a traditional procurement approach, in

which government engaged the design consultant and the main contractor separately

in the design and construct of the road project. The design was completed before the

tender stage. An invitation to tender was advertised to seek contractors that satisfy

the minimum pre-qualifications to tender for the project in an open tender; eight

contractors submitted their bids for this project. This project was awarded to a main

contractor using a standard construct-only contract. The operations and maintenance

were carried out internally by state government agencies, and as part of a network of

roads.

The following sections discuss the results of applying the procurement model

to this project.


5.3.1 Stage 1/Task A: Activity analysis

As explained in Section 3.2.2, the scope of the analysis of key activities starts

from (and includes) schematic design, and extends to (and includes) operations and

maintenance of an asset, and the focus is on the key production activities of design,

construct, operations and maintenance. A total of 18 key activities, each representing

distinct technology and/or distinct knowledge and skills set(s), were identified in this

road case study.

Design activities are categorised into the design of construction, and the design

or planning of maintenance of the project. In terms of construction design, this

includes road civil engineering design and traffic engineering design. The road civil

engineering design includes structural and civil design, geometric design, drainage,

pavement, and intersections and interchanges of the road structure. Traffic

engineering design includes the design of Intelligent Transportation Systems (ITS),

traffic management and controllers, access management, speed design, and the

performance specification of operations for traffic signals, lighting, ITS and traffic

operations. The design activities include schematic design, developed design, and

contract documentation design; however, it excludes performance specifications as

these are readily available in standard guides and manuals.

As the aim of the model is to identify a configuration of key procurement

dimension most likely to set the project on a path to deliver superior VfM across the

whole-of-life of the asset, the design and planning of maintenance of the road

project, according to the required level of quality, is also included in the activity

analysis. This includes planning of inspections and data collection, and planning of

implementation of reactive, routine and preventive maintenance for all elements,

such as roads, pavement, furniture, drainage, ITS, traffic signals, and lighting.

With regards to construction activities, the 12 key activities with distinct

technology and/or distinct knowledge or skill sets include: 1. bulk earthworks; 2.

traffic management; 3. supply and installation of drainage and culverts; 4.

construction of retaining walls; 5. protective treatments and works; 6. plant-mixed

stabilised pavements; 7. sprayed bituminous surfacing and cover aggregate; 8.

asphalt, including its preparation, surfacing and grading; 9. roadside structures

(mainly precast concrete barriers); 10. noise barriers; 11. pavement or line markings;


and 12. landscaping. Electrical works related to the project include: electrical pits,

conduits and fittings, road lightings, and ITS ducts and pits.

The operational activities of this road project involve traffic operations,

including: traffic signals, ITS, ramp metering, variable speed limits, lane

management, camera surveillance, speed cameras, and variable road signs. These

activities all belong to a similar skill set and can be carried out by a single person

trained in all the operational-related activities.

Lastly, the maintenance activities of the road project are categorised into three

types; namely, routine, preventive and emergency. ‘Routine’ maintenance refers to

the reactive maintenance to roads or pavement that does not require a speedy

response, such as the repair of: potholes, localised areas, furniture, drainage, ITS,

traffic signals, and lighting. While ‘Preventive’ maintenance is similar to reactive

routine maintenance, it is planned as part of a program for bulk maintenance of the

network; for example, bitumen reseals, gravel re-sheeting, and asphalt overlay.

‘Emergency’ maintenance is a type of reactive maintenance that requires a very

quick response time and which, by definition, involves little or no planning, as is the

case in the aftermath of a major flood or natural disaster. Maintenance in the form of

rehabilitation has been excluded as a key activity, as it can be considered as another

road project in its own right.

5.3.2 Stage 1/ Task B: Make-or-buy analysis

Having identified the key activities for Case Study #R1, a project manager in

the government road agency that delivered this project, then answered the questions

for each key activity for this case study. In answering the questions, the project

manager was mindful that their responses should reflect the competencies and

capabilities of the government and the market at the time the actual procurement

decision was made (in 2009). The answers to each question (on a 7-point scale) for

each variable in the integrative framework (shown earlier in Table 3.1) were

designed to generate a pattern comprising different combinations of symbols (-, 0

and +) for each activity. The empirical or actual patterns of TCE and RBT variables

were then matched with the theoretical or predicted patterns in the integrative

framework, and the activity was assigned to one of the levels in which there was the

closest match between the empirical and theoretical patterns. As explained


previously, Levels 1 to 4a represent activities that are most efficiently internalised,

and Levels 4b to 7 represent activities that are most efficiently externalised.

Table 5.6 (below) provides a summary of the responses to the questionnaire

instrument, and the patterns and corresponding activity levels with respect to the

design activities of the Case Study #R1.

Table 5.6 Make-or-buy analysis for design activities for #R1

No Key activity Capacity Rarity Costly-to-imitate

Asset specificity

Uncertainty Frequency Pattern Level

Design 1 Road civil engineering

design 3 1,4 1,1 1,2 1,2 2 (-, 0, 0, 0 to

+, 0, 0) 5

2 Traffic engineering design

3 2,3 1,1 1,1 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

3 Plan for inspections and data collection for 2 types of maintenance for all elements

6 7,7 4,4 6,6 1,1 6 (++, + to +++, 0, 0 to ++, 0 to ++, + to +++)

2

In terms of the design of the civil engineering and traffic engineering

components of Case Study #R1 in Activities 1 and 2, the score of ‘3’ on the capacity

variable indicates that it is not within the comfortable limits of the government

agency to provide the services without additional investments in human resources.

(This is despite the fact that the state government agency had already gained some

competence in the supervision of other projects and had a certain level of internal

capacity and technical capability). This matches with the response to the frequency

variable which indicates that the project is not typical, and is moderately sized at

$10-100 million. Furthermore, all civil and traffic engineering firms have the

required civil and traffic engineering knowledge required for Case Study #R1, and

there were many firms available to provide these services in the 2009 market, as

indicated by the low score on ‘rarity’, and the corresponding symbol ‘0’. The low

score of ‘1’ on the imitability variable also demonstrates that it is extremely

straightforward for the associated market firm to develop the required knowledge and

skill to deliver the activity.

In summary, the scores and symbols generate a Level 5 pattern for Activities 1

and 2, and indicate that the market is organisationally superior to the government in

terms of providing a less costly and/or more speedy response. This is because of its

position in the industry which enables it to aggregate higher levels of work and to

generate work flow, while at the same time increasing efficiencies and making lesser

mistakes when carrying out similar activities.


Activity 3 refers to the actual planning and the specifications of the quality of

the road maintenance, and specifically comprises of: a plan for inspections and data

collection; a plan for reactive (routine) maintenance; and a plan for preventive

(programmed) maintenance. Activity 3 is a Type A maintenance activity (as

previously discussed in Section 4.4.6), and is a Level 2 activity.

Table 5.7 (below) is a summary of the results of the make-or-buy analysis of

construction activities of Case Study #R1.

Table 5.7 Make-or-buy analysis for construction activities for #R1

No Key activity Capacity Rarity Costly-to-

imitate

Asset specificity


Construction 4 Earthworks 3 1,1 1,1 1,1 1,3 2 (-, 0, 0, 0 to

+, 0, 0) 5

5 Traffic management 3 1,1 1,1 1,1 1,3 2 (-, 0, 0, 0 to +, 0, 0)

5

6 Drainage and culverts 3 1,1 1,1 1,1 1,2 2 (-, 0, 0, 0 to +, 0, 0)

5

7 Retaining walls and protective treatments

3 3,1 1,1 1,1 1,3 2 (-, 0, 0, 0 to +, 0, 0)

5

8 Plant-mixed stabilised pavements

3 1,3 1,1 1,4 3,2 2 (-, 0, 0, 0 to +, 0, 0)

5

9 Sprayed bituminous surfacing and cover aggregate

3 1,1 1,1 1,1 2,2 2 (-, 0, 0, 0 to +, 0, 0)

5

10 Asphalt, including preparation, surfacing and grading

3 1,1 1,1 1,2 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

11 Roadside structures, mainly precast concrete barriers

3 1,1 1,1 1,3 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

12 Noise barrier 3 1,1 1,1 1,3 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

13 Pavement/line markings and markers; and anti-graffiti protection

3 1,2 1,1 1,2 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

14 Electrical pits, conduits, fittings, including road lightings and ITS ducts and pits

3 1,4 1,1 1,3 2,3 2 (-, 0, 0, 0 to +, 0, 0)

5

15 Landscaping, including. road furniture

3 1,1 1,1 1,1 1,1 2 (-, 0, 0, 0 to +, 0, 0)

5

Activities 4 to 15 represent all the key construction activities and were assigned

to Level 5 based on the responses and patterns generated. Similar to civil and traffic

engineering design activities 1 and 2, all the RBT and TCE variables generate a low

score ranging from 1 to 3. In terms of capacity and, mindful that procurement

decision was made in 2009, the government was constrained in terms of providing

these activities in a project that was non-typical and low frequency, in comparison to

the portfolio of projects in the government agency.


The knowledge and skills required in these activities are widely available and

there is a plentiful supply of these firms in the market. These activities are also not

critical, and there are some tolerances in response time and, therefore, lesser extent

of hold-up to the government in terms of quick response. Furthermore, these

activities are more predictable per se, with weather conditions being the main source

of uncertainty. More importantly, the market has the advantage in terms of

aggregating a greater volume of work to generate workflow; this, in turn, leads to

more competence in organising, and making lesser mistakes in planning,

coordination, procurement and management of supply chain, and a decrease in the

bad behaviour of suppliers.

As explained in Section 3.2.3a, Level 3 and Level 5 are concerned with

supervision; that is, with make-or-buy in obtaining resources that are readily

available, and not dependent on critical workload and the organisation of these

resources. This helps to explain the relationship between the issue of lack of

capacity, either in relation to budget and/or supervision, and margins of workload. It

might not be cost efficient for these construction activities to be internalised at Level

3 when demand drops, and/or when there are more mistakes through overstretched

supervision. For this reason, all of the construction activities in Case Study #R1 were

assigned Level 5; this means that the market could provide the activity for a

relatively lower cost and employ more efficient methods of delivery than the

government, by virtue of its position in the industry which allows it to aggregate

higher levels of work and generate work flow.

Table 5.8 (below) gives the summary of the responses to questions regarding

the implementation of operations and maintenance activities.


Table 5.8 Make-or-buy analysis for operations and maintenance activities for #R1

No Key activity Capacity Rarity Costly-to-imitate

Asset specificity


Implementation of Operations 16 Intelligent Transport

Systems (ITS) and traffic operations

5 2,2 1,1 1,1 1,1 6 (+, 0, 0, 0 to +, 0, + to +++)

3

Implementation of Maintenance 17 Inspections & data

collection, including routine and programmed maintenance

4 1,1 1,1 6,6 6,6 6 (-/+, 0, 0, + to +++, + to +++, + to +++ )

4a

18 Reactive (emergency) maintenance

4 1,1 1,1 7,7 7,7 6 (-/+, 0, 0, + to +++, + to +++, + to +++ )

4a

Activity 16 refers to the operations of the road in terms of ITS and traffic

operations, and is assigned as a Level 3 activity. Given the advantages of economies

of scale and knowledge of the network of roads, the government is organisationally

positioned to provide cheaper and/or quicker responses than the market; however,

might not be better than those provided by the market in terms of quality.

Both Activity 17 and Activity 18 refer to the actual maintenance of the road, as

well as inspections and data collection. Activity 17 refers to the actual

implementation of reactive (routine) maintenance and implementation of preventive

(programmed) maintenance to roads (or pavement) and furniture, while Activity 18

refers to the implementation of maintenance with a reactive (or emergency)

component; that is, a critical requirement of quick response time. As earlier

discussed in Section 4.4.6, Activity 17 is a Type B maintenance activity, and Activity

18 is a Type C maintenance activity; therefore, both activities are Level 4a activities.

As previously noted, the design and planning of maintenance (Activity 3) is a

Type A maintenance activity and is internalised, along with the implementation of

maintenance (Activities 17 and 18) and the implementation of operations (Activity

16). The remaining design and construction activities of Case Study #R1 are

externalised. Table 5.9 (below) is a summary list of the internalised and externalised

activities.


Table 5.9 Summary of make-or-buy analysis for #R1 No Key internalised activity Design 3 Plan for inspections and data collection; plan for reactive (routine) maintenance and plan for preventive

(programmed) maintenance to all elements (Level 2) Operations 16 Intelligent Transport Systems and traffic operations (Level 3) Maintenance 17 Inspections and data collection for reactive routine and preventive maintenance; and implementation of

reactive (routine) maintenance and implementation of preventive (programmed) maintenance to roads or pavement and furniture (Level 4a)

18 Implementation of reactive (emergency) maintenance (Level 4a) Key externalised activity Design 1 Road civil engineering design (Level 5) 2 Traffic engineering design (Level 5) Construction 4 Earthworks (Level 5) 5 Traffic management (Level 5) 6 Drainage and culverts (Level 5) 7 Retaining walls and protective treatments (Level 5) 8 Plant-mixed stabilised pavements (Level 5) 9 Sprayed bituminous surfacing and cover aggregate (Level 5) 10 Asphalt, including preparation, surfacing and grading (Level 5) 11 Roadside structures, mainly precast concrete barriers (Level 5) 12 Noise barrier (Level 5) 13 Pavement/line markings and markers; and anti-graffiti protection (Level 5) 14 Electrical pits, conduits, fittings, including road lightings and ITS ducts and pits (Level 5) 15 Landscaping, including road furniture (Level 5)

5.3.3 Stage 1/Task C: Market analysis

As already mentioned in Section 3.2.4, this market analysis task used SCP or

market structure analysis to perform a check on each activity to corroborate the

predicted activity levels identified in make-or-buy analysis. Based on the latter, there

are no Level 7 activities arising out of scale of the key activity’s work in the project.

All the activities are identified as ‘Level 5’, and the corresponding market structure

is closest to the stereotypical perfect competition; that is, numerous firms capable of

delivery of their particular activity in the market.

An on-line search was carried out to determine the potential number of market

firms supplying each activity to Case Study #R1 within the region or state in order to

determine if this number was in line with the level assigned to the activity through

the (above) process of matching empirical patterns with theoretical patterns in the

integrative framework. That is:

1. If the key activity is assigned as a Level 4b or Level 5 activity, then the expected number of firms in the market is expected to be greater than eight.

2. If the key activity is assigned a Level 6 activity, then the expected number of firms in the market is expected to be between four and eight.


3. If the activity is assigned a Level 7 activity, then the expected number of firms in the market is expected to be between one and three.

Table 5.10 (below) is the summary of the expected number and the

corroborated number of firms on the externalised key activities.

Table 5.10 Market structure analysis of key activities for #R1 No Key activity Level Expected

number Corroboration number

1 Road civil engineering design

5 >8 33 approved firms in highway construction category in State road department approved list

2 Traffic engineering design 5 >8 35 approved traffic engineering firms approved by State road department

4 Earthworks 5 >8 At least 112 civil contractors capable of earthworks, pavement and drainage work across small to large projects in State road department approved list; 290 earthmoving contractors in Yellow Pages®

5 Traffic management 5 >8 58 Traffic management firms in State road department approved list

6 Drainage and culverts 5 >8 around 95 medium-sized civil contractors capable of drainage and culverts; and 17 precast culverts suppliers in State road department approved list

7 Retaining walls and protective treatments

5 >8 183 Retaining wall suppliers and subcontractors to region in State from search in Yellow Pages®

8 Plant-mixed stabilised pavements

5 >8 Approximately 95 medium-sized civil contractors in State road department approved list

9 Sprayed bituminous surfacing and cover aggregate

5 >8 Approximately around 95 medium-sized civil contractors with bituminous surfacing capability in State road department approved list; 50 sprayed bituminous surfacing contractors in State in Yellow Pages®

10 Asphalt, including preparation, surfacing and grading

5 >8 Approximately 95 medium-sized civil contractors with asphalt and grading capability according to State road department approved list; 24 asphalt and bitumen paving contractors to region in Yellow Pages®

11 Roadside structures (precast concrete barriers)

5 >8 8 precast concrete suppliers for precast concrete road barriers to region in State road department approved supplier list

12 Noise barrier 5 >8 17 Precast concrete suppliers in State road department approved list

13 line markings and markers 5 >8 17 Linemarkers to region in State in Yellow Pages® 14 Electrical pits, conduits,

fittings, lightings 5 >8 14 steel structure suppliers, 9 road safety barrier systems and

device suppliers to region in State road department approved supplier list

15 Landscaping 5 >8 225 Landscape contractors to region in State in Yellow Pages®

The State road department approved list of design consultants was available

online, and was used to determine the number of road civil engineering and traffic

engineering consultant firms with design capabilities similar to those in Case Study

#R1. There are 92 approved firms across civil engineering design, traffic

engineering, and transport planning in the market in the entire State. Some firms are

multidisciplinary and can provide one or more of the engineering services. In terms

of civil engineering design capability of highway construction, including widening

and overlay of existing roads, there are approximately 33 consultant firms on the pre-

qualification list. As for traffic engineering capability, the state department has

approved 35 consultant engineering firms operating in the region. From the above

analysis, the market for both civil and traffic engineering design services associated


with projects similar to Case Study #R1 appears to be closest to the stereotypical

perfect competition.

With regards to roadwork activities, the department has listed approximately

95 civil contractors with financial levels capable of projects similar to Case Study

#R1 that are capable of earthworks, pavement and drainage work across small to big

projects. The activities include Activity 4 – earthworks, Activity 6 – drainage and

culverts, Activity 8 – plant-mixed stabilised pavements, Activity 9 – sprayed

bituminous surfacing and cover aggregate, and Activity 10 – Asphalt (including

preparation, surfacing and grading).

The supply of these activities was further examined. For Activity 4 –

earthworks, a keyword search for ‘excavating and earth-moving contractors’ listed in

Yellow Pages® online, revealed approximately 290 excavating and earth-moving

contractors and 60 excavating and earthmoving equipment suppliers listed in the

region and in the State. In terms of the supply of the precast culverts and drainage,

the state road department has 17 approved suppliers in the region. For Activity 9 –

sprayed bituminous surfacing, approximately 50 firms were generated from the

keyword search for ‘bitumen spraying’ in the Yellow Pages® listed in the region,

and 12 firms are further listed under the ‘bitumen spraying’ category. This could

indicate that these 12 firms are likely to own the machinery and equipment required

for bituminous surfacing. As for Activity 10 – asphalt surfacing, and grading, the

keyword search using ‘asphalt and bitumen paving contractors’ to the region in the

state in the Yellow Pages® generated 58 firms, and a further sub-category under the

heading of ‘asphalt and bitumen paving contractors’ listed 24 firms.

For Activity 6 – there are 17 suppliers of precast culverts and drainage,

roadside safety structures, and noise barriers in the state road department approved

suppliers list. The state department has also approved 58 traffic management

companies, 14 steel structure suppliers, and nine suppliers of road safety barrier

systems and other related road safety devices. Lastly, for landscaping works, there

are 225 companies listed under ‘Landscape contractors’ in the Yellow Pages®.

Consistent with the activity levels assigned, the numerous numbers of firms

across the activities indicates a market structure that is closest to the stereotypical

perfect competition market structure for Activities 1 to 15; this corroborates the level

assigned to these activities in the integrative framework.


5.3.4 Stage 2: Bundling analysis

In this task, the focus is now on second-order issues concerning the

organisation and management across key activities. As there are no troublesome

residual Level 7 or Level 4b activities in the make-or-buy analysis to consider, the

model then goes straight to considering opportunities to bundle key externalised

activities into main activities, and main activities into major activities. For each main

activity bundle, the model prompts the user to consider whether bundling is likely to

generate a Level 7. The key activities of similar nature are bundled or grouped into

two main bundles of activities, namely ‘design’ and ‘construction’. The activities in

each main activity are summarised in Table 5.11 (below). As seen in the table, the

main design bundle comprises of road civil design and traffic engineering design key

activities, and is unlikely to develop into a Level 7 activity if bundled together.

The questionnaire survey of civil and building contractors determined that

there are six civil contractors capable of Construct-only contracts (of at least $50

million) in the state where this case study was situated. Of these six contractors, five

indicated significant spare capacity and spare capacity, and were able to tender for

the road project Case Study #R1 in mid-2010. Factoring the number to 100 percent

gives an estimate of 15 contractors who might have been available to bid for the

project; therefore, bundling the construction activities into a main bundle is unlikely

to result in a Level 7 bundle of main activity.

Table 5.11 Main design activity bundle and main construction activity bundle for #R1

Bundles Main design activity

1 Road civil engineering design (Level 5) 2 Traffic engineering design (Level 5)

Main construction activity

4 Earthworks (Level 5)

5 Traffic management (Level 5) 6 Drainage and culverts (Level 5)

7 Retaining walls and protective treatments (Level 5)

8 Plant-mixed stabilised pavements (Level 5) 9 Sprayed bituminous surfacing and cover aggregate (Level 5)

10 Asphalt, including preparation, surfacing and grading (Level 5) 11 Roadside structures, mainly precast concrete barriers (Level 5)

12 Noise barrier (Level 5)

13 Pavement/line markings and markers; and anti-graffiti protection (Level 5) 14 Electrical pits, conduits, fittings, including road lightings and ITS ducts and pits (Level 5)

15 Landscaping, including road furniture (Level 5)


Hence, there are no discernible viable DCM, DCO or DCOM major activity

bundles to be considered, either as PPPs or as contracts. The model then defaults to a

separate procurement approach to each of the two main externalised activities, in

terms of two separate contracts organised by the government; that is, direct to

government or via a government agency or project management (PM) consultant.

In summary, the bundles of activities are translated into two contracts, one

contract for the civil and traffic engineering design, and another for the entire

construction and delivery of the road project:

• Contract #1: Design of civil engineering and traffic engineering of the road to the

same engineering consultant (2 activities)

• Contract #2: Construction of the entire road project to one main contractor (12

activities)

5.3.5 Stage 3: Exchange relationship analysis

Finally, measures of the TCE variables only for Contract#1 and Contract#2 are

taken, mindful of the inclusion of the planning and coordination components across

key activities, with an anticipated typical market firm at the head of each contract or

counterparty to government. The theoretical or predicted nature of the exchange

relationship can be observed by matching the empirical pattern to the patterns

illustrated in Figure 3.2 on exchange relationship. The summary of the scores for the

two main activities or contracts are given in Table 5.12 (below).

Table 5.12 Exchange relationship analysis for #R1

no Contract Asset specificity

Uncertainty Frequency Pattern Nature of exchange

1 Contract 1# – Designer responsible for planning; coordination and overall delivery of civil engineering design (Level 5) and traffic engineering design

1,1 1,1 2 (0,0,0) Discrete and efficient

2 Contract 2# – Main contractor responsible for planning; coordination and overall delivery and construction of the remaining Level 5 activities

1,3 1,1 2 (0,0,0) Discrete and extremely efficient

For Contract 1#, the patterns indicate that the exchange between state

government and the market firm for both the design main activity and the

construction main activity should be discrete and efficient or arms-length in nature.

Based on Table 3.3, there is better goal alignment between principal and agent when

this discrete exchange connects to an outcome-based exchange or contract; that is, a

fixed-price contract with high power incentive, where the agent assumes the risks to


completion. Risks include: agency costs, specification of outcomes, verification of

outcomes, risk premium, suitability for information asymmetry, and outcome

certainty. Hence, the principal-agent characteristics of this discrete exchange that

connects to an outcome-based exchange are as follows:

a. Fixed-price b. High power incentive c. Agent's risk to completion

• agency costs

• specification of outcomes • verification of outcomes

• risk premium • suitability for information asymmetry

• outcome certainty • better goal alignment

For Contract #2, which is related to the planning, coordination and overall

delivery and construction of the remaining activities, the patterns indicate an

extremely efficient discrete exchange, using a standard or neo-classical contractual

agreement. The principal-agent characteristics of this discrete exchange that connect

to an outcome-based exchange are as follows:


• agency costs

• specification of outcomes • verification of outcomes • risk premium

• suitability for information asymmetry • outcome certainty

• better goal alignment

5.3.6 Procurement strategy

Based on the recommendation of the arms-length nature of exchange, the

procurement model considers the most efficient approach to procuring the design and

construction of Road Case Study #R1, using separate standard fixed-price, lump-sum

contracts. The overall procurement strategy informed by the procurement model for

Case Study #R1 is as follows:

1. Internalise the design, plan and implementation of maintenance


2. Internalise the implementation of operations

3. Engage an engineering design consultant for the planning, coordination and

overall delivery of civil engineering and traffic engineering design, using a

single, fixed-price lump-sum contract

4. Engage a main contractor for the planning, coordination and overall delivery of

the construction, using a single, fixed-price lump-sum contract

5.3.7 Hypothesis testing (including theoretical competition)

The above theoretical procurement strategy is extremely close to the actual

practice in terms of internalisation and externalisation of each activity, the level of

bundling of key, main, or major bundles of activities, and the nature of exchange

relationship of each contract. In practice, the design and construct activities were

procured directly and separately by the government, using fixed-price lump-sum

contracts, while internalising the operations and maintenance of Case Study #R1; this

is consistent with the outcomes of the procurement model. The first-order decision-

making model, in other words, very strongly supports Case Study #R1 as being

efficiently procured in terms of VfM in whole-life terms.

The results of case study analysis of Case Study #R1 match with Scenario 1 of

the refutability procedures in Section 3.3.5; that is, when the actual level of

competition is within the optimal level of 5-8 EoI, then the actual procurement is

expected to be similar to the theoretical procurement. This result supports the

hypothesis, and provides very strong support for the validity and reliability of the

first-order procurement decision-making model. However, this validity and

reliability of the model is subject to the outcomes of all four cases.

The next Case Study #R2, which is much larger and more complex than Case

Study #R1, is concerned with the development of a section of a busway, on-grade

pavement, and driven and cut-and-cover tunnels.

5.4 Road Case Study #R2 (Theoretical procurement)

Road Case Study #R2, on the other hand, had only two EoI, which is much

lower than the optimal range of EoI. Road Case Study #R2 comprised a busway,

including bridges and ramps, driven tunnel, including cut-and-cover tunnels, and two

bus stations. The excavation of the driven tunnel was the most challenging,

complicated by the sensitive heritage-listed building with shallow foundations above,


varying geological conditions along the alignment of the tunnel. This project was

procured using an alliance approach; that is, the government collaborates with the

proponent team, which comprised of the designer and main contractor. After

subsequent discussions, an agreement on the commercial parameters and contractual

framework was reached, and finalised in the interim alliance agreement. During the

project development or pre-construction phase, the client and the proponent team

undertook activities jointly as an integrated alliance team to develop and agree on

target outturn costs and other performance targets. With respect to payment terms,

the proponent team was guaranteed reimbursement of their direct project costs and

corporate project overheads in an open-book arrangement, subject to the agreed gain-

share or pain-share model or regime.

The following presents the outcomes of the analyses of applying the

procurement model to Case Study #R2.


Road Case Study #R2 is much larger in scope and more complex than Case

Study #R1, with the added complexities and uncertainties of the driven tunnel under

an existing building, a cut-and-cover tunnel under a high platform and railway lines,

and connection to an existing busway. The activities for the design, operations and

maintenance are similar to those for Case Study #R1, while the construction

activities are significantly larger in scope and size. Construction activities of similar

nature were categorised into the different sections of the project. A total of 61 key

activities were identified in Road Case Study #R2. The design activities of Case

Study #R2 is similar to that of Case Study #R1 (though larger in scope); that is, the

design of new construction works, and the design of performance specifications for

maintenance.

The design of construction works comprises of six key activities, namely: 1)

civil and structural engineering design (excluding the driven tunnel, and including:

rail track relocation, storm water drainage, utilities relocations, and busway station

design); 2) civil and structural engineering design to the driven tunnel; 3) traffic

engineering design; 4) mechanical and electrical engineering design; 5) fire safety

design for tunnels; and 6) landscaping and urban finishes design.


The design of performance specifications for maintenance refers to: the plan

for inspections and data collection for routine and planned programmed

maintenance; the plan for reactive routine maintenance; and the plan for preventive

maintenance to specialist linings, mechanical and electrical and fire elements in the

driven tunnel, and to all other elements in the project.

Forty-nine key construction activities were categorised into the various

sections of the road project due to the differences in the nature of the complexities

associated with an at-grade road and a cut-and-cover tunnel; namely: cut-and-cover

tunnels, driven tunnel, at-grade road, bridge, ramps, median, walkway and bikeway

structures, and bus stations. The road project was analysed at an activity or first-order

level in terms of its different sections. Even though similar activities, say bulk

excavation, can occur in both a cut-and-cover tunnel and at-grade road, bulk

excavation for at-grade road is considered separately as a different activity from bulk

excavation for a cut-and-cover tunnel. Finally, the operations and maintenance

activities are similar to those in Case Study #R1.

5.4.2 Stage 1/Task B: Make-or-buy analysis

Due to the differing nature of the complexities associated with the different

sections in Case Study #R2, the analysis of the activity levels are grouped into:

design activities; construction of cut-and-cover tunnels; construction of driven

tunnel; construction of road at-grade; construction of bridge, ramps, median,

walkway and bikeway structures; construction of bus stations; other construction

activities; implementation of operations; and implementation of maintenance.

Table 5.13 (below) summarises the responses to the case study questionnaire

for design activities in Case Study #R2, and includes the corresponding pattern and

activity level assigned to each activity.


Table 5.13 Make-or-buy analysis for design activities for #R2 No Activity Capacity Rarity Costly-to-

imitate Asset specificity


Design 1 Civil engineering design and

structural engineering design 3 2,3 2,2 2,3 3,3 4 (-, 0, 0, 0 to +,

0, 0) 5

2 Civil engineering design and structural engineering design for the driven tunnel

1 7,7 7,7 5,7 5,5 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

3 Traffic engineering design 3 2,3 3,1 1,1 2,1 4 (-, 0, 0, 0 to +, 0, 0)

5

4 Mechanical and electrical engineering design, including air quality and ventilation

3 4,4 3,3 3,3 3,2 4 (-, 0, 0, 0 to +, 0, 0)

5

5 Fire safety design for tunnels 1 6,6 6,6 3,3 2,2 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

6 Landscaping and urban finishes design

3 2,2 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0)

5

7 Planning for inspections and data collection for reactive and programmed maintenance to all elements in project (except driven tunnel); plan for reactive (routine) maintenance to roads/pavement and furniture, drainage, ITS, traffic signals and lighting; and plan for preventive (programmed) maintenance to roads/pavement and furniture

6 7,7 4,4 6,6 1,1 6 (++, + to +++, 0, 0 to ++, 0 to ++, + to +++)

2

8 Planning for inspections and data collection for reactive and programmed maintenance to specialist linings, mechanical and electrical and fire elements in driven tunnel; and plan for reactive (routine) maintenance and plan for preventive (programmed) maintenance to specialist linings, membranes, mechanical, electrical and fire elements in driven tunnel

1 7,7 7,7 5,5 5,5 3 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

The activities related to the driven tunnel design – civil engineering design and

structural engineering design in Activity 2, fire safety design in Activity 5, and the

plan and specifications for maintenance of specialist items in Activity 8 in the driven

tunnel (where the market ability is technically superior to that of the government) –

are assigned as Level 7. This is mainly due to the expert knowledge required for

tunnel design, and government does not have the adequate internal capacity. An

overseas tunnel design specialist was engaged to contribute to the design aspects of

the driven tunnel. For this reason, the scores for ‘rarity’, ‘costly-to-imitate’, ‘asset

specificity’, and ‘uncertainty’ are high, ranging from 5 to 7.

In terms of the design of the busway, although the state government has a

certain level of internal capacity and technical capability, the contractor or market is


organisationally better than the government to provide a cheaper and/or quicker

response because of its ability to aggregate a greater amount of work, increase

efficiencies, and make less mistakes in carrying out similar activities. Thus, activities

excluding the driven tunnel – that is, civil and structural engineering design, traffic

engineering design, mechanical and electrical engineering design, landscaping and

urban finishes design activities of the project – are assigned Level 5. Finally, the

design or planning for maintenance of the busway (excluding the driven tunnel) is a

Type A maintenance activity and is assigned Level 2, for the same network

advantage reasons given for Case Study #R1.

Table 5.14 (below) provides a summary of the construction of cut-and-cover

tunnels at both ends of the driven tunnel, and under the middle road, dual gauge and

suburban rail lines, and the existing freeway.

Table 5.14 Make-or-buy analysis of construction activities in cut-and-cover tunnels for #R2

No Activity Capacity Rarity Costly-to-imitate

Asset specificity


Cut and cover tunnels (portals to driven tunnel; under rail lines; connections to existing busway and under existing freeway) 9 Relocation of existing public

utility plant 7 7,7 7,7 4,7 5,5 7 (+++, + to +++, + to

+++, 0 to +++, 0 to +++, + to +++)

1

10 Removal works – fencing, pavement, walls, sewer/water lines

3 1,1 1,1 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

11 Traffic management 3 1,1 1,1 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

12 Bored piles 4 2,3 2,2 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

13 Excavate and shotcrete to bored piles

4 3,3 1,2 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

14 Earthworks including: access; bulk excavation; strip topsoil; and batter stabilisation; backfill to top of slab

4 2,2 2,2 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

15 Structural works – concrete cast insitu – column, headstock/ abutments, planks, topping slab

4 2,2 1,1 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

16 Precast concrete – concrete barriers and kerbs

3 2,2 2,2 2,2 2,2 4 (-, 0, 0, 0 to +, 0, 0) 5

17 Waterproofing 4 2,2 2,2 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

18 Drainage 4 1,1 1,1 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

19 Pavement (a layer of no fines concrete and a paving layer)

4 3,3 3,3 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

20 Modifications to existing bridge and footpath beams and slab

3 3,3 2,2 2,1 2,3 4 (-, 0, 0, 0 to +, 0, 0) 5

21 Demolition works, including building and bridge structures, walls and pavements

3 1,1 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5

22 Realignment of rail track 7 7,7 7,7 4,7 5,5 7 (+++, + to +++, + to +++, 0 to +++, 0 to +++, + to +++)

1


Both Activity 9 and Activity 22 can only be carried out by government

agencies or subsidiaries, which are the only authorities that possess the specialised

monopolised knowledge to execute these activities. Activity 9 refers to the relocation

of public utility plant that requires expert knowledge of this plant and can only be

carried out by the operator who installed it. Activity 22 refers to the realignment of

railway tracks, which can only be carried out by the related public railway provider

that has the internal capability and competence to realign its tracks and to coordinate

the scheduling of train times while carrying out this work. Therefore, Activities 9 and

22 are assigned Level 1, with corresponding high scores of 7 for value, rarity and

costly-to-imitate variables. In other words, market firms are not able to match

government’s capability and competence in relation to these activities.

The remaining cut-and-cover construction activities, Activities 10 to 21, are

externalised. Although the government has a moderate level of internal capability

and capacity to carry out these activities (as indicated by their moderate scores of

between 3 and 4), carrying them out internally would create constraints, such as

budgeting and number of staff, and the level of resources and investment would need

to be increased. In addition, the frequency variable was scored as a non-typical and

extremely large project, and this corroborates the score of the value or capacity

variable.

The uncertainties in relation to the start and finish times caused by the

uncertainty with the progress of the driven tunnel, the constraints in relation to

coordination with rail operators, and the staging of a constrained site in connection

with an existing busway, are the main contributing factors in creating high levels of

uncertainty and complexity in the cut-and-cover tunnel. The willingness of the firm

supplying the activity to make the required progress to suit the project timeline and

weather conditions can critically affect that timeline and hold-up government. Thus,

these activities exhibit higher scores (ranging from 5 to 7) on asset specificity and

uncertainty variables, indicating a higher potential for negative opportunistic

behaviour by the contractor. Activities 10 to 15, and 17 to 19 – removal works,

traffic management, bored piles, shotcrete to bored piles, earthworks, concreting,

waterproofing, drainage, and pavement works – are assigned with Level 4b. For

Activity 16 (precast concrete), and Activities 20 and 21 (modification and demolition

works to existing structures), the scores reflect a lower level of asset specificity and


uncertainty, and are less likely to hold-up the government agency in terms of project

progress and timeline. They are, therefore, Level 5 activities.

The following Table 5.15 lists the responses to the questions pertaining to the

activities involved in the construction of the driven tunnel.

Table 5.15 Make-or-buy analysis for construction activities in driven tunnel for #R2


Asset specificity


Driven tunnel 23 Excavation in tunnel

and shotcrete 4 2,4 4,2 6,7 6,7 4 (-/+, 0, 0, + to

+++, + to +++, 0)

4b

24 Waterproofing 4 2,4 4,2 7,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

25 Structural works (including fibre reinforced polymer (FRP), headstocks, planks, FRP roof slab and topping slab)

4 2,4 2,2 6,6 6,6 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

26 Precast concrete – barriers, kerbs and wall

3 1,1 1,1 1,1 3,3 4 (-, 0, 0, 0 to +, 0, 0)

5

27 Drainage 4 2,2 2,2 6,6 6,6 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

28 Trimming and backfill of main tunnel to finish (surface preparation)

4 1,1 1,1 6,6 6,6 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

29 Pavement (a layer of no fines concrete and a paving layer)

4 3,3 3,3 4,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

30 Ventilation fans 2 5,6 3,3 2,5 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

With the exception of Activity 26, Activities 23 to 29 for the driven tunnel are

assigned Level 4b. These activities have a higher score on asset specificity and

uncertainty variable, which is created by the uncertainty and unpredictability of

geological conditions. This is the case, for instance, in Activity 23 where the extent

of excavation works and quantity of shotcrete are highly dependent on the settlement

trends of the geological conditions. The contractor, designer and geotechnical

engineer would review the construction and geological data from the previous day

using face mapping, and make decisions regarding construction methodology and

strength of tunnel support on a daily basis. These activities related to the driven

tunnel exhibit a high level of uncertainty, mainly concerning the start and finish

times and/or durations in cut-and-cover tunnels created by the uncertainty of the

progress with driven tunnel, as well as cooperation from rail operators, and other

factors, such as, staged and site constraints regarding connections to the existing


freeway, and the criticality of the willingness of the firm supplying the activity to

make the required progress to suit the project timeline. Even if these difficulties are

overcome with strategic decisions and cooperation within timelines, progress of

activities can still be substantially inhibited by weather conditions.

The ventilation fans in Activity 30, are large fans that facilitate airflow in the

tunnel. They are relatively straightforward to install and, therefore, low in terms of

asset specificity and uncertainty. However, they can only be procured from a limited

number of specialised suppliers in the market; therefore, the activity has a high score

of 6 for rarity variable, and the overall pattern reflects Level 6.

Table 5.16 (below) gives the responses and corresponding patterns for at-grade

concrete pavement or road.

Table 5.16 Make-or-buy analysis of construction activities for at-grade road, bridge and ramp structures, and two bus stations (#R2)


Asset specificity


Road at-grade 31 Bulk excavation 3 1,1 1,1 1,1 3,4 4 (-, 0, 0, 0 to +, 0, 0) 5 32 Subgrade preparation 3 1,1 1,1 1,1 3,4 4 (-, 0, 0, 0 to +, 0, 0) 5 33 Drainage 3 1,1 1,1 1,1 2,3 4 (-, 0, 0, 0 to +, 0, 0) 5 34 Concrete pavement 3 3,3 3,3 3,4 2,3 4 (-, 0, 0, 0 to +, 0, 0) 5 35 Precast concrete –

barriers, kerbs 3 1,1 1,1 1,1 3,3 4 (-, 0, 0, 0 to +, 0, 0) 5

36 Retaining walls 3 2,2 1,1 2,2 2,2 4 (-, 0, 0, 0 to +, 0, 0) 5 37 Asphalt pavement 3 2,3 1,1 1,3 2,3 4 (-, 0, 0, 0 to +, 0, 0) 5 38 Re-alignment of

existing busway 3 2,3 1,1 1,3 2,3 4 (-, 0, 0, 0 to +, 0, 0) 5

39 Traffic management 3 1,1 1,1 1,1 1,3 4 (-, 0, 0, 0 to +, 0, 0) 5 Bridge, ramps, median, walkway and bikeway structures

40 Traffic management 3 1,1 1,1 1,1 1,3 4 (-, 0, 0, 0 to +, 0, 0) 5 41 Earthworks 3 1,1 1,1 1,1 3,4 4 (-, 0, 0, 0 to +, 0, 0) 42 Pile foundation 3 3,4 2,2 1,3 4,4 4 (-, 0, 0, 0 to +, 0, 0) 5 43 Concrete structural

works, 3 2,2 1,1 2,2 2,2 4 (-, 0, 0, 0 to +, 0, 0) 5

44 Precast concrete – barriers, kerbs

3 1,1 1,1 1,1 3,3 4 (-, 0, 0, 0 to +, 0, 0) 5

Bus stations 45 Water and stormwater

services 3 1,1 1,1 1,1 1,1 4 (-, 0, 0, 0 to +, 0, 0) 5

46 Electrical and communication services

3 3,3 1,1 1,2 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5

47 Pile foundations 3 3,4 2,2 1,3 4,4 4 (-, 0, 0, 0 to +, 0, 0) 5 48 Cast insitu concrete (lift

well, platforms and bus bays)

3 1,1 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5

49 Structural steelwork 3 1,1 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5 50 Roofing and drainage 3 1,1 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5 51 Cladding and louvres 3 1,1 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5 52 Glazing 3 1,3 1,1 1,3 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5 53 Mechanical services 3 2,2 1,1 1,1 1,2 4 (-, 0, 0, 0 to +, 0, 0) 5 54 Lift installation 2 1,6 1,1 1,5 1,2 4 (--, + to +++, 0, 0 to

++, 0 to ++, 0) 6


It can be seen in Table 5.16 (above), that Activities 31 to 53 have been

assigned Level 5, and only Activity 54 – lift installation – is a Level 6 activity, based

on the patterns generated by the responses. The state government agency has most of

the capabilities required to carry out the activities; however, it may be constrained in

performing activities of this size and extent. Hence, the score for value or capacity

tends towards the ‘beyond capacity’ option. The score of ‘1’ and corresponding

symbol (-) indicates a lack of capacity.

For Activity 54 – lift installation – the lack of capacity is greater, as indicated

by the symbol (--), corresponding to a lower score of ‘2’. The lack of capacity

indicated by scores of ‘2’ and ‘3’, matches with the frequency score ‘4’, which

represents a non-typical project of extremely large value (greater than $250 million).

A frequency score between ‘1’ and ‘4’ indicates a non-typical project, and is

represented by the symbol ‘0’. In terms of rarity across the Level 5 activities, there is

a plentiful supply of firms in the market that can perform these activities, and most of

these have the knowledge and skills required in the activity. Similarly, in terms of

imitability of the knowledge required across the Level 5 activities, guidelines and

specifications are readily available, and it is extremely straightforward for the firm to

develop the skills and knowledge required in the activity. Hence, the score for both

variables across these activities is either a ‘1’ or ‘2’, which corresponds to a ‘0’

symbol.

Finally, Activity 54 – lift installation – has similar scores and patterns to the

rest of the activities, with the exception of the rarity variable. This is because there

are only 4 to 6 contractors in the market who can undertake this more specialised

activity. Thus, the score is 6, and generates a ‘+++’ symbol. This activity also has a

higher score of ‘5’ on asset specificity, because the project timeline will be greatly

affected if the government needs to replace the contractor. Hence, this activity has an

overall pattern indicating a Level 6, where the market is technically more competent

than the government because of its firm-specific technical knowledge that cannot be

readily sourced. Similar to the patterns and reasoning for activities in Table 5.16

(above), Activities 55, 56 and 58 in Table 5.17 (below) have been assigned Level 5,

and Activity 57 has been assigned Level 6.


Table 5.17 Make-or-buy analysis of other construction activities in multiple parts of #R2


Asset specificity


Other (activities in more than of the above of parts of the project) 55 Line marking and signage 3 1,2 1,1 1,2 1,1 4 (-, 0, 0, 0 to

+, 0, 0) 5

56 Landscaping 3 1,1 1,1 1,1 1,1 4 (-, 0, 0, 0 to +, 0, 0)

5

57 M&E (power supply, lighting and fire services)

2 5,6 1,1 3,3 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

58 ITS 3 1,4 1,1 1,3 2,3 4 (-, 0, 0, 0 to +, 0, 0)

5

The following Table 5.18 gives the make-or-buy analysis of the activities

related to operations and maintenance of Case Study #R2.

Table 5.18 Make-or-buy analysis for operations and maintenance activities of #R2


Asset specificity


Implementation of operations

59 Intelligent Transport Systems and traffic operations, including traffic signals

5 2,2 1,1 1,1 1,1 4 (+, 0, 0, 0 to +, 0, + to +++)

3

Implementation of maintenance 60 Inspections and data

collection for reactive (routine) maintenance (inspections ad hoc) and

preventive (programmed) maintenance (inspection planned);

And implementation of reactive (routine)

maintenance to all parts of project (including driven tunnel): roads/pavement and furniture, drainage, ITS, traffic signals, and lighting;

preventive (programmed) maintenance to all parts of project (including driven tunnel): roads/pavement and furniture

4 1,1 1,1 6.6 6.6 4 (-/+, 0, 0, + to +++, + to +++, + to +++ )

4a

61 Inspections and data collection, and implementation of reactive (emergency) maintenance to all elements in project (except driven tunnel): roads/pavement and furniture, drainage, ITS, traffic signals, and lighting

4 1,1 1,1 7,7 7,7 4 (-/+, 0, 0, + to +++, + to +++, + to +++ )

4a

The scores for the RBT and TCE variables for operations and maintenance of

Case Study #R2 are similar to those for Case Study #R1. Similar to Activity 16 in

Case Study #R1, Activity 59 – ITS and traffic operations – is assigned Level 3. This


is because government has the advantage of knowledge and access to the road

network, and the capability to deploy this knowledge in tactical planning resulting in

efficiencies and economies of scale.

Activity 60 (similar to Activity 17 in Case Study #R1) refers to reactive and

preventive maintenance. It is a Type B maintenance activity, and is assigned Level

4a. This is because there is a greater uncertainty and asset specificity in the extent

and nature of maintenance, which gives rise to variations in a contract if it is locked

in for a certain period of time.

Activity 61 in emergency maintenance (similar to Activity 18 in Case Study

#R1) is a Type C maintenance activity, and has been assigned Level 4a due to the

requirement for a critical response (which led to a high score on asset specificity and

uncertainty). The benefits of internalising this activity are to prevent hold-up to a

speedy response, to have greater control over managing the response, and to thus

avoid costly variations.

Table 5.19 and Table 5.20 (below) summarise the list of internalised and

externalised activities respectively.

Table 5.19 Summary of internalised activities for #R2

No Key internalised activity Design

7 Planning for inspections and data collection for reactive and programmed maintenance to all elements in project (except driven tunnel); plan for reactive (routine) maintenance to roads/pavement and furniture, drainage, ITS, traffic signals and lighting; and plan for preventive (programmed) maintenance to roads/pavement and furniture (Level 2)

Construction Cut and cover tunnels

(portals to driven tunnel; under rail lines; connections to existing busway and under existing freeway) 9 Re-location of existing public utility plant (Level 1) 22 Re-alignment of rail track (Level 1) Operations 59 Intelligent Transport Systems and traffic operations, including traffic signals (Level 3) Implementation of maintenance 60 Inspections and data collection for reactive (routine) maintenance (inspections ad hoc) and

preventive (programmed) maintenance (inspection planned) (Level 4a) And implementation of • reactive (routine) maintenance to all parts of project (including driven tunnel): roads/pavement,

furniture, drainage, ITS, traffic signals, and lighting (Level 4a) • preventive (programmed) maintenance to all parts of project (including driven tunnel): roads/pavement

and furniture (Level 4a) 61 Inspections and data collection, and implementation of reactive (emergency) maintenance to all elements in

project (except driven tunnel): roads/pavement and furniture, drainage, ITS, traffic signals, and lighting (Level 4a)


Table 5.20 Summary of externalised activities for #R2 No Key externalised activity Design

1 Civil engineering design and structural engineering design (Level 5) 2 Civil engineering design and structural engineering design to the driven tunnel (Level 7) 3 Traffic engineering design (Level 5) 4 Mechanical and electrical engineering design, including air quality and ventilation (Level 5) 5 Fire safety design for tunnels (Level 7) 6 Landscaping and urban finishes design (Level 5) 8 Plan for inspections and data collection, and plan for preventive maintenance in driven tunnel (Level 7) Construction - Cut and cover tunnels 10 Removal works: fencing, pavement, walls, sewer/water lines (Level 4b) 11 Traffic management (Level 4b) 12 Bored piles (Level 4b) 13 Excavate and shotcrete to bored piles (Level 4b) 14 Earthworks including: access; bulk excavation; strip topsoil; batter stabilisation; backfill (Level 4b) 15 Structural works: concrete cast insitu – column, headstock/abutments, planks, topping slab (Level 4b) 16 Precast concrete: concrete barriers and kerbs (Level 5) 17 Waterproofing (Level 4b) 18 Drainage (Level 4b) 19 Pavement (a layer of no fines concrete and a paving layer) (Level 4b) 20 Modifications to existing bridge and footpath beams and slab (Level 5) 21 Demolition works, including building and bridge structures, walls and pavements (Level 5) Construction - Driven tunnel 23 Excavation in tunnel and shotcrete (Level 4b) 24 Waterproofing (Level 4b) 25 Structural works, including fibre reinforced polymer (FRP), headstocks, planks, FRP roof slab (Level 4b) 26 Precast concrete: barriers, kerbs and wall (Level 5) 27 Drainage (Level 4b) 28 Trimming and backfill of main tunnel to finish (surface preparation) (Level 4b) 29 Pavement (a layer of no fines concrete and a paving layer) (Level 4b) 30 Ventilation fans in the driven tunnel (Level 6) Construction - Road at grade 31 Bulk excavation (Level 5) 32 Subgrade preparation (Level 5) 33 Drainage (Level 5) 34 Concrete pavement (Level 5) 35 Precast concrete: barriers, kerbs (Level 5) 36 Retaining walls (Level 5) 37 Asphalt pavement (Level 5) 38 Re-alignment of existing busway (Level 5) 39 Traffic management (Level 5) Construction - Bridge, ramps, median, walkway and bikeway structures

40 Traffic management (Level 5) 41 Earthworks (Level 5) 42 Pile foundation (Level 5) 43 Structural works, including cast-in situ columns, headstock, abutments, planks, concrete pavement (Level 5) 44 Precast concrete: barriers, kerbs (Level 5) Construction - Bus stations 45 Water and stormwater services (Level 5) 46 Electrical and communication services (Level 5) 47 Pile foundations (Level 5) 48 Cast insitu concrete (lift well, platforms, and bus bays) (Level 5) 49 Structural steelwork (Level 5) 50 Roofing and drainage (Level 5) 51 Cladding and louvres (Level 5) 52 Glazing (Level 5) 53 Mechanical services (Level 5) 54 Lift installation in bus stations (Level 6) Remaining construction activities in multiple parts of the project 55 Line marking and signage (Level 5) 56 Landscaping (Level 5) 57 M&E (power supply, lighting, and fire services) (Level 6) 58 ITS (Level 5)


5.4.3 Stage 1/Task: C Market analysis

Based on make-or-buy analysis, the following Level 7 activities have been

identified:

1. Activity 2 - Civil engineering design and structural engineering design to the

driven tunnel

2. Activity 5 - Fire safety design for tunnels

3. Activity 8 - Planning for inspections and data collection for reactive and

preventive maintenance to specialist linings, and mechanical and electrical, and

fire elements in driven tunnel; and plan for reactive (routine) maintenance and

plan for preventive (programmed) maintenance to specialist linings,

membranes, mechanical and electrical, and fire elements in driven tunnel

Other than the above Level 7 activities, which are related to the design for the

driven tunnel, the remaining activities have similar activity levels and market

structures to the activities in Case Study #R1. Based on market analysis and

information from the case study interviews, there are limited firms in the market that

are capable of designing the driven tunnel in this case study, and the above Level 7

activities are not generated due to the size or scale of the workload in the activities

created by the size of the case study, but is generated in each of these design

activities by the rare and costly-to-imitate technical knowledge inherent in these

activities. Therefore, there is no opportunity to divide these activities into multiple

contracts in order to transpose the Level 7 to Level 5 or Level 6; thus, these activities

require further consideration in the next task.


The focus in this task is on second-order issues concerning the organisation and

management across key activities. The first step is to exclude any potentially

troublesome Level 7 and Level 4b activities from the bundling analysis, and these

activities are given in Table 5.21 (below).


Table 5.21 Bundles of Level 7 and Level 4b activities for #R2

No Key externalised activity Level 7 Design

2 Civil engineering design and structural engineering design to the driven tunnel (Level 7) 5 Fire safety design for tunnels (Level 7) 8 Plan for inspections and data collection, and plan for preventive maintenance in driven tunnel (Level 7)

Level 4b Construction - Cut and cover tunnels 10 Removal works: fencing, pavement, walls, sewer/water lines (Level 4b) 11 Traffic management (Level 4b) 12 Bored piles (Level 4b) 13 Excavate and shotcrete to bored piles (Level 4b) 14 Earthworks including: access; bulk excavation; strip topsoil; batter stabilisation; backfill (Level 4b) 15 Structural works – concrete cast insitu – column, headstock/abutments, planks, topping slab (Level 4b) 17 Waterproofing (Level 4b) 18 Drainage (Level 4b) 19 Pavement (a layer of no fines concrete and a paving layer) (Level 4b) Construction - Driven tunnel 23 Excavation in tunnel and shotcrete (Level 4b) 24 Waterproofing (Level 4b) 25 Structural works, including fibre reinforced polymer (FRP), headstocks, planks, FRP roof slab (Level 4b) 27 Drainage (Level 4b) 28 Trimming and backfill of main tunnel to finish (surface preparation) (Level 4b) 29 Pavement (a layer of no fines concrete and a paving layer) (Level 4b)

Government is limited in the extent to which it can use the prospect of future

work to check potential hold-up arising in the Level 7 and 4b activities, because of

the requirements for transparency and equity in tendering policy. In terms of the

Level 4b activities, even if a firm at the head of the supply chain can use the prospect

of future work to check negative opportunistic behaviour by subcontractors

delivering these activities, it might well still be inclined to construct negative

opportunistic behaviour on its own account. This is the result of tendering constraints

with government, and the desire to retain all of the hold-up premiums (quasi-rents)

from variations associated with Level 4b activity. The model recommends separating

the contract(s) for the potentially troublesome activities from the major bundle of

activities, and to contract these directly between government and firm(s); that is,

separate contracts for the design of the driven tunnel, and the construction of driven

tunnel, cut-and-cover tunnels, and bus stations.

Opportunities to bundle the remaining key externalised activities into main

activities and main activities into major activities are as follows:

1. The main design activity comprises of the following key activities, and the scale

of design work is unlikely to create a Level 7 activity at the level of this main

activity.


Table 5.22 Main design activities bundle for #R2

Bundles Main design activity Level 5

1 Civil engineering design and structural engineering design (Level 5) 3 Traffic engineering design (Level 5) 4 Mechanical and electrical engineering design, including air quality and ventilation (Level 5) 6 Landscaping and urban finishes design (Level 5)

2. The main construction activity comprises a total of 32 key activities (as follows

in Table 5.23); again, the scale of construction work is unlikely to generate a

Level 7 at the level of this main activity.

Table 5.23 Main construction activities bundle for #R2

Bundles Main construction activity Construction - Road at grade 31 Bulk excavation (Level 5) 32 Subgrade preparation (Level 5) 33 Drainage (Level 5) 34 Concrete pavement (Level 5) 35 Precast concrete – barriers, kerbs (Level 5) 36 Retaining walls (Level 5) 37 Asphalt pavement (Level 5) 38 Re-alignment of existing busway (Level 5) 39 Traffic management (Level 5) Construction: Bridge, ramps, median, walkway and bikeway structures

40 Traffic management (Level 5) 41 Earthworks (Level 5) 42 Pile foundation (Level 5) 43 Structural works, including: cast insitu columns, headstock, abutments, planks, concrete pavement (Level 5) 44 Precast concrete – barriers, kerbs (Level 5) Construction - Bus stations 45 Water and stormwater services (Level 5) 46 Electrical and communication services (Level 5) 47 Pile foundations (Level 5) 48 Cast insitu concrete (lift well, platforms and bus bays) (Level 5) 49 Structural steelwork (Level 5) 50 Roofing and drainage (Level 5) 51 Cladding and louvres (Level 5) 52 Glazing (Level 5) 53 Mechanical services (Level 5) Remaining construction activities in multiple parts of the project 16 Precast concrete – concrete barriers and kerbs (Level 5) (similar to Activity 26) 20 Modifications to existing bridge and footpath beams and slab (Level 5) 21 Demolition works, including building and bridge structures, walls and pavements (Level 5) 55 Line marking and signage (Level 5) 56 Landscaping (Level 5) 58 ITS (Level 5) Level 6 30 Ventilation fans in the driven tunnel 54 Lift installation in bus stations 57 M&E (power supply, lighting, and fire services)

There is no viable DCM, DCO or DCOM major activity bundle to consider,

either as a PPP or as a contract. Thus, the default strategy for Road Case Study #R2


is to separate the procurement of the 4 externalised activities into 4 separate contracts

between the contractors (or consultants) and government, or via a government

agency or PM consultant:

• Contract #1: Design of driven tunnel, including fire safety design (level 7)

• Contract #2: Design of remaining part of project (main design bundle) • Contract #3: Construction of tunnels both cut-and-cover tunnels, and driven

tunnel (Level 4b) • Contract #4: Construction of remaining part of project (main construction

bundle)

Contracts #1 and #2 might or might not be awarded to the same consultancy,

and are determined through a competitive bidding process; similarly, Contracts #3

and #4 might or might not be awarded to the same main contractor, and are also

determined through a competitive bidding process.


Finally, in this stage, the TCE variables are measured relating to Contract#1,

#2, #3 and #4, which now include the planning and coordination across key activities

with an anticipated typical market firm at the head of each contract or counterparty to

government. Table 5.24 (below) gives the summary of the scores for the TCE

variables and the corresponding patterns.

Table 5.24 Exchange relationship analysis of #R2

No Contract Asset specificity

Uncertainty Frequency Pattern Exchange relationship

1 Contract #1: Main activity – Consultant responsible for planning, coordination, and overall delivery of design of driven tunnel, including fire safety design

5,7 5,5 4 (++,+,+/0) Very inefficient discrete

2 Contract #2: Main activity – Consultant responsible for planning, coordination, and overall delivery of design of project (with the exception of the driven tunnel)

2,2 3,3 4 (0,0,+/0) Extremely efficient discrete

3 Contract #3: Main activity – Contractor responsible for planning, coordination, and overall delivery of the construction of all tunnels

5,7 6,7 4 (+,+++,+/0)

Relational exchange

4 Contract #4: Main activity – Contractor responsible for planning, coordination, and overall delivery of all construction in the project (with the exception of tunnels)

3,4 3,3 4 (0,0,+/0) Extremely efficient discrete

When matched to Figure 3.2, the patterns indicate that the exchange between

state government and the market firm for Contract 1#, where the consultant is


responsible for the planning, coordination and overall delivery of design of driven

tunnel (including fire safety design), exhibits a very inefficient discrete exchange

using a bespoke consultancy agreement with credible threats for non-performance.

The principal-agent characteristics of this discrete exchange, which connect to an

outcome-based exchange, are as follows:


• agency costs • specification of outcomes

• verification of outcomes • risk premium

• suitability for information asymmetry • outcome certainty • better goal alignment

As for Contract 2#, where the consultant is responsible for the planning,

coordination and overall delivery of design of project, except for the driven tunnel,

the patterns point to an extremely efficient discrete exchange using a standard neo-

classical consultancy agreement. The principal-agent characteristics of this discrete

exchange, which connect to an outcome-based exchange, are as follows:


• agency costs • specification of outcomes

• verification of outcomes • risk premium • suitability for information asymmetry


For Contract #3 which is related to the planning, coordination and overall

delivery of the construction of all tunnels, the TCE variables have high scores, which

give a pattern indicating a relational exchange. The principal-agent characteristics of

this relational exchange, which connect to a mixed outcome-based and behaviour-

based exchange (or hybrid exchange), are as follows:


a. target outturn costs/guaranteed construction sum linked with gain share/pain share regime

b. risks balanced between agent and principal c. suitable for outcome uncertainty

For the last contract, Contract #4 – the planning, coordination, and overall

delivery of all the construction in the project except for tunnels – the TCE patterns

reveal an extremely efficient discrete exchange using a standard neo-classical

construction contract. The corresponding principal-agent characteristics of this

discrete exchange, which connect to an outcome-based exchange, are:


• agency costs

• specification of outcomes • verification of outcomes • risk premium

• suitability for information asymmetry • outcome certainty



The procurement strategy recommended by the first-order procurement

decision-making model is substantially different to actual practice. The model only

supports actual procurement in so far as the internalisation and externalisation of

activities, and the incorporation of an alliance component in part of the externalised

activities. The major differences recommended by the procurement strategy are as

follows:

1. The model recommends four contracts for the externalised activities; this is in

contrast to the actual procurement that used only one alliance contract to

procure all externalised activities.

2. Of these four contracts, the model recommends only one alliance mode, and

this is for the construction of driven and cut-and-cover tunnels.

3. The other three contracts comprise a bespoke, extreme arms-length contract for

the design of driven tunnel and two standard fixed-priced contracts – one for

the design of all works excluding the driven tunnel, and one for the

construction of all works excluding the driven and cut-and-cover tunnels.



The recommended procurement strategy for Case Study #R2 demonstrates that

appreciable improvements are theorised through the use of multiple contracts (with

one alliance component) and a greater reliance on standard fixed price terms in lieu

of the whole project being procured in alliance mode. These differences generate a

greater number of EoIs than the low number of two EoIs generated by the actual

procurement approach.

According to the survey of civil and building contractors, most of the larger

civil contractors from this case study state who were capable of Alliancing contracts

of at least $350 million indicated that they were operating at near full capacity.

Furthermore, 3 smaller firms, capable of Construct only contracts up to $200 million,

indicated spare capacity. Upon rationalising the number to 100 percent, this gives

(potentially) approximately 8 contractors who may be able to bid for the project; this

is greater than the original 2 or 3 EoI, and increases the EoI to within the optimal 5 to

8. This matches with Scenario 4 of the refutability procedures in Section 3.3.5; that

is, at sub-optimal competition level of relatively low EoI, if the empirical

procurement is appreciably dissimilar from the model’s prediction, then the predicted

procurement is likely to generate a greater number of EoI, to within the optimal

number of 5 to 8. This provides support to the model’s ability to strategise

procurement approaches that avoid market failure ex ante.

In conclusion, Road Case Study #R2 supports the hypothesis in so far as there

are some appreciable differences in the procurement identified by the model and

actual practice, given the very low number of EoI generated by the actual

procurement approach (2 EoI) and the expectation that the predicted procurement

will generate an optimal EoI number. The following section is an analysis of Health

Case Study #H2.

5.5 Health Case Study #H2 (Theoretical procurement)

Health Case Study #H2 had a sub-optimal EoI of 15 contractors expressing

interest in tendering; this is much higher than the optimal EoI. This project was part

of an expansion of an existing hospital. Stage A comprised of a four-storey new

building, a two-storey new building, and extensions to existing buildings totaling

over 20 000m2 GFA. Stage B consisted mainly of refurbishment works within parts


of the existing building. The greatest challenge was to ensure minimum disruption to

the existing hospital services, sensitive equipment and patients. The project was

procured using the managing contracting approach. With approximately 75 percent

of the schematic design and documentation completed, tender invitations were issued

to four out of the 15 contractors that expressed interest. The selected managing

contractor was paid based on a fixed management fee, and incentive payments for

achieving cost savings. The fixed Guaranteed Construction Sum (GCS) of the project

was finalised through a series of negotiations, rather than the competitive tendering

process in the traditional procurement approach. The design of the hospital was

updated as the construction progressed. The operations and maintenance of the

hospital were either sourced separately or internally; for example, cleaning and

security services were outsourced to an external service provider, while the reactive

maintenance of non-specialised items were carried out by full-time on-site staff.

The outcomes of applying the procurement model to the project are as follows.


Determining the activities for Health Case Study #H2 is less straightforward

than for the road case studies; specifically, with respect to determining the

parameters of a key activity, especially the activities in the operations and

maintenance stage. The scope of activity analysis became clearer during the

deliberation of the extent of activity analysis. That is, core or front-line services that

did not involve the physical asset of the project were not considered key activities for

analysis; for example: clinical and administrative services, such as patient records;

ancillary services, such as linen and porterage services; and the design, supply,

installation or positioning of loose equipment or furniture and plug-in equipment,

such as x-ray equipment and computed tomography (CT) scanners. A total of 55 key

activities across key production activities of design, construct, operations and

maintenance, were identified for Health Case Study #H2.

The design activities comprise of two main groups: the design of new

installation or construction works (which includes functional brief, conceptual

design, design document, and developed design), and the design of performance

specification of maintenance or the planning for maintenance schedule. The design

components of new construction works comprise of eight distinct key activities

requiring differing skill sets, namely: 1. architectural; 2. civil and structural


engineering; 3. mechanical and electrical services [including building management

system (BMS), security, information and communications technology (ICT)] and

hydraulics (including wet fire systems); 4. dry fire system; 5. highly specialised

hydraulics, including medical gases and pneumatic tubes; 6. kitchen design; 7.

landscape design and 8. traffic engineering and systems design.

The hospital has highly specialised hydraulic requirements (such as medical

gases and pneumatic tube systems), and other requirements that are unique to the

hospital design, such as BMS and kitchen design. During the course of answering the

case study questions, highly specialised services specific to the hospital design

emerged as a separate activity, distinct from the rest of the design of building

engineering services.

The design of maintenance specification and the planning of maintenance

schedule have been considered as a single activity. This activity involves the

planning of building, engineering, and maintenance services using computer software

which schedules the timing for all reactive and preventive maintenance, be it routine

or emergency, or specialised or non-specialised in nature.

In relation to construction activities, there are 36 key activities: 1. demolition

works; 2. site establishment works; 3. excavation works; structural works, such as, 4.

bored piers, 5. concrete, 6. masonry, 7. structural steelwork; internal finishes, such

as, 8. carpentry, 9. Metalwork, 10. doors and door frames, 11. automatic door, 12.

vinyl floor and wall finishes, 13. carpet floor finishes, 14. ceramic floor or wall tiles,

15. painting, 16. signage and way-finding; external finishes, such as 17. roofing, 18.

roof safety system, 19. glazing and windows, and façade, 19. fixtures, furnishings

and equipment (FFE) for non-specialised items, such as, whiteboards and AV

equipment; 20. FFE for specialised items, such as sterilising equipment; 21.

commercial kitchen equipment, such as dishwasher, ovens and refrigerators; 22. cold

room installation; 23. heating, ventilation, and air conditioning (HVAC); 24.

electrical, including in-ground electrical services; 25. BMS; 26. security, including

security cameras, duress alarms, access readers; 27. communications; 28. hydraulics,

including plumbing and drainage; 29. hydraulics on medical gases, such as oxygen

gas, breathing air; 30. hydraulics on pneumatic tubes; 31. fire services, including fire

detectors, fire indicator panel (FIP); 32. emergency warning and intercommunication

system (EWIS), including fire extinguishers and blankets; 33. lifts; and external


works, such as, 34. soft and hard landscaping; 35. car parking; and 36. covered

walkways.

Two activities have been identified for the operations stage which involve

interaction with the physical component of the hospital building; namely, the

operations of BMS and operations of security services, such as operation of security

cameras in the hospital. Both activities require 24-hour supervision and constant

monitoring of the building and engineering services. For instance, when the BMS

signals an anomaly in the pressure or temperature of a certain location, it needs to be

checked and fixed immediately.

The maintenance of the building and engineering services in the Health Case

Study #H2 includes: mechanical, electrical, plumbing and carpentry works; and

floor, wall and ceiling lining and finishes. Maintenance across all building and

engineering services has been categorised into two broad categories, namely, reactive

and preventive.

Reactive maintenance refers to maintenance that includes routine and

emergency maintenance that can be specialised or non-specialised in nature. Reactive

routine maintenance refers to maintenance activities that are carried out on a regular

basis, or do not require critical or immediate response. Routine reactive maintenance

might require specialised knowledge (such as fixing specialised equipment) or might

not require such knowledge (such as fixing door handles). On the other hand,

reactive emergency maintenance requires a speedy response and generally warrants

specialised servicing, such as tuning thermostatic-mixing valves or examining

malfunctioning of medical equipment.

In contrast to reactive maintenance, preventive maintenance refers to

maintenance activities that have been scheduled or programmed to prolong

equipment life and to minimise any unexpected maintenance activity: HVAC,

electrical, building fabric maintenance (vinyl, carpet, ceramic tiling, painting, and

waterproofing), hardware, fixtures and fittings, hydraulics (valves, thermostatic

mixing pumps, hot water system, pumps, pneumatic tubes and medial gases),

landscaping and external works, lifts, painting, pumps, roller shutter and entry gates.

Preventive maintenance has also been grouped into specialised and non-

specialised categories. The maintenance of building and engineering services has


been categorised specifically into six activities: 1. reactive routine (non-critical) and

non-specialised; 2. reactive routine (non-critical) and specialised; 3. reactive

emergency (critical) and non-specialised; 4. reactive emergency (critical) and

specialised; 5. preventive, programmed and non-specialised; and 6. preventive,

programmed and specialised. Cleaning is also another important maintenance

activity in the hospital, and needs to be carried out on a daily basis, as it is critical in

providing a hygienic environment and reducing the risk of infections. The last

activity concerns the landscaping of the external works of the hospital.


Table 5.25 (below) is a summary of the patterns and corresponding levels of

each design activity, bearing in mind that the responses are reflective of the

competencies and capabilities of the government and market in 2008. Design

activities begin from design briefs to schematic design, and developed design

(including contract documentation. The architect and engineer need to be

professionally accredited and ensure building code compliance and disable access

compliance in the building design. Table 5.25 shows that architectural, civil and

structural, dry fire system, landscape and traffic engineering design activities are

assigned Level 5, while the remaining building engineering services and specialised

commercial kitchen design are assigned Level 6. The state government agency has

minimal capacity across all design activities, which are beyond their technical

capability, especially for Level 6 activities. This lack of capacity matches low

frequency or a non-typical project in the government agency.

Across the design activities, the design of building engineering services in

Activities 3 and 4 – with the exception of dry fire design – has received higher scores

for asset specificity and uncertainty, because of the difficulties and complexities in

verifying as-built information, and in integrating the various types of building

engineering services with the existing hospital. Similarly, for kitchen design (which

requires specialised knowledge of the design requirements of a hospital), there is a

corresponding limited supply of firms with the necessary capability and experience.

The expertise in Level 6 activities cannot be easily acquired or developed in a short

span of time. The design for the majority of building engineering services –

excluding the highly specialised hydraulic services – have a higher score of ‘5’ on

asset specificity and uncertainty variables, while highly specialised hydraulic


services have a lower score of ‘4’. This is because these services are more discrete

and isolated in terms of their design component and have fewer integration issues.

Table 5.25 Make-or-buy analysis of design activities for #H2

no Activity Capacity Rarity Costly-to-

imitate

Asset specificity


1 Architectural design (concept – from brief to space design; schematic design; developed design; contract documentation design, including building code compliance and review; disabled access compliance; fixture and fittings)

3 3,4 3,3 3,4 2,3 4 (-, 0,0,0 to +,0,0)

5

2 Civil and structural design (schematic design; developed design)

3 1,2 1,2 2,1 1,1 4 (-, 0,0,0 to +,0,0)

5

3 Building engineering services design: Mechanical and electrical (including BMS, Security, ICT) and hydraulics (including wet fire systems)

2 5,5 4,4 5,5 5,5 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

4 Building engineering services design: Highly specialised hydraulics (including medical gases and pneumatic tubes)

2 5,5 4,4 4,4 4,4 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

5 Building engineering services design: Dry fire systems

3 4,4 2,2 3,3 3,3 4 (-, 0,0,0 to +,0,0)

5

6 Kitchen Design 2 6,6 1,1 3,3 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

7 Landscape design 3 3,3 2,2 2,2 1,1 4 (-, 0,0,0 to +,0,0)

5

8 Traffic engineering and systems design

3 1,1 1,1 2,2 1,1 4 (-, 0,0,0 to +,0,0)

5

9 Planning for building, engineering services maintenance comprising: Reactive (non-critical/routine/non-specialised); reactive (critical/emergency /non-specialised and critical/emergency/specialised); preventive (programmed/non-specialised and programmed/specialised)

5 2,2 2,2 5,5 3,3 4 (+, 0,0,0 to +,0,0)

Mismatch: Either Level 3 or 5 (Model recommends Level 5)

The government has greater capacity for, and competence in Activity 9

(planning for engineering services maintenance) than it has in Activities 1 to 8.

However, the low frequency in this activity mismatches with the positive capacity,

and the pattern indicates either a Level 3 or Level 5. Although the state government

can benefit from their on-site maintenance staff’s knowledge of the building, a

contractor can also acquire this knowledge by directly employing these staff and/or

by acquiring site records. Furthermore, the Managing Contractor engages the trades

or specialist contractors in the construction of this health project and maintenance in

other projects and, thus, has the specific knowledge of all works and the potential


supply chain advantage over state government. It is noted that, in this case, the

Managing Contractor is currently on-site as part of a variation and is helping to

manage staged hand-over in the defects liability period. This situation is

demonstrating some of the advantages that the Managing Contractor has over state

government staff and gives us a better understanding of the initial transitional phase

in maintenance. In addition to these practical reasons, the frequency variable takes

theoretical precedence; thus, Activity 9 is assigned Level 5.

As the construction activities are carried out in a single location,

compartmentalised and isolated, they are exposed to similar uncertainties and

difficulties bounded by the site conditions. Thus, the results of make-or-buy analysis

of the construction activities are compared and contrasted among different groups of

activity levels. Table 5.26 (below) lists the activities that are assigned Level 7. These

activities require specialised knowledge or proprietary technology that the state

government cannot supply or readily source or acquire. The market is technically

and/or organisationally better than government at the time of making the

procurement decision.

Table 5.26 Make-or-buy analysis of Level 7 construction activities for #H2


imitate

Asset specificity


36 Building management services (BMS)

1 7,7 5,5 5,5 5,5 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

31 Fixtures, Furnishings & Equipment (FFE) – supplied and install specialised items (e.g. sterilising equipment)

1 6,6 5,5 3,3 1,1 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

41 Hydraulics – Pneumatic tubes complete with pumps, compressors and outlets etc.

1 7,7 5,5 5,5 5,5 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

Activity 36 refers to BMS, a proprietary computer-based system that controls

and monitors the heating, ventilation, air-conditioning mechanical services in the

hospital to improve energy efficiency and provide a comfortable environment. It is

beyond the government’s technical capability to supply and install the software and

hardware of the BMS which needs to be integrated with the existing hospital BMS.

Activity 41 represents the pneumatic tube system that is a computer-controlled

vacuum transportation system within the hospital to deliver drugs, blood samples and


documents. Activity 31 refers to the supply and installation of specialised equipment,

such as sterilising equipment that needs to meet medical standards.

Across these activities, the required resources are extremely rare and costly-to-

imitate; that is, the tacit knowledge and expertise required are possessed by only a

limited supply of firms. Furthermore, installation of proprietary systems makes

imitations or maintenance almost impracticable to imitate; in other words, there are

high switching costs or practical lock-in in some services. Correspondingly, the value

variable has a low value score of ‘1’ and symbol ‘---’ which represents that a

substantial government investment will be required if the resources are acquired

internally. The activities are assigned Level 7, given the strong dominance of the

RBT variables. While scoring across these Level 6 and Level 7 services, we could

then proceed to separate activities into proprietary and non-proprietary. Although

changes in technology are likely, these relate more specifically to plug-in health

equipment; therefore, inertia exists around spaces such as theatres, suicide-proof

spaces, and other specific spaces and so this reduces uncertainty arising from

technological changes.

Table 5.27 (below) summarises the scores and corresponding patterns for Level

6 construction activities.



imitate

Asset specificity


23 Automatic door 2 5,5 3,3 2,2 1,1 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

27 Façade 2 5,5 2,2 2,2 1,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

34 HVAC 2 5,5 2,2 5,5 5,5 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

35 Electrical, including in- ground electrical

2 5,5 2,2 2,2 1,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

38 Communications 2 6,6 2,2 5,5 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

40 Hydraulics - Medical gases, complete with copper piping, pumps, compressors & outlet

2 6,6 2,2 3,3 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

42 Fire services, incl. fire detectors, FIP, EWIS,

2 6,6 2,2 2,2 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

43 Lifts 2 5,5 2,2 3,3 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

Although the activities in Table 5.27 – namely, automatic door, façade system,

HVAC, electrical works, communications, hydraulics, fire services, and lifts –

require specialised technical knowledge (that is, they are tacit in nature and cannot be

readily sourced), these services can also be applied to buildings other than hospitals.


The government agency might have gained some experience or knowledge in these

activities from the supervision of other projects, and they might not be significantly

beyond the government’s capacity, compared with Level 7 activities. However, it

will still require a certain level of training and investment to internalise these

activities. While the supply of these activities is greater than for those in level 7, they

might still not be widely available. Although, the expertise required for these

activities are less specialised, they require firms with the expert knowledge and

skills, to be able to apply in a hospital environment. The overall pattern matches with

Level 6 in the integrative framework which represents that the market is technically

better than the government in performing these activities.

The following Table 5.28 gives the scores and patterns corresponding to Level

5 for the remaining construction activities. This table indicates that the majority of

the construction activities are Level 5, and 26 of these are identified. Based on the

low scores for the TCE and RBT variables, all of these Level 5 activities exhibit

characteristics which confirm that the market is organisationally better than the

government. Even though the government agency has the tacit knowledge and skills,

further investment will still be required to perform these activities. However, this

investment will not be as great as it would be for Level 6 and 7 activities. In terms of

rarity, these activities are more easily sourced in the market, and the technology and

tacit knowledge in the activities are more easily imitated, in comparison with Level 6

and 7 activities. However, the market has the competitive advantage of accessing the

supply chain, aggregating greater work flow, gaining competence in organising, and

making fewer mistakes, in comparison with the government.



no Activity Capacity Rarity Costly to

imitate

Asset specificity


10 Demolition works and clearance

3 1,1 1,1 1,1 2,2 4 (-, 0,0,0 to +,0,0) 5

11 Site establishment works, including temporary services, site fencing and hoarding, accommodation

3 1,1 1,1 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5

12 Excavation works (Bulk and detailed)

3 2,2 2,2 1,2 3,3 4 (-, 0,0,0 to +,0,0) 5

13 Bored piers 3 2,1 1,1 1,2 3,3 4 (-, 0,0,0 to +,0,0) 5 14 Concrete; formwork;

and reinforcement 3 1,2 1,2 1,2 2,3 4 (-, 0,0,0 to +,0,0) 5

15 Masonry 3 1,1 1,1 1,2 2,3 4 (-, 0,0,0 to +,0,0) 5 16 Structural steel work 3 3,3 1,1 1,3 2,2 4 (-, 0,0,0 to +,0,0) 5 17 Carpentry works,

including wall and ceiling construction (studwork, ceiling lining and framing)

3 1,1 1,1 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5

18 Metalwork, including balustrade handrails, privacy screens etc.

3 3,3 1,1 1,2 1,2 4 (-, 0,0,0 to +,0,0) 5

19 Roofing (Corrugated metal roof sheeting)

3 2,2 2,2 1,1 1,2 4 (-, 0,0,0 to +,0,0) 5

20 Roof safety system 3 3,3 2,2 1,1 1,2 4 (-, 0,0,0 to +,0,0) 5 21 Glazing and windows 3 2,3 2,2 1,2 1,2 4 (-, 0,0,0 to +,0,0) 5 22 Doors and door frames 3 2,2 1,1 2,2 1,1 4 (-, 0,0,0 to +,0,0) 5 24 Vinyl floor and wall

finishes 3 2,2 1,1 2,2 1,1 4 (-, 0,0,0 to +,0,0) 5

25 Carpet floor finishes 3 1,1 1,1 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5 26 Ceramic floor/wall tiles 3 1,1 1,1 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5 28 Painting 3 1,1 1,1 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5 29 Signage and way-

finding 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5

30 Fixtures, Furnishings & Equipment (FFE) – supplied and install non-specialised items (e.g. whiteboards and AV equipment)

3 2,2 2,2 2,2 1,1 4 (-, 0,0,0 to +,0,0) 5

32 Commercial kitchen equipment, including dishwasher, ovens and fridges

3 2,2 2,2 1,1 1,1 4 (-, 0,0,0 to +,0,0) 5

33 Cold room installation 3 3,3 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 37 Security, including

security cameras, duress alarms, access readers etc.

3 3,3 2,2 4,4 2,2 4 (-, 0,0,0 to +,0,0) 5

39 Hydraulics, including plumbing and drainage

3 4,4 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5

44 External works – Hard and soft landscaping

3 2,2 1,1 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5

45 External works – Carparking

3 2,2 1,1 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5

46 External works - Covered walkways

3 2,2 1,1 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5


Table 5.29 (below) summarises the make-or-buy analysis for operation

activities.

Table 5.29 Make-or-buy analysis of operation activities for #H2


imitate

Asset specificity


47 Building management system (BMS)

5 1,1 1,1 2,2 1,1 4 (+, 0,0,0 to +,0,0)


48 Protective services /security

5 1,1 1,1 2,2 1,1 4 (+, 0,0,0 to +,0,0)


Similar to Activity 9, Activities 47 and 48 receive conflicting scores between

the value variable and frequency variable, and the activities can be either Level 3 or

Level 5. Activity 47 refers to the operation of BMS which requires only a small

amount of training or time for an engineer with some technical knowledge. This

activity can then be organised or managed internally or externally. However, there

may be greater synergies when connected with the outcome from Activity 9.

Therefore, as Activity 9 is transposed to externalisation, then advantages also accrue

to Activity 47 in terms of external provision. Given these practical reasons, including

precedence by frequency variable, the model proposes to externalise the activity, and

it is a Level 5 activity.

Activity 48 refers to the protective or security services of the hospital which

require 24-hour supervision of the CCTV cameras. This activity does not require an

appreciable amount of training or time (say, a few weeks) to familiarise security

guards who could be doing security work in other non-health establishments. Again,

this activity can be organised or managed internally or externally. Given that

theoretical frequency variable takes precedence, the model proposes the activity to

proceed with externalisation as Level 5.

Table 5.30 (below) gives the make-or-buy analysis for the maintenance

activities. As discussed earlier, there are six types of maintenance across all building

and engineering services, in addition to cleaning and landscaping works in the

maintenance stage.


Table 5.30 Make-or-buy analysis of maintenance activities for #H2


imitate

Asset specificity


49 Building, engineering services maintenance: Reactive (non-critical/routine/non-specialised)

5 1,1 1,1 2,2 1,1 4 (+, 0,0,0 to +,0,0)


50 Building, engineering services maintenance: Reactive (non-critical/routine/specialised)

1/2 6/7,6/7

6/7,6/7

3,3 2,2 4 (---, + to +++,+ to +++ , 0 to +++, 0 to +++, 0)

Level 6/7

51 Building, engineering services maintenance: Reactive (critical/emergency /non-specialised)

5 1,1 1,1 4,4 1,1 4 (+, 0,0,0 to +,0,0)


52 Building, engineering services maintenance: Reactive (critical/emergency/specialised)

1/2 6/7,6/7

6/7,6/7

3,3 4,5 4 (---, + to +++,+ to +++ , 0 to +++, 0 to +++, 0)

Level 6/7

53 Building, engineering services maintenance: Preventive (programmed/non-specialised)

5 2,2 2,2 2,2 2,2 4 (+, 0,0,0 to +,0,0)


54 Building, engineering services maintenance: Preventive (programmed/ specialised)

1/2 6/7,6/7

6/7,6/7

3,3 4,4 4 (---, + to +++,+ to +++ , 0 to +++, 0 to +++, 0)

Level 6/7

55 Cleaning 5 1,1 1,1 2,2 1,1 4 (+, 0,0,0 to +,0,0)


56 Landscaping and external works

5 2,2 2,2 2,2 2,2 4 (+, 0,0,0 to +,0,0)


Activities 49, 51 and 53 refer to the implementation of reactive maintenance,

that are non-specialised in nature, and can be organised or managed either internally

or externally. If connected with the outcomes from Activity 9, there will be greater

synergies. As the scores on the value variable indicate that the government has

positive capacity to organise or manage these activities, there is a mismatch in the

frequency variable; this indicates that the activities in the project are not typical


within the portfolio of projects. However, as Activity 9 is transposed to

externalisation, then advantages might also accrue in Activities 49, 51 and 53 in

terms of external provision. Given these reasons and precedence of the frequency

variable, Activities 49, 51 and 53 are theoretically assigned Level 5.

For maintenance activities that require specialised attention in the actual

implementation of maintenance – either non-critical reactive (Activity 50), critical

reactive (Activity 52), or programmed preventive (Activity 54) – there are no

conflicting scores between value and frequency variables. The scores are

straightforward in indicating that the activities are either Level 6 or 7, and were

beyond the government’s technical capability at the time when the procurement

decision was made.

Activity 55 refers to the cleaning activities in the hospital which do not require

an appreciable amount of training time (perhaps a few weeks or so). This activity can

then be organised or managed internally as well as externally. Similarly, Activity 56

does not need an appreciable amount of training time; perhaps a few weeks or so will

be enough to become familiarised with the hospital site. This activity can then be

organised or managed internally as well as externally. Given the frequency variable

takes precedence, then Activity 55 is assigned Level 5.

To summarise, and based on the activity analysis, the model proposes that all

the activities across key production activities of design, construction, operations and

maintenance be externalised (no internalised activities). This does not match with the

activities on the maintenance of non-specialised building and engineering services

and cleaning services which were internalised in practice. Table 5.31 (below)

summarises all the activities that match with current practice, and Table 5.32 (also

below) summarises those that do not match.


Table 5.31 Summary of externalised activities that match practice for #H2

No Key activity Design

Design of new installation/construction works 1 Architectural design (concept – from brief to space design; schematic design; developed design; contract

documentation design including building code compliance and review; disabled access compliance; fixture and fittings) (match: Level 5)

2 Civil and structural design (schematic design; developed design) (match: Level 5) 3 Building engineering services design: Mechanical and electrical (including BMS, Security, ICT) and

hydraulics (including wet fire systems) (match: Level 6) 4 Building engineering services design: Highly specialised hydraulics (including medical gases and

pneumatic tubes) (match: Level 6) 5 Building engineering services design: Dry fire systems (match: Level 5) 6 Kitchen Design (match: Level 6) 7 Landscape design (match: Level 5) 8 Traffic engineering and systems design (match: Level 5)

New installation/construction works 10 Demolition works and clearance (match: Level 5) 11 Site establishment works, including temporary services, site fencing and hoarding, accommodation

(match: Level 5) 12 Excavation works (Bulk and detailed) (match: Level 5) 13 Bored piers (match: Level 5) 14 Concrete; Formwork; and Reinforcement, including beams (match: Level 5) 15 Masonry (match: Level 5) 16 Structural steel work (match: Level 5) 17 Carpentry works, including wall and ceiling construction (studwork, ceiling lining and framing) (match:

Level 5) 18 Metalwork, including balustrade handrails, privacy screens etc. (match: Level 5) 19 Roofing (Corrugated metal roof sheeting) (match: Level 5) 20 Roof safety system (match: Level 5) 21 Glazing and windows (match: Level 5) 22 Doors and door frames (match: Level 5) 23 Automatic door (match: Level 6) 24 Vinyl floor and wall finishes (match: Level 5) 25 Carpet floor finishes (match: Level 5) 26 Ceramic floor/wall tiles (match: Level 5) 27 Façade (match: Level 6) 28 Painting (match: Level 5) 29 Signage and way-finding (match: Level 5) 30 Fixtures, Furnishings & Equipment (FFE) – supplied and installed non-specialised items (e.g. whiteboards

and AV equipment) (match: Level 5) 31 Fixtures, Furnishings & Equipment (FFE) – supplied and installed specialised items (e.g. sterilising

equipment) (match: Level 7) 32 Commercial kitchen equipment, including dishwasher, ovens and fridges (match: Level 5) 33 Cold room installation (match: Level 5) 34 HVAC (match: Level 6) 35 Electrical including in ground electrical (match: Level 6) 36 Building management services (BMS) (match: Level 7) 37 Security, including security cameras, duress alarms, access readers etc. (match: Level 5) 38 Communications (match: Level 6) 39 Hydraulics, including plumbing and drainage (match: Level 5) 40 Hydraulics - Medical gases (such as oxygen gas, breathing air), complete with copper piping, pumps,

compressors and outlets etc. (match: Level 6) 41 Hydraulics – Pneumatic tubes complete with pumps, compressors and outlets etc. (match: Level 7) 42 Fire services, including fire detectors, fire indicator panel (FIP), Emergency Warning and

Intercommunication System (EWIS), fire extinguishers and blankets (match: Level 6) 43 Lifts (match: Level 6) 44 External works – Hard and soft landscaping (match: Level 5) 45 External works – Carparking (match: Level 5) 46 External works - Covered walkways (match: Level 5) 50 Building, engineering services maintenance: Reactive (non-critical/routine/specialised) (match: Level 6/7) 52 Building, engineering services maintenance: Reactive (critical/emergency/specialised) (match: Level 6/7) 54 Building, engineering services maintenance: Preventive (programmed/specialised) (match: Level 6/7


Table 5.32 Summary of externalised activities that do not match practice for #H2

No Key activity Design Design of performance specification of maintenance 9 Planning for building, engineering services maintenance comprising: Reactive (non-critical/routine/non-

specialised); reactive (critical/emergency/non-specialised and critical/emergency/specialised); preventive (programmed/non-specialised and programmed/specialised) (mismatch: positive value and low frequency: frequency takes theoretical precedence, plus practical reasons; thus, not Level 3, but proceed as Level 5)

New installation/construction works Implementation of operations 47 Building management system (BMS) (mismatch: positive value and low frequency: frequency takes

theoretical precedence, plus practical reasons; thus, not Level 3, but proceed as Level 5) 48 Protective services/security (mismatch: positive value and low frequency: frequency takes theoretical

precedence, plus practical reasons; thus, not Level 3 but proceed as Level 5) Implementation of maintenance 49 Building, engineering services maintenance: Reactive (non-critical/routine/non-specialised) (mismatch:

positive value and low frequency: frequency takes theoretical precedence, plus practical reasons; thus, not Level 3 but proceed as Level 5)

51 Building, engineering services maintenance: Reactive (critical/emergency/non-specialised) (mismatch: positive value and low frequency: frequency takes theoretical precedence, plus practical reasons; thus, not Level 3 but proceed as Level 5)

53 Building, engineering services maintenance: Preventive (programmed/non-specialised) (mismatch: positive value and low frequency: frequency takes theoretical precedence, plus practical reasons; thus, not Level 3 but proceed as Level 5)

55 Cleaning (mismatch: positive value and low frequency: frequency takes theoretical precedence, plus practical reasons given in footnote on the activity page; thus, not Level 3 but proceed as Level 5)

56 Landscaping and external works (mismatch: positive value and low frequency: frequency takes theoretical precedence, plus practical reasons; thus, not Level 3 but proceed as Level 5)


From the above task, there are a mixture of Level 5, 6 and 7 activities. In

corroborating with the levels identified, there are numerous firms for Level 5

activities and several firms for Level 6 activities in the market, and there is no

indication of a monopoly or duopoly market structure. As for Level 7 activities, there

are three activities from the make-or-buy analysis, that is:

1. Activity 31 – Fixtures, furnishings and equipment (FFE) – supply and install

specialised items, such as sterilising equipment (Level 7)

2. Activity 36 – Building management services (BMS) (Level 7)

3. Activity 41 – Hydraulics – Pneumatic tubes, complete with pumps, compressors

and outlets (Level 7)

From the market analysis, these Level 7 activities are supplied by a limited

number of firms capable of performing these activities, as a result of rare and costly-

to-imitate technology, and not as a result of the size of the activity. Therefore, no

further consideration is required here, and we can proceed to bundling analysis.



From the make-or-buy analysis summary and market analysis, there are no

troublesome Level 7 activities arising as a result of scale or any Level 4b activities;

however, there are three Level 7 activities arising out of rare and costly-to-imitate

technology:

1. Activity 31 – Fixtures, furnishings and equipment (FFE) – supply and install

specialised items, such as sterilising equipment (match: Level 7)

2. Activity 36 – Building management services (BMS) (match: Level 7)

3. Activity 41 – Hydraulics – Pneumatic tubes, complete with pumps, compressors

and outlets (match: Level 7)

While there may be some leverage that the main contractor can exercise in

terms of the threat of losing future work, this may not be appreciable given that the

firms delivering these activities are in thin market structures. However, the biggest

constraint in this project is that it will not be practical to have separate contracts

directly to government outside of the main contractor’s contract, given the extremely

high level of integration of activities that is required. Therefore, the mechanism used

to first address the possibility of hold-up post-contract is via an extremely inefficient

exchange incorporated with credible threats. Secondly, an inefficient exchange also

helps to reduce the likelihood of lack of competition in the main construction

contract; for instance, if any of these firms is only willing to bid with certain

contractors. Not only can this lack of competition be checked via the exchange

relationship but, at the same time, it incorporates the process of nomination of these

specialist firms so that no main contractor has a competitive advantage or

disadvantage, and the competition at the level of the main contractor can then be

protected.

It is then time to consider opportunities to bundle key externalised activities

into main activities, and main activities into major activities and, while doing so, to

consider each main activity and whether bundling is likely to generate a Level 7 at

the main activity bundle.

1. The main design bundle #1 comprising the following key activities is shown in

Table 5.33 (below). The scale of the bundle is unlikely to create a Level 7 at

the main activity bundle.


Table 5.33 Main design activities bundle for #H2

No Bundle 3 Building engineering services design: Mechanical and electrical (including, BMS, Security, ICT) and

hydraulics (including wet fire systems) (match: Level 6) 4 Building engineering services design: Highly specialised hydraulics (including medical gases and

pneumatic tubes) (match: Level 6) 5 Building engineering services design: Dry fire systems (match: Level 5)

2. The rest of the design activities remain as separate key activities and contract

directly with government via separate contracts: architectural design; civil and

structural design; kitchen design; landscape design; and traffic engineering and

systems design; except for Activity 9 – planning for building, engineering

services maintenance that is reactive, preventive, specialised and non-

specialised in nature which is included in the main maintenance bundle. Table

5.34 (below) is a summary of the list of separate key design activities.

Table 5.34 List of separate design activities for #H2

No Key activities Design of new installation/construction works 1 Architectural design (concept – from brief to space design; schematic design; developed design; contract

documentation design including building code compliance and review; disabled access compliance; fixture and fittings) (match: Level 5)

2 Civil and structural design (schematic design; developed design) (match: Level 5) 6 Kitchen Design (match: Level 6) 7 Landscape design (match: Level 5) 8 Traffic engineering and systems design (match: Level 5)

3. The main construction bundle #2 (See Table 5.35 below) comprises of all the

key construction activities (10 – 46), including Level 7 activities. As the

construction cost of the project is around $250 million, and there is a sufficient

supply of main contractors available to deliver projects of this size, a Level 7

due to size is unlikely.


Table 5.35 Main construction activities bundle for #H2

No Bundle New installation/construction works 10 Demolition works and clearance (match: Level 5) 11 Site establishment works, including temporary services, site fencing and hoarding, accommodation

(match: Level 5) 12 Excavation works (Bulk and detailed) (match: Level 5) 13 Bored piers (match: Level 5) 14 Concrete; Formwork; and Reinforcement, including beams (match: Level 5) 15 Masonry (match: Level 5) 16 Structural steel work (match: Level 5) 17 Carpentry works, including wall and ceiling construction (studwork, ceiling lining and framing) (match:

Level 5) 18 Metalwork, including balustrade handrails, privacy screens etc. (match: Level 5) 19 Roofing (Corrugated metal roof sheeting) (match: Level 5) 20 Roof safety system (match: Level 5) 21 Glazing and windows (match: Level 5) 22 Doors and door frames (match: Level 5) 23 Automatic door (match: Level 6) 24 Vinyl floor and wall finishes (match: Level 5) 25 Carpet floor finishes (match: Level 5) 26 Ceramic floor/wall tiles (match: Level 5) 27 Façade (match: Level 6) 28 Painting (match: Level 5) 29 Signage and way-finding (match: Level 5) 30 Fixtures, Furnishings & Equipment (FFE) – supplied and install non-specialised items (e.g. whiteboards

and AV equipment) (match: Level 5) 31 Fixtures, Furnishings & Equipment (FFE) – supplied and install specialised items (e.g. sterilising

equipment) (match: Level 7) 32 Commercial kitchen equipment, including dishwasher, ovens and fridges (match: Level 5) 33 Cold room installation (match: Level 5) 34 HVAC (match: Level 6) 35 Electrical, including in ground electrical (match: Level 6) 37 Security, including security cameras, duress alarms, access readers etc. (match: Level 5) 36 Building management services (BMS) (Level 7) 38 Communications (match: Level 6) 39 Hydraulics, including plumbing and drainage (match: Level 5) 40 Hydraulics - Medical gases (such as oxygen gas, breathing air), complete with copper piping, pumps,

compressors and outlets etc. (match: Level 6) 41 Hydraulics – Pneumatic tubes complete with pumps, compressors and outlets etc. (Level 7) 42 Fire services, including fire detectors, fire indicator panel (FIP), Emergency Warning and

Intercommunication System (EWIS), fire extinguishers and blankets (match: Level 6) 43 Lifts (match: Level 6) 44 External works – Hard and soft landscaping (match: Level 5) 45 External works – Carparking (match: Level 5) 46 External works - Covered walkways (match: Level 5)

4. The operations activities, which are mainly BMS and protective or security

services, remain as two key activities and are not bundled together into a main

activity.

5. Main maintenance bundle #3 comprises of the key activities of building,

engineering services maintenance that are reactive, preventive and non-

specialised, and including Activity 9 – planning for maintenance (as

summarised in Table 5.36, below). This bundle can be carried out by a single

contractor and is unlikely to generate a Level 7 due to size.


Table 5.36 Main maintenance activities bundle for #H2

No Bundle 49 Building, engineering services maintenance: Reactive (non-critical/routine/non-specialised) (Mismatch:

Level 3 or 5) (Model recommends Level 5) 51 Building, engineering services maintenance: Reactive (critical/emergency/non-specialised) (Mismatch:

Level 3 or 5) (Model recommends Level 5) 53 Building, engineering services maintenance: Preventive (programmed/non-specialised) (Mismatch: Level 3

or 5) (Model recommends Level 5) Design of performance specification of maintenance

9 Planning for building, engineering services maintenance comprising: Reactive (non-critical/routine/non-specialised); reactive (critical/emergency/non-specialised and critical/emergency/specialised); preventive (programmed/non-specialised and programmed/specialised) (Mismatch: Either Level 3 or 5) (Model recommends Level 5)

The remaining five types of maintenance activities of the building, engineering

services (reactive, preventive and specialised), cleaning, landscaping, and external

works remain as separate key activities, and are listed in Table 5.37 (below).

Table 5.37 List of separate key maintenance activities for #H2

No Key activity 50 Building, engineering services maintenance: Reactive (non-critical/routine/specialised) (match: Levels 6/7) 52 Building, engineering services maintenance: Reactive (critical/emergency/specialised) (match: Levels 6/7) 54 Building, engineering services maintenance: Preventive (programmed/specialised) (match: Levels 6/7 55 Cleaning (Mismatch: Level 3 or 5) (Model recommends Level 5) 56 Landscaping and external works (Mismatch: Level 3 or 5) (Model recommends Level 5)

From the above main bundles, there appears to be a viable DCOM major

activity bundle, which comprises of main activities and key activities, to consider.

That is, first, the DCOM bundle is market is sounded as PPP, then market tested as a

contract, then via government. There would either be one PPP contract, or one

DCOM contract, or 13 contracts direct to government (comprising of the 3 main

contracts and the 10 key activities listed above), with Activity 9 – planning for

maintenance included in the implementation of maintenance.


Consider either the PPP contract or the DCOM contract and re-measure only

the TCE variables with an anticipated typical market firm or consortium at the head

of these contracts or counterparty to government. The scores are given in Table 5.38.

Table 5.38 Exchange relationship analysis of #H2

no Contract Asset specificity


1 DCOM major activity – comprising of all the design, construct, operations and maintenance main activities and key activities

6,6 5,5 4 (++,+,+/0) inefficient to extremely inefficient via a discrete exchange

By matching the patterns to Figure 3.2, the patterns indicate that the exchange

between government and the market firm for the major bundle is inefficient to


extremely inefficient via a discrete exchange, using a bespoke non-standard contract

with significant credible threats. The principal-agent characteristics of this discrete

exchange that connect to an outcome-based exchange (with reference to Table 3.3)

are as follows:


• agency costs

• specification of outcomes • verification of outcomes

• risk premium • suitability for information asymmetry



The model or theoretical procurement strategy is appreciably different to the

actual procurement. That is, both operations and all maintenance activities (including

the planning of maintenance) are indicated by the model as externalised. This is

unlike actual procurement where operations activities are internalised, and a

proportion of the maintenance – comprising of planning maintenance and non-

specialised maintenance as well as cleaning, landscaping and external works

maintenance – is also internalised. These differences indicate a DCOM contract

rather than the Design and Construct contract being reflected in the managing

contract deployed.

The model’s theoretical prediction supports actual procurement to the extent of

all the externalised activities and the bundling of design and construction under the

organisation of an external firm; that is, managing contractor in the actual

procurement.


The above procurement strategy recommends some appreciable improvement

to the actual procurement through the externalisation of operations and maintenance,

and the procurement of a DCOM bundle in terms of either a PPP contract or a

government financed contract. It is expected that these differences in the

procurement identified by the model and actual practice will generate fewer EoIs


than the number generated by the actual procurement approach. The survey of civil

and building contractors showed that there were three building contractors in this

case study state who indicated spare capacity and who were capable of DCOM.

Upon rationalising to 100 percent, approximately seven contractors could have

tendered for the project. This is less than half of the original 15 EoI, and downwards

and within the optimal EoI.

This lower EoI matches Scenario 4 in the refutability procedures in Section

3.3.5; that is, at the sub-optimal competition level of relatively high EoI, if the

empirical procurement is appreciably dissimilar from the model’s prediction, then the

predicted procurement is likely to generate a lower number of EoI, and closer to the

optimal number of 5 to 8 EoI. This provides strong support for the model’s ability to

avoid market failure ex post, and to generate an optimal EoI. Once again, this case is

only one of the four cases conducted, and validation of this conclusion depends on

the outcomes of the hypothesis testing on all four cases. This testing will provide

very strong empirical support for the VfM virtues of the model.

In summary, the results of Case Study #H2 support the hypothesis associated

with the first-order decision-making model, and are consistent with VfM and EoI.

The next section now analyses the development of Health Case Study #H1.

5.6 Health Case Study #H1 (Theoretical procurement)

Health Case Study #H1 had 5 EoI and is within the optimal range of EoI. The

project comprised of the delivery of a new main hospital building, two new buildings

for mental health facilities and the upgrade of four existing buildings. The main

hospital building is over 35 000m2 (including helicopter roof pad access). Case Study

#H1 was delivered as a PPP in a DCOM bundle. Five consortia expressed interest to

tender. Each consortium consisted of the 1. finance and consortium leader; 2. design

and construct main-contractor; and 3. facilities management, maintenance and non-

core support services provider. The EoIs received were evaluated and three were

shortlisted to submit a detailed proposal. The consortium undertook the financing,

design, construction and commissioning of the new hospital and health facilities. The

consortium also undertook the operations and maintenance of the new and existing

hospitals and health facilities for 25 years. This includes: building and FFE


maintenance; and facilities management and delivery of ancillary non-clinical

services. Clinical services were provided by a separate health service provider.

The following sections demonstrate the process and results of the analytical

steps in the first-order decision making model.


A total of 79 key activities were identified in Case Study #H1. These key

activities are similar in nature to the key activities in Case Study #H2 across the

design, construct, operations and maintenance activities. The scope of key design

activities starts with the design of construction works. The two main types of design

activities are: schematic and performance specification design, and fully developed

design and contract documentation. The design of the schematic and functional

relationships is complex and occurs at the planning phase of the project.

The first eight activities are concerned with the schematic design and detailed

performance specifications of new works: 1. architectural design; and 2. civil and

structural design; building engineering services design for the following (3-8)

activities: 3. mechanical and building management control system (BMCS); 4.

electrical (including security and communications); 5. hydraulics (including wet fire

systems and medical gases and pneumatic tubes); 6. dry fire systems; 7. landscape

design; 8. helicopter landing design.

The next set of key activities 9 to 18 are related to the full developed design

and contract documentation design of the above key activities, that is: 9. architectural

design; 10. civil and structural design; 11. building engineering services design; 12.

BMCS; 13. electrical (including security and communications); 14. hydraulics

(including wet fire systems); 15. hydraulics (medical gases and pneumatic tubes); 16.

dry fire systems; 17. landscape design; and 18. helicopter landing design. The last

two design activities are related to the design of performance specification of

operations and maintenance or the planning for maintenance schedule of 19.

operations such as, BMCS, utilities management services, and security services; and

20. maintenance for building, engineering services, comprising of preventive

maintenance that is programmed or both non-specialised and specialised in nature.

A total of 47 key construction activities were identified: 21. site establishment

works (including temporary services, site fencing and hoarding, and site


accommodation); 22. demolition works (including existing buildings and wards, and

existing roadwork and clearance); 23. bulk and detailed excavation works (including

cut-and-fill and retaining walls); 24. concrete, formwork and reinforcement; 25. pre-

stressed floors; 26. structural steelwork; 27. masonry; 28. carpentry works, including

wall and ceiling construction (studwork and framing); 29. plasterboard linings; 30.

operable walls; 31. metalwork (including platforms and walkways, metal stairs,

ladders, pipe railings, balustrades and handrails, fences and screens, guardrails, crash

barriers, loading dock bumpers, and bollards); 32. stainless steelwork (including

theatres and sterile processing); 33. joinery and pathology furniture; 34. roofing and

cladding (including insulation, barriers, and waterproofing); 35. external windows

(aluminium framed and glazed elements with clear anodised or powder-coated

finishes), and glazed curtain walls; 36. doors (external and internal); 37. internal

windows and mirrors; 38. glazed screens; 39. doors and windows (internal and

external hardware); 40. cement floors; 41. epoxy floors; 42. carpet and carpet floor

tiles; 43. ceramic floor and wall tiles; 44. vinyl floor and wall finishes; 45. plaster

walls; 46. suspended ceilings and tiles; 47. plasterglass and plasterboard ceilings; 48.

polyester powder-coated ceilings; 49. painting; 50. signage and way-finding; 51.

curtains and blinds; 52. supply and installation of fixtures, furnishings and fittings

equipment of non-specialised items, such as, whiteboards and AV equipment (either

new or existing – by the state health department or project company); 53. supply and

installation of fixtures, furnishings and fittings equipment of specialised items (again,

either new or existing – by the government or project company); 54. HVAC; 55.

electrical (including lighting and power); 56. BMCS; 57. security (including duress

alarm system); 58. communications; 59. hydraulics (including plumbing and

drainage); 60. hydraulics/medical gases (such as oxygen gas, breathing air), complete

with copper piping, pumps, compressors and outlets; 61. hydraulics/pneumatic tubes

(complete with pumps, compressors and outlets); 62. fire services [including fire

detectors, fire indicator panel (FIP), emergency warning and intercommunication

system (EWIS), fire extinguishers and blankets]; 63. sterilising system; 64. cool

rooms, blood fridges, mortuary cabinets; 65. lifts; 66. landscaping (including

protection and remedial work to existing trees, turfing, planting, and irrigation); 67.

external works/pavements (including roads, parking areas, loading bays, paths, and

hard-standing areas); and lastly, 68. external works/covered ways (terraces,

balconies, decks and structures, balustrades and fencing).


Three key activities are identified in the operational stage of the hospital: 1.

BMCS; 2. utility management services, including maintenance and continuity of

supply and monitoring of electricity, gas, fuel oil, water, sewerage, surface water,

stormwater and in-ground water disposal; and 3. security services.

Lastly, maintenance activities are categorised according to the requirements of

the hospital. Even though the functionality of Case Study #H2 might be different

from that of Case Study #H1, the general maintenance across hospitals can be

categorised in a similar manner, such as: 1. cleaning; 2. landscaping and external

work; and 3. building engineering maintenance services (BEM). BEM can be either

reactive (that is, routine or emergency); or planned (that is, preventive, including

programmed replacement of specialised, non-specialised or highly specialised

equipment). More specifically, BEM services are: 1. reactive and non-specialised; 2.

reactive and specialised; 3. reactive and highly specialised (including BMCS and

pneumatic tubes); 4. planned and non-specialised; 5. planned and specialised; and 6.

planned and highly specialised.

After identifying the key activities across design, construction, operations and

maintenance, the next task is to carry out make-or-buy analysis for each of the

activities identified, using the case study questionnaire.


Given that Case Study #H1 might not be located in the same state as Case

Study #H2, and considering that the decisions were made reflective of the

competencies and capabilities of the market and government in 2006-07, the

responses pertaining to similar activities might be dissimilar and need to be

considered separately. The responses to the case study questionnaire, including the

matching patterns and activity levels of design activities, are summarised in Table

5.39 (below).

Table 5.39 assigns Level 4b to, the schematic design and associated outline

specifications, to the detailed performance specifications of: architectural, structural

and civil, mechanical, BMCS, electrical (including security and communications),

hydraulics (including wet fire systems and medical gases and pneumatic tubes), dry

fire systems, landscape, and helicopter landing design. The full developed design,

including contract documentation of these activities are assigned Level 5, with the


exception of BMCS and specialised hydraulics (which are assigned Level 6), and the

helicopter landing facility (which is assigned Level 7). Although the state

government has indicated a lack of capacity across all design activities, it has a

greater competency for schematic design and performance specifications (as

indicated by the scores of ‘4’) than for the development of complete design and

contract documentation, which is beyond its technical capability (as indicated by

scores ranging from ‘1’ to ‘3’). Across the design activities, the low scores on the

value variable match the low scores on the frequency variable; thus, this represents a

non-typical project in the state department project portfolio.

For Activities 1 to 8 (related to the schematic design and detailed performance

specification), the scoring of uncertainty and asset specificity are high (as indicated

by the scores ‘5’ and ‘6’) because of the highly complex process to liaise and co-

ordinate amongst the various functional groups and stakeholders within the hospital

and government can be held-up by the amount of time required to develop the

schematics and detailed performance specifications. As a result, the overall pattern

results in a Level 4b. These reasons also apply to the performance specification for

operations in Activity 19, and for maintenance in Activity 20.

Of the remaining design activities (9 to 18) that require full developed design

and contract documentation, the design of the helicopter landing facility is the only

activity that has extremely high scores on the RBT variables. This indicates that the

government cannot match the market (at procurement decision date) in the extremely

rare and costly-to-imitate resources and knowledge required for the activity;

therefore, it is assigned Level 7. An SCP analysis could show that there is a limited

supply of firms capable of designing the helicopter landing facility. Activity 12

(BMCS design) and Activity 15 (specialised hydraulics, such as medical gases and

pneumatic tubes) are also dominated by RBT variables but to a lesser extent, as there

are a greater number of firms with the required technology and knowledge for these

provisions. The remaining activities have overall low scores across RBT and TCE

variables, and generate a Level 5 pattern. This reflects that the market is

organisationally better placed than the state government to provide a cheaper

response because of its competitive advantage in aggregating workloads, and its

greater efficiencies and competence.


Table 5.39 Make-or-buy analysis of design of new installation works for #H1

no Activity Capacity Rarity Costly-to-imitate

Asset specificity


1 Architectural design (from schematic and associated outline specification to detailed performance specification)

4 3,3 4,4 6,6 6,6 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

2 Structural and civil design (detailed performance specification)

4 3,1 2,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

3 Building engineering services design (detailed performance specification): Mechanical, BMCS

4 2,3 3,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

4 Building engineering services design (detailed performance specification): Electrical (including Security and Communications)

4 2,2 2,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

5 Building engineering services (detailed performance specification): Hydraulics (including wet fire systems, and medical gases and pneumatic tubes)

4 2,2 2,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

6 Building engineering services design (detailed performance specification): Dry fire systems

4 2,2 2,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

7 Landscape design (detailed performance specification)

4 2,2 2,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

8 Helicopter landing design (detailed performance specification)

3/4 4,4 4,4 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

9 Architectural design (developed design; contract documentation design, including building code compliance and review; disabled access compliance; fixture &fittings)

3 3,3 4,4 2,4 3,3 4 (-, 0,0,0 to +,0,0) 5

10 Structural and civil design (developed design to contract documentation design)

3 3,1 2,2 2,1 2,3 4 (-, 0,0,0 to +,0,0) 5

11 Building engineering services design (developed design to contract documentation design): Mechanical

3 2,3 3,2 2,4 2,4 4 (-, 0,0,0 to +,0,0) 5

12 Building engineering services design (developed design to contract documentation design): BMCS

2 6,6 4,4 4,4 2,3 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

13 Building engineering services design (developed design to contract documentation design): Electrical (including Security and Communications)

3 2,2 2,2 2,4 2,4 4 (-, 0,0,0 to +,0,0) 5

14 Building engineering services (developed design to contract documentation design): Hydraulics (including wet fire systems)

3 2,2 2,2 2,4 2,4 4 (-, 0,0,0 to +,0,0) 5

15 Building engineering services design (developed design to contract documentation design): Hydraulics (medical gases and pneumatic tubes)

2 6,6 3,3 1,2 1,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

16 Building engineering services design (developed design to contract documentation design): Dry fire systems

3 2,2 2,2 2,4 2,4 4 (-, 0,0,0 to +,0,0) 5

17 Landscape design (developed design to contract documentation design)

3 2,2 2,2 2,2 2,1 4 (-, 0,0,0 to +,0,0) 5

18 Helicopter landing design (developed design to contract documentation design)

1 7,7 5,5 4,3 2,2 4 (---, + to +++, + to +++, 0 to +++, 0 to +++, 0)

7

19 Performance specifying for operations (BMCS; utilities management services; security services)

4 2,2 3,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b

20 Performance specifying for building, engineering services maintenance comprising: Preventive (programmed/non-specialised/specialised)

4 2,3 3,2 5,5 5,5 4 (-/+, 0, 0, + to +++, + to +++, 0)

4b


Table 5.40 and Table 5.41 (below) list the scores, matching patterns, and

activity levels for all construction activities.

Table 5.40 Make-or-buy analysis of construction activities 21 to 52 for #H1 no Activity Capacity Rarity Costly-to-

imitate Asset specificity


21 Site establishment works, including temporary services, site fencing and hoarding, accommodation etc.

3 2,2 2,2 2,4 2,3 4 (-,0,0,0 to + ,0,0) 5

22 Demolition works, including existing buildings and wards, and existing roadwork and clearance

3 2,3 2,2 2,4 3,3 4 (-, 0,0,0 to +,0,0) 5

23 Excavation works, including cut and fill and retaining walls

3 2,2 2,2 2,3 2,4 4 (-, 0,0,0 to +,0,0) 5

24 Concrete; formwork; and reinforcement, including beams

3 2,2 2,2 2,3 2,4 4 (-, 0,0,0 to +,0,0) 5

25 Prestressed floors 3 3,3 2,2 2,3 2,4 4 (-, 0,0,0 to +,0,0) 5 26 Structural steel work 3 2,2 2,2 2,3 2,4 4 (-, 0,0,0 to +,0,0) 5 27 Masonry 3 2,2 2,2 2,3 2,4 4 (-, 0,0,0 to +,0,0) 5 28 Carpentry works, including wall and

ceiling construction (studs & framing) 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5

29 Plasterboard linings 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 30 Operable walls 3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0) 5 31 Metalwork, including: platforms;

bollards walkways; metal stairs; ladders; pipe railings, balustrades and handrails, fences and screens; guardrails, crash barriers, loading dock bumpers

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5

32 Stainless steel, including theatres; sterile processing

3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0) 5

33 Joinery and pathology furniture 3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0) 5 34 Roofing and cladding including

insulation and barriers and waterproofing

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5

35 External windows (aluminium framed and glazed elements with clear anodised/or powder-coated finishes) & glazed curtain walls

3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0) 5

36 Doors (external and internal) 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 37 Internal windows and mirrors 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 38 Glazed screens 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 39 Doors and window internal and

external hardware 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5

40 Cement floors 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 41 Epoxy floors 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5 42 Carpet and carpet floor tiles 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5 43 Ceramic floor and wall tiles 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 44 Vinyl floor and wall finishes 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5 45 Plaster walls 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 46 Suspended ceilings and tiles 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 47 Plasterglass and plasterboard ceilings 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 48 Polyester powdercoated ceilings 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5 49 Painting 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 50 Signage and way-finding 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5 51 Curtains and blinds 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0) 5 52 Fixtures, furnishings & fittings

equipment (FF& FE Groups 1; 2; 2T) – provide (new by project company/existing nominated by project company) & provided (new/existing by State Health Department) & install non-specialised items (e.g. whiteboards, AV equipment)

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0) 5


Table 5.41 Make-or-buy analysis for construction activities 53 to 68 for #H1


imitate

Asset specificity


53 Fixtures, Furnishings & Fittings Equipment (FF& FE Groups 1; 2; 2T) – provide (new by Project Company/existing nominated by project Company) and provided (new/existing by State Health Department) and install specialised items

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

54 HVAC 3 2,3 2,2 2,4 2,4 4 (-, 0,0,0 to +,0,0)

5

55 Electrical, including lighting and power

3 2,3 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

56 Building Management Control System (BMCS)

2 5,5 2,2 2,2 2,2 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

57 Security (including Duress Alarm System)

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

58 Communications 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

59 Hydraulics, including plumbing and drainage

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

60 Hydraulics – Medical gases such as oxygen gas, breathing air (complete with copper piping, pumps, compressors and outlets etc.)

3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0)

5

61 Hydraulics – Pneumatic tubes (complete with pumps, compressors and outlets etc.)

2 6,6 2,2 2,2 2,4 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

62 Fire services, including fire detectors, fire indicator panel (FIP), Emergency Warning and Intercommunication System (EWIS), fire extinguishers and blankets

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

63 Sterilising system 3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0)

5

64 Cool rooms, blood fridges, mortuary cabinets

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

65 Lifts 3 4,4 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

66 Landscaping, including protection and remedial work to existing trees; turfing; planting; irrigation

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

67 External works – Pavements, including roads, parking areas, loading bays, paths, and hard-standing areas

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

68 External works – Covered ways; terraces; balconies; decks and structures; balustrades and fencing

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

With the exception of the two Level 6 activities – Activity 56 (BMCS) and

Activity 61 (specialised hydraulics), Table 5.40 and Table 5.41 assign all

construction activities a Level 5. Overall, the Level 5 activities have low scores on

all TCE and RBT variables. This reflects that the market is organisationally better or


better positioned than the government in terms of competence, greater efficiencies

and economies of scale. In contrast to these Level 5 activities, BMCS and specialised

hydraulics require specialised knowledge and proprietary technology that are less

easily sourced in the market. Thus, these activities display a higher score on the

rarity variable.

Table 5.42 (below) lists the scores, matching patterns and activity levels for the

operation activities; that is, BMCS, utility management services, and security

services; these activities are also assigned a Level 5. Similar to Case Study #H2, the

operation of these services requires little training and they can thus be organised or

managed either internally or externally. However, given that the frequency variable

takes precedence in theory, the model proposes that the activity be externalised as

Level 5, which is in line with actual procurement.

Table 5.42 Make-or-buy analysis of the implementation of operations for #H1


imitate

Asset specificity


69 Building Management Control System (BMCS)

3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

70 Utility management services (including maintenance, continuity of supply and monitoring of electricity, gas, fuel oil, water, sewerage, surface water, stormwater; and in-ground water disposal)

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

71 Security services 3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

Table 5.43 summarises the scores, matching patterns and corresponding

activity levels for maintenance activities. In Table 5.43, all these activities for Case

Study #H1 are assigned Level 5, with the exception of Activities 74 and 77 that have

been assigned a Level 6. The latter assignation is due to the fact that the rarity of the

specialised knowledge and technology associated with BMCS and specialised

hydraulics makes their maintenance, beyond the capacity of the state department to

provide.


Table 5.43 Make-or-buy analysis of the implementation of maintenance for #H1

no Activity Capacity Rarity Costly-to-imitate

Asset specificity


72 Building, engineering services maintenance: Reactive (non-critical/routine and critical/emergency - non-specialised)

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

73 Building, engineering services maintenance: Reactive (non-critical/routine and critical/emergency - specialised)

3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0)

5

74 Building, engineering services maintenance: Reactive (non-critical/routine and critical/emergency - highly specialised: BMCS and Pneumatic tubes)

2 6,6 2,2 2,2 2,4 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

75 Building, engineering services maintenance: Planned and Preventative including programmed replacement (non-specialised)

3 2,2 2,2 2,2 2,3 4 (-, 0,0,0 to +,0,0)

5

76 Building, engineering services maintenance: Planned and Preventative including programmed replacement (specialised)

3 2,2 2,2 2,2 2,4 4 (-, 0,0,0 to +,0,0)

5

77 Building, engineering services maintenance: Planned and Preventative including programmed replacement (highly specialised: BMCS and Pneumatic tubes)

2 6,6 2,2 2,2 2,4 4 (--, + to +++, 0, 0 to ++, 0 to ++, 0)

6

78 Cleaning 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

79 Landscaping and external works 3 2,2 2,2 2,2 2,2 4 (-, 0,0,0 to +,0,0)

5

In contrast to Case Study #H2, in Case Study #H1 there are no activities with a

mismatch in the value and frequency variables; furthermore, all the activities are

assigned Level 5, with the exception of the activities listed in Table 5.44 (Below).

Table 5.44 Summary of Levels 4b, 6 and 7 activities

No Key activity Design of new installation/construction works 1 Architectural design (from schematic and associated outline specification to detailed performance specification) (Level 4b) 2 Structural and civil design (detailed performance specification) (Level 4b) 3 Building engineering services design (detailed performance specification): Mechanical and BMCS (Level 4b) 4 Building engineering services design (detailed performance specification): Electrical (including Security and

Communications) (Level 4b) 5 Building engineering services (detailed performance specification): Hydraulics (including wet fire systems and medical

gases and pneumatic tubes) (Level 4b) 6 Building engineering services design (detailed performance specification): Dry fire systems (Level 4b) 7 Landscape design (detailed performance specification) (Level 4b) 8 Helicopter landing design (detailed performance specification) (Level 4b) 12 Building engineering services design (developed design to contract documentation design): BMCS (Level 6) 15 Building engineering services design (developed design to contract documentation design): Hydraulics (medical gases and

pneumatic tubes) (Level 6) 18 Helicopter landing design (developed design to contract documentation design) (Level 7) Design of performance specification of operations and maintenance 19 Performance specifying for operations (BMCS; utilities management services; security services) (Level 4b) 20 Performance specifying for building, engineering services maintenance comprising: Preventive (programmed/non-

specialised and specialised) (Level 4b) New installation/ construction works 56 Building Management Control System (BMCS) (Level 4b) 61 Hydraulics – Pneumatic tubes complete with pumps, compressors and outlets etc. (Level 6) Implementation of maintenance 74 Building, engineering services maintenance: Reactive (non-critical/routine and critical/emergency - highly specialised:

BMCS and Pneumatic tubes) (Level 6) 77 Building, engineering services maintenance: Planned and Preventative, including programmed replacement (highly

specialised: BMCS and Pneumatic tubes) (Level 6)



The above task shows a mixture of Levels 5, 6 and 7 activities. In terms of

corroborating with the levels identified, there are numerous firms in the market for

Level 5 activities and several firms for Level 6 activities; therefore, there is no

indication of a monopoly or duopoly market structure. While there is one Level 7

activity (the design of the helicopter landing facility), it is not considered to be a

result of the scale of the key activity but of the rare and costly-to-imitate technology

required for the activity. Hence, no further action is required to be taken in this

regard.

5.6.4 Stage 2 Bundling analysis

In this task, the focus is now on second-order issues concerning the

organisation and management across key activities and, in particular, on troublesome

Level 4b and Level 7 (arising from technology) activities. The specific focus is to

separate the ten Level 4b activities and one Level 7 activity (arising out of rare

technology) that have been identified, from any bundles related to the organisation

and management across key activities. These activities are listed in Table 5.45.

Table 5.45 List of Level 4b and Level 7 design activities for #H1

No Key activity Schematic and associated outline specification to detailed performance specification

1 Architectural design (Level 4b) 2 Structural and civil design (detailed performance specification) (Level 4b) 3 Building engineering services design (detailed performance specification): Mechanical and BMCS (Level 4b) 4 Building engineering services design (detailed performance specification): Electrical (including Security and

Communications) (Level 4b) 5 Building engineering services (detailed performance specification): Hydraulics (including wet fire systems

and medical gases and pneumatic tubes) (Level 4b) 6 Building engineering services design (detailed performance specification): Dry fire systems (Level 4b) 7 Landscape design (detailed performance specification) (Level 4b) 8 Helicopter landing design (detailed performance specification) (Level 4b) Design of performance specification of operations and maintenance

19 Performance specifying for operations (BMCS; utilities management services; security services) (Level 4b) 20 Performance specifying for building, engineering services maintenance: Preventive (programmed/non-

specialist and specialist) (Level 4b) 18 Helicopter landing design (developed design to contract documentation design) (Level 7)

There are ten Level 4b activities which concern the development of detailed

performance specification of the facilities and functional relationships, and which

involve a complex planning period (involving investigation and user engagement in

design evaluation). In terms of addressing opportunistic behaviour, it is restricted

in this case as the government is required to tender each contract in accordance

with tendering guidelines and this limits its ability to negotiate with single firms.

As these are early and upstream activities in which the government’s internal


resources are heavily involved ( including user and coordination activity by the

Health Department concerned), the model recommends the exclusion of these Level

4b activities from any subsequent bundle that includes construction.

There is one Level 7 activity, which is related to the helicopter landing

developed design. As is the case for the (above) ten Level 4b activities, there is

likely to be little checking of opportunistic behaviour by the government or main

contractor for this activity, given that there is an extreme lack of market supply

associated with the developed design. In contrast with the above activities, the key

coordination activity involves construction, and it is recommended that it be

included with main design activity and coordinated by the contractor. However, it

is also recommended that the government establish a special relationship (for

example, prime cost sum or nomination arrangement) with the firm developing the

design.

Consider the opportunities to bundle key externalised activities into main

activities, and main activities into major activities. In so doing, consider each

main activity and any major activity in terms of whether a higher level of bundling

could generate a Level 7. In summary, this gives:

1. One main design bundle for Case Study #H1; that is, one main activity for the

developed design and contract documentation of building engineering services in

Design Bundle #1. The scale of the bundle is unlikely to create a Level 7 at the

main activity bundle.

Table 5.46 Main Design Activity Bundle #1 for #H1

No Bundle 11 Building engineering services design (developed design to contract documentation design): Mechanical (Level 5) 12 Building engineering services design (developed design to contract documentation design): BMCS (Level 6) 13 Building engineering services design (developed design to contract documentation design): Electrical (including

Security and Communications) (Level 5) 14 Building engineering services (developed design to contract documentation design): Hydraulics (including wet fire

systems) (Level 5) 15 Building engineering services design (developed design to contract documentation design): Hydraulics (medical gases

and pneumatic tubes) (Level 6) 16 Building engineering services design (developed design to contract documentation design): Dry fire systems (Level 5)

The rest of the design activities remain as key activities; namely, the ten Level

4b key activities 1-8, 19 and 20 (related to detailed performance specification), and

the four developed design activities 9, 10, 17 and 18 (developed design related to

architectural, civil and structural, landscape, and helicopter landing) remain as

separate key activities. They are summarised in Table 5.47.


Table 5.47 List of individual key design activities for #H1

No Key activities 1 Architectural design (Schematic and associated outline specification to detailed performance specification)

(Level 4b) 2 Structural and civil design (detailed performance specification) (Level 4b) 3 Building engineering services design (detailed performance specification): Mechanical and BMCS (Level 4b) 4 Building engineering services design (detailed performance specification): Electrical (including Security and

Communications) (Level 4b) 5 Building engineering services (detailed performance specification): Hydraulics (including wet fire systems

and medical gases and pneumatic tubes) (Level 4b) 6 Building engineering services design (detailed performance specification): Dry fire systems (Level 4b) 7 Landscape design (detailed performance specification) (Level 4b) 8 Helicopter landing design (detailed performance specification) (Level 4b) Design of performance specification of operations and maintenance

19 Performance specifying for operations (BMCS; utilities management services; security services) (Level 4b) 20 Performance specifying for building, engineering services maintenance: Preventive (programmed/non-

specialist and specialist) ) (Level 4b) 9 Architectural design (developed design; contract documentation design, including building code compliance

and review; disabled access compliance; fixture and fittings) (Level 5) 10 Structural and civil design (developed design to contract documentation design) (Level 5) 17 Landscape design (developed design to contract documentation design) (Level 5) 18 Helicopter landing design (developed design to contract documentation design) (Level 7)

2. The main construction bundle #2 comprises of key activities 21-68. As the

construction cost of the project is around $250 million, and there is a sufficient

supply of main contractors available to deliver projects of this size, a Level 7

(due to size) is unlikely.

3. The main operations bundle #3 includes Activity 69 – BMCS, Activity 70 –

utility management services, and Activity 71 – security services.

4. The main maintenance bundle #4 comprises of the key activities of building,

engineering services maintenance (maintenance activities that are reactive,

preventive, specialised and non-specialised; and cleaning, and landscaping and

external works maintenance), as shown in Table 5.48 (below). This bundle is

unlikely to generate a Level 7 due to size.

Table 5.48 Main maintenance activities bundle for #H1 No Bundle 69 Building Management Control System (BMCS) (Level 5) 70 Utility management services (including maintenance and continuity of supply and monitoring of electricity,

gas, fuel oil, water, sewerage, surface water, stormwater; and disposal of in-ground water) (Level 5) 71 Security services (Level 5) 72 Building, engineering services maintenance: Reactive (non-critical/routine, and critical/emergency - non-

specialised) (Level 5) 73 Building, engineering services maintenance: Reactive (non-critical/routine, and critical/emergency -

specialised) (Level 5) 74 Building, engineering services maintenance: Reactive (non-critical/routine, and critical/emergency - highly

specialised BMCS and Pneumatic tubes) (Level 6) 75 Building, engineering services maintenance: Planned and Preventative, including programmed replacement

(non-specialised) (Level 5) 76 Building, engineering services maintenance: Planned and Preventative, including programmed replacement

(specialised) (Level 5) 77 Building, engineering services maintenance: Planned and Preventative, including programmed replacement

(highly specialised: BMCS and Pneumatic tubes) (Level 6) 78 Cleaning (Level 5) 79 Landscaping and external works (Level 5)


The above indicates a viable DCOM major activity bundle, comprising of the

DCOM Main Activities #1 to #4 and four separate key activities; that is: Activity 9 –

architectural developed design, Activity 10 – civil and structural developed design,

Activity 17 – landscape developed design, and Activity 18 – helicopter landing

developed design (excluding the ten Level 4b activities of detailed performance

specifications from the major DCOM bundle).

From the survey of civil and building contractors, it was found that there were

three building contractors in this case study state with spare capacity and capability

to undertake DCOM. Upon rationalising to 100 percent, approximately 5 contractors

could have tendered for the project. This happened to coincide with the actual EoI in

the case study, and indicates that all the contractors with spare capacity had

expressed interest for the project and generated the optimal (or 5) EoI as in this case

study. This shows that bundling the design and construct with operations and

maintenance is unlikely to create a Level 7 due to size. In summary, beyond the 10

Level 4b detailed performance key activities, in order of preference:

• One PPP contract that equals the DCOM major bundle (comprising four main

activities and four key developed design activities); if this does not attract

sufficient EoI, then

• One DCOM contract (government funded) that equals the DCOM major bundle,

comprising four main activities and four key developed design activities; if this

does not attract sufficient EoI, then

• Four main activity contracts four developed design key activity contracts – all

government funded, and totalling eight separate contracts direct to government.


Consider either the PPP contract or the DCOM contract and re-measure only

the TCE variables, with an anticipated typical market firm or consortium at the head

of these contracts or counterparty to government. The scores are given in Table 5.49.

Table 5.49 Exchange relationship analysis for #H1

No Contract Asset specificity


1 DCOM major activity – comprising of all the design, construct, operations and maintenance main and key activities

6,6 5,5 4 (++,+,+/0) Inefficient to extremely inefficient discrete exchange using bespoke non-standard contract with significant credible threats


When matching the patterns in Figure 3.2, the patterns indicate that the

exchange between government and the market firm for the major bundle should be

‘inefficient’ to ‘extremely inefficient’ via a discrete exchange, using a bespoke non-

standard contract with significant credible threats. By again referring to Table 3.3,

the principal-agent characteristics of this discrete exchange that connect to an

outcome-based exchange are as follows:


� Agency costs � Specification of outcomes � Verification of outcomes � Risk premium � Suitability for information asymmetry � Outcome certainty � Better goal alignment


The first-order decision-making model’s theoretical procurement strategy for

Case Study #H1 is very close to the actual procurement via a PPP approach and,

therefore, indicates that Case Study #H1 is being efficiently procured. Coincidentally,

Case Study #H1 is of similar size and scope to Case Study #H2, and the

recommendations for Case Study #H1 are similar to those for Case Study #H2. This

again provides support for the model’s recommendation in Case Study #H1. This

recommendation can be further validated by ongoing monitoring in this case study in

terms of the approach to externalising some of the operations and maintenance

activities, instead of the original internalisation. There can also be further on-going

monitoring of other projects with regard to the relative merits of internalisation and

externalisation of the activities associated with developing the full detailed

performance specification. This case is the last of the four cases conducted to test the

hypothesis associated with the first-order decision-making model.


Given the close match in actual competition with optimal EoI; and the very

close match between recommended procurement strategy and actual practice (that is,

internalisation and externalisation of each activity, the bundling of DCOM activities

in one major bundle and the nature of exchange relationship of each contract), the


results of case study analysis of Case study #H1 match with Scenario 1 of the

refutability procedures in Section 3.3.5. This outcome supports the procurement-

competition/flexibility hypothesis and that Case Study #H1 has been efficiently

procured in terms of VfM in whole-life terms. It represents the predictive strength of

the model in deriving a procurement approach in pursuance of VfM in whole-life

terms, and indicates that empirical research supports theory.

The results of the four case studies indicate that theoretical and literal

replication has been achieved. The extent to which the predicted procurement

strategy supports or contradicts the hypothesis, is considered in the following section.

5.7 Overall results of hypothesis testing

The above case study analyses provide very strong evidence of analytical

generalisation; and demonstrates the economic or VfM benefits of using the model

either backwards to review past projects and/or forwards to guide decision-making in

procurement. As seen in Table 5.50, all of the completed case studies support the

hypothesis. As expected, the procurement model predicts a similar procurement

approach in Case Study #H1 and Case Study #R1, which are at optimum EoI, and

proposes a different procurement approach from practice at low or high level EoI in

Case Study #H2 and Case Study #R2.

Table 5.50 Testing the procurement-competition/flexibility hypothesis

Case #

Box A Actual Procurement

Box B Theoretical Procurement

Box A and B Match or Mismatch

Box C Optimum EoI

Box D Actual EoI

Box C and D Match or Mismatch

Hypothesis Supported Scenarios (1 and 4) or Not Supported (Scenarios 2 and 3)

R1 CO $50-100 m

See Section 5.3.6

Match Optimal Optimal (8 EoI)

Match Supported (Scenario 1)

R2 AC $250-500 m

See Section 5.4.6

Mismatch Potential 8 EoI

Sub-optimal (Low- 2 EoI)

Mismatch Supported (Scenario 4)

H1 PPP $250-500 m

See Section 5.6.6

Match Optimal Optimal (5 EoI)

Match Supported (Scenario 1)

H2 MC $250-500 m

See Section 5.5.6

Mismatch Potential 7 EoI

Sub-optimal (High – 15 EoI)

Mismatch Supported (Scenario 4)

In both Case Study #R1 and Case Study #H1, the close match of theoretical

procurement with actual procurement at optimal level of EoI has been replicated in

two different and distinct sectors, and this provides strong support for the first-order

procurement decision-making model in terms of the economic or VfM benefits of

using the model.


Furthermore, based on theoretical replication, the model is tested at sub-

optimal levels of high EoI in Case Study #H2, and low EoI in Case Study #R2. These

results indicate further evidence of the validity of the model. More specifically, in

Case Study #H2, the model derives a procurement strategy that bundles design,

construction, operations and maintenance and which could have been potentially a

PPP. Based on the data from the questionnaire survey of civil and building

contractors, this approach would most likely give an appreciably reduced number of

EoI – downwards and within the optimal (5-8) EoI range from the higher, actual

EoI.

In Case Study #R2, an Alliance major road project, the procurement model

derives a procurement strategy that comprises four contracts, rather than the one

overall Alliance contract. The four contracts proposed by the model comprise of two

design contracts and two construction contracts. The design contract for the driven

tunnel includes a bespoke agreement which incorporates strict penalties for non-

performance, and the contract for the remainder of the design (largely routine civil

engineering design) uses a conventional consultant agreement that is fixed-fee or

fixed-price. The construct only contract for the regular on-grade works and routine

evaluated structures is a standard fixed-price contract, and the one for the driven

tunnel and cut-and-cover tunnels (under rail, and involving existing highway

realignment) uses an alliance agreement. Again, based on data from the questionnaire

survey of civil and building contractors, this approach will most likely give an

appreciably increased number of EoI – upwards and within the 5-8 optimal EoI

range from the low number of actual EoI.

When reflecting on the theoretical underpinning of the model, the literal

replication of the two sub-optimal cases – Case Study #R2 and Case Study #H2 –

provides further indication that the model economises on transaction costs and

produces an efficient configuration of bundles, which is attractive to the private

sector and can achieve an EoI number closer to the optimal EoI.

5.8 Discussion of findings

On the basis of avoiding market failure at high and low EoI (Please refer to

Sections 3.3.2 and 3.3.3 for discussion in relation to VfM), and having established

replication and analytical generalisation at optimal and sub-optimal EoI, the


hypothesis can now be applied to the sample of 79 major health and road projects

(eight projects of the 87 projects have missing EoI data).

For 34 (or 43 percent) of the projects with optimal EoI (5 to 8 EoI) in the

sample, actual procurement is perceived to indicate VfM, and the model is expected

to derive a procurement strategy that would have (at least substantially) matched the

actual procurement mode. With regards to the remaining 45 (or 57 percent of)

projects with sub-optimal EoIs, the actual procurement can be associated with market

failure, and the model is expected to derive a mismatching procurement approach

with actual procurement, and generate optimal EoI. In other words, only 43 percent

of the sample of projects can be inferred as having developed an efficacious

approach to procurement, and it is suggested that there is appreciable room for VfM

improvement in the remaining 57 percent. In other words, based on the sample of 79

major health and road projects, the new model (which includes the determination of a

PPP or non-PPP approach) is expected to approximately double the chances of the

procurement approach setting a project on the path to delivering superior VfM.

In addition to checking the efficacy of procurement approaches in current

practice in the sample of road and health projects in this research, the findings from

the analysis of four case studies corroborate a logical expectation that procurement

patterns emerging from the use of the new model are likely to differ appreciably

from the procurement patterns observed in the sample. For example, the model is

expected to lead to:

1. A greater proportion of smaller projects (up to $800 million)

Notwithstanding changes in the supply, EoI displays a dramatic decline in

projects over $800 million. Thus, a greater number of smaller projects might be

expected using the new model on the basis that the first-order approach prevents

unduly over-sizing the project, and unduly allocating the project as a single contract.

This expected difference concerning smaller projects, is illustrated by the fact that

the two cases matched by the procurement model (Case Study #R1 and Case Study

#H1) are both less than $800 million. Furthermore, in one of the mismatching cases

(#R2) – a single alliance contract over $250 million – the procurement model

recommends four main contracts, including two design contracts and two construct

only contracts. The latter comprise of one contract for the construction of the regular


on-grade road and elevated structures, and one contract for the inherently uncertain

works for the driven tunnel and cut-and-cover tunnels.

In other words, while mindful of supply, the first-order model promotes a

search for the right size of project and, where appropriate, suggests multiple

contracts with multiple exchange relationships. As a consequence, it is conceivable

that this could mean more than one PPP contract in a project.

2. Increasing rationalisation of procurement across sectors

In health projects, there are key learning economies and economies of scale

arising from the network which is represented by the BMS and bounded by the

buildings’ foot-print. This affords the market advantages in design, construction,

operations and maintenance associated with highly specialised services. Therefore,

for health projects, the procurement model is likely to reveal greater scope to bundle

operations and maintenance with design and construction. It follows, then, that the

procurement model is also likely to prompt a greater number of smaller health

projects to be considered as PPPs. This expected difference in greater scope to

bundle operations and maintenance is again illustrated in the mismatching Case

Study #H2 where a viable design, construction, operations and maintenance bundle is

identified and could have been market-sounded as a potential PPP.

In contrast to health projects, key learning economies and economies of scale

arising from the network of roads affords government advantages, and particularly if

different levels of government (say, state and local government) collaborate with

respect to widely disseminated technology in routine, programmed, and emergency

maintenance to the vast majority of the road network in a state. In this case, for

conventional on-grade road and straightforward elevated structures, the procurement

model is more likely to indicate less scope for bundling operations and maintenance

with design and construction. However, the implications of this outcome would vary

across the country according to the status quo/preferences of particular states. For

example:

• For some states, the status quo is to largely internalise road maintenance

• Some states are contemplating shifting road maintenance to externalised

provision to resist this change


• If states have recently shifted road maintenance to externalised provision, the

model is likely to indicate reversing this change

• If states have begun with internalisation, shifted to externalisation, and now

employ a mix of internalisation and externalisation, the model is likely to

encourage a return to substantial internalisation

These expected prescriptions are once again illustrated in the two road case

studies in which the model supports the internalisation of operations and

maintenance. Moreover, the state of Western Australia provides experiential

evidence to corroborate the theory underlying the procurement model, and empirical

evidence to test the model through its experience with alternative delivery of

maintenance. It first employed an internalised approach, then an externalised

approach, then a mixed internalised and externalised delivery (Alliancing). However,

the state is now on the path back to internalisation.

However, this does not imply that the procurement model will not identify

opportunities for PPPs in the road sector. It does indicate, however, that the model is

expected to surface these opportunities in road sector projects that are far removed

from regular road and bridge structures. In other words, the model is likely to

promote the use of PPPs in structures that are very large and complex, and which

have a much higher percentage of their total cost represented by operations and

maintenance costs than is the case for routine roads and bridges. In the larger, more

complex projects, the market is afforded advantages by virtue of its specialised

knowledge of these complex structures, and by the opportunities presented to effect

design innovations in pursuance of significantly reducing whole-life costs.

By directing this increasing rationalisation of procurement across sectors, we

could see less reliance on stereotypical procurement that tends to create incentives to

minimise capital costs and/or minimise time taken to reach the opening day of the

asset (as indicated in Figure 5.4); for example, DC, MC, and ECI. In terms of

Alliancing, the model might be less likely to promote reliance on this mode for the

entire project. Rather, the procurement model is likely to display more discretion in

deploying Alliancing for some parts of the project only, in an efficient multiple

contract approach, as was the outcome in Case Study #R2.


Figure 5.4: Increasing rationalisation of procurement

3. More time for planning and design development

More fundamentally, the model is likely to guide these changes on the basis of

cost and benefit improvements derived from more time for planning and design

development and which includes briefing stage (Tang, Shen, Skitmore, & Cheng,

2013). In health sector projects, by directing this increasing rationalisation of

procurement across sectors, we are likely to see or confirm a shift towards fully

developing the performance specification with extremely careful planning and

scoping or future- proofing by front-line personnel, with a view to inviting DCOM

EoI and bids (including PPP EoI and bids). Similarly, in conventional road and

bridge structures, there is room in more projects to take the design much further

including to the development of working drawings, before inviting EoI and bids.

In both health and road sectors, although different levels of design are being

advocated, the market is being allowed to work as efficiently as possible in the

provision of a fixed-price, and in the interests of whole-life VfM. This expected

difference of more time for planning and design development was seen in the four

case studies, where the two cases matched by the procurement model involved a

fixed-price or price in advance for the works based on full design completed prior to

the award of the construction contract (in Case Study #R1), and substantial planning

and design development completed prior to the award of the PPP contract (in Case

Study #H1).


For example Construct -Only

For example DCOM (Including PPPs)

Bundling Analysis (2nd order – Property Rights Theory)

(e.g. Design and Construct; Managing Contractor; Early Contractor Involvement)

Higher levels of competitive tension in establishment of contract sum/budget from firms selected from 5-8 EoI

Higher levels of competitive tension in establishment of contract sum/budget from firms selected from 5-8 EoI

Exchange Analysis (3rd order – Transaction Cost Economics and Principal Agent Theory)

Ranging from Traditional/arms-length exchange

To Relational exchange (including aspects of Alliancing)

Long-term Incentives for VfM

Short-term Incentives for VfM

Government advantages (scale economies and learning economies)

Market advantages (scale economies and/or learning economies)

Size (1st order – Activities; Make-or-Buy; Market Analysis)

(Resource-Based Theory and Transaction Cost Theories)

Long-term Incentives for VfM


More generally, however, it is expected that the model’s assumptions

concerning the allowance of more time for planning and design development are

realistic and will hold on the basis (at least to-date) that governments are willing to

contemplate a PPP in major projects despite the lengthy process involved, including

the development of PSC to financial close. The strong research on comparing PPPs

to non-PPPs to-date shows that the former have advantages in their lack of variance

in time and cost to opening day, but not in their minimum time to opening day

(Raisbeck, Duffield and Xu, 2010). Furthermore, recent reforms to Victoria’s PPP

policy (2 May 2013) include the proposal to use the principles of PPPs in smaller

projects. This proposal reflects the relevance of the model’s recommendation of

smaller PPP projects.

At the same time, however, given the inconsistent trinity of time, cost and

quality (Ive and Chang 2007), the key weakness of the procurement model is that it

relegates minimum time from schematic to the completion of construction works.

Clearly, there will always be projects in which time to completion of construction is

an imperative; in these cases, the role of the procurement model is to highlight the

differences between the procurement approach being adopted to deliver the project

(that is, in minimum time) and the procurement approach suggested by the model

(that is, in pursuance of superior VfM). In this way, a better feel and estimate can be

made of the premium that is being afforded to the project by virtue of differences

being highlighted.

5.9 Summary

The possible appreciable differences in procurement patterns arising from the

new model, and in contrast to those procurement patterns created by MAUA, are

logically grounded in: i) a more sophisticated procurement selection yielded by

deploying Nobel Prize-winning theories and dominant Strategic Management theory;

and ii) a procurement model that is wholly economic, with a long-term orientation to

VfM in relative terms, and which considers the entire process starting from design

through to operations and maintenance.

The model addresses the weaknesses of current practice by resisting non-

economic influences (including any short-term influence from political and/or

financial sources highlighted, as discussed in Section 1.3.1), and by addressing the


issues of competition (affected by size) and flexibility (affected by predictability) in

determining the suitability of PPP.

The next chapter summarises the study’s contributions to theory and research

methods, explores its practical implications, and makes recommendations for further

research.

Chapter 6: Conclusions 255

Chapter 6: Conclusions

6.1 Introduction

This research has developed a procurement model that explains the effect of

key dimensions of procurement on competition and flexibility, and which is able to

determine (in a novel way) the suitability or otherwise of PPPs. Fundamentally, the

model resists any short-term political or financial influences, and provides a more

transparent and scientific approach to procurement selection based on economic

analysis. This research into the development, deployment and testing of this first-

order decision-making model (in the context of major road and health infrastructure

in Australia) has implications for procurement practice and theory, and for research

methods.

This concluding chapter begins with a succinct summary of the entire research

framework. It then highlights the scope and applicability of the research

contributions to the theory underpinning the procurement model (Research

Objectives 1-3) and research methods (Research Objective 4). It then concludes by

outlining the research implications for practice, the limitations of the model, and

recommendations for future research. First, however, the chapter provides a

summary of the research framework.

In contrast to current practice, the procurement model analyses a project’s key

design, construction, operations and maintenance at the activity (or first-order) level.

As seen in Figure 6.1 (below), key activities in a project are measured in accordance

with two (Nobel Prize-winning) theories from NIE (TC and TCE theories relating to

key aspects of procurement), and the dominant theory from the field of Strategic

Management relating to procurement (RBT). These theories are configured within a

set of procedures in the procurement model. At the core of these procedures is an

integration of the three theories: the integrative framework of vertical integration

which has been adopted to measure the transaction characteristics, and the relative

capabilities and competencies of government and market surrounding these

activities. This framework then informs the procurement strategy by delineating three

key procurement dimensions: 1. size of project, 2. extent of bundling, and 3. the

256 Chapter 6: Conclusions

nature of the exchange relationship (including the method of remuneration) between

the government and private sector firm at the head of each project bundle or contract.

Figure 6.1: Summary of research framework

Experiential evidence supports the importance of these three key procurement

dimensions. Based on substantial experiential evidence, the UK House of Lords

Select Committee on Economic Affairs (2010) for example, concludes that

competition (affected by project size) and flexibility (affected by predictability) are

two key drivers in determining the suitability of PPPs to deliver VfM. Size is the first

key procurement dimension informed by the procurement model in this research;

furthermore, the level of bundling – the second dimension studied – also affects

project size. The exchange relationship – the third key procurement dimension

informed by the procurement model – directly speaks to the ability of the buyer or

government to cope with any lack of predictability, and to exercise flexibility post-

contract.

The approach to testing the validity and reliability of the procurement model in

pursuance of VfM (Points 1 to 3 in Figure 6.1) revolves around EoI. In terms of

validity, the avoidance of market failure (both pre-contract and post-contract) is a

key rationale for any government. It can be connected to the competition and

flexibility drivers mentioned by the House of Lords Select Committee report on PPPs

(noted above and see the relationship between Points 2 and 3 in Figure 6.1). Thus,

the use of EoI in this research to test whether the procurement mode (or

configuration of the three procurement dimensions) increases the chance of the

project being on a path to achieving superior VfM (Point 3 relationship in Figure

6.1).

Nobel Prize winning theories

New first-order procurement decision-making model


Size Bundling Exchange

Procurement strategy

EoI Avoidance of market failure (Pre-contract and post-contract)

VfM (starting on path toward superior ratio of costs and benefits in whole-life terms)

Point 1

Point 2

Point 3

Transaction characteristics

Government and market

capabilities and competence

Key activities

Project schematic


On the one hand, low EoI and the possibility of pre-contract market failure

(akin to monopoly supply) has a negative effect on production costs because of a

lack of downward pressure on price or capital costs, and a lack of incentive to

incorporate innovations in design (that is, in bids that include design); the latter, in

turn, negatively affect whole-life costs and performance. On the other hand, high EoI

and the possibility of post-contract market failure (again, monopoly supply and

bilateral trade) can lead to negative opportunistic behaviour (such as the

appropriation of monies and/or more time and/or better terms) by the supplier or

consortium in the event of a change or variation in the works.

More specifically, high EoI (over 8) is empirically shown (Gupta, 2002;

Skitmore, 2002) to yield little production benefits in terms of lower prices and lower

incentives for design innovations and, at the same time high EoI can indicate the

prospect of market failure ex post or post-contract, due to potential negative

opportunistic behaviour as a result of lack of flexibility (Sweeney, 2009; Williamson,

1985). On the other hand, low EoI (4 or less) are not sufficient to avoid oligopoly

pricing constraints and resultant ineffective competition (Beattie, Goodacre, &

Fearnley, 2003; Selten, 1973; Shepherd, 1982), and can indicate market failure ex

ante or pre-contract, including as a result of the issue of size or high level of

bundling. Indeed, in the determination of low levels of competition, Selten (1973)

shows that 5 competitors represent the dividing line between few and many, when

modelled as moves in a non-cooperative game and pertaining to a bidding scenario.

For these reasons, 5-8 (inclusive) EoI is derived as an optimal level of competition,

and consistent with VfM for the purposes of this research.

Having established the validity (and optimal range) of EoI as an indicator of,

and a checking point for whether the procurement mode (incorporating the three key

procurement dimensions) is setting the project on a whole-life path towards

delivering superior VfM, the survey of major road and health projects then seeks to

test the reliability of EoIs in testing the procurement model by assessing whether

there is at least a statistically significant relationship between EoI and each of the

three key procurement dimensions (Point 1 relationship in Figure 6.1). The empirical

results support the effects of procurement on EoI, and also support EoI as a valid and

reliable test of the procurement model. Thus, the following hypothesis was adopted:


Actual competition is expected to be in the optimum range of competition (5 to 8 EoI) in cases in which actual procurement matches the theoretical procurement (informed by the first-order procurement decision-making model) and outside the optimum range in cases where there is an appreciable mismatch between actual procurement and the theoretical procurement.

In order to test this hypothesis, and in pursuance of analytical generalisation

(Yin, 2009), four major infrastructure project case studies were selected.

Additionally, two very different sectors were chosen to demonstrate the flexibility

and transferability of the model. All four case studies gave strong and collective

support to the procurement approaches proposed by the first-order procurement

decision-making model. This very strongly demonstrates the efficacy of the model.

The survey of civil and building contractors provided data representative of the

supply-side or market in the five year period of the sample from which the four case

studies were selected. This survey was used to estimate the likely EoI in sub-optimal

EoI Case Study #R2 and Case Study #H2, with respect to the procurement strategy

predicted by the procurement model, which was appreciably different than that of the

actual procurement approach. It was notable that in both these mismatching cases,

the estimate of theoretical EoI based on the model’s predicted procurement took

opposite directions; however, both took a path towards the optimal number of 5-8

EoI.

The following section highlights the scope of the contributions of this research,

with respect to each research objective.

6.2 Scope and certainty of contributions to knowledge

6.2.1 New Institutional Economics and Resource-based Theory

Objective 1: to integrate TC, TCE and RBT (in a first-order stage) with respect

to the make-or-buy decision to determine the size and scope of externalised activities

within a piece of infrastructure

Based on the (four) case studies’ validation of the model, it was expected to be

able to predict the internalisation (make) and externalisation (buy) decisions of key

activities in the Stage 1/Task B: Make-or-buy analysis in the procurement of major

infrastructure projects. The model was applied in explaining make-or-buy decisions

at the activity level in four different types of projects and across road and health

sectors. This broadened the scope of the framework to explain make-or-buy across

multiple supply chains in an entire project, and from the perspective of government.


In addition, this research tested and validated the seven refutable levels in the

integrative framework of vertical integration (Bridge, 2008) in the construction

industry. As a whole, this represents an extension of scope (to whole-of-life of

projects) and certainty (more activities) of the integrative framework of vertical

integration. It thus represents a level of empirical testing of the three theories

included in the framework (TC, TCE and RBT) that is appreciably beyond the level

of other studies in the field of Construction Economics to date.

Furthermore, the model has been tested on the combination of value and

frequency variables. This has extended the model’s ability to predict an appropriate

level of capacity of the focal firm, in addition to the make-or-buy decision. In

practice, this implies the scope of the model to extend to the state or network level,

and consider procurement of major infrastructure to encompass operations and

maintenance, which can contribute towards VfM in long-term or whole-life terms. In

theory, significant progress has been made in the principles of measuring frequency

and value, which demonstrates a closer relationship between the frequency variable

and the value variable. This progress represents: i) an important step in

operationalising Coasian Theorem; ii) advancement towards the integration of TC,

TCE and RBT; and iii) the development of the theory of the firm.

Objective 2: to determine (in a second-order stage) an optimal bundle of

externalised activities which seeks to: i) minimise bureaucratic costs through greater

single-point contact with a private sector firm across a wide range of activities (that

is, the transfer of control to a private sector firm based on PRT); and ii) minimise

hold-up through the greater use of internalised management (or agents) and control

over private sector firms

Based on its validation, the model is expected to be able to identify optimal

bundles of activities and to investigate any bundles of activities within projects that

may be suitable as potential PPPs. It can determine the most efficient bundle from a

property rights’ perspective in Stage 2: Bundling Analysis by excluding potentially

troublesome activities from a bundle. Thus, the research contributes to the existing

bundling and PRT literature in terms of the minimisation of bureaucratic costs

(through the transfer of control to private sector firms), and the minimisation of hold-

up (through the greater use of internalised management and control over private

sector firms).


Objective 3: i) to deploy TCE (in a third-order stage) to determine the most efficient

type of exchange relationship for each externalised bundle or contract (to further

minimises external transaction costs); and ii) to deploy PAT to create greater

alignment between principal and agent (which also reduces transaction costs, and

promotes production benefits beyond this cost reduction)

Again based on its validation, the model is expected to predict the most

efficient external exchange relationship for each key activity, or main or major

bundle, by deploying the TCR trinity in the context of infrastructure procurement in

Stage 3: Exchange Relationship Analysis. In addition, the model is expected to align

principal-agent characteristics with the continuum of discrete to relational exchange

relationship to give an additional layer of information on the nature of external

relationship by indicating appropriate contractual payment terms and risk allocation

between client and agents. In doing so, this research answers Eisenhardt’s (1989) call

for PAT to be adopted in a principal-agent relationship after determining the

governance structure. This also represents the first attempt to align the continuum of

exchange relationship (given in Figure 3.2) with contractual outcomes based on PAT.

On this basis, the model potentially encourages greater goal alignment between

government and contractor, which reduces agency costs and promotes production

benefits beyond cost reduction.

6.2.2 Research methods

Objective 4: to develop expression of interest (EoI) as a key indicator of competition

and flexibility, and as proxy for market failure and VfM, in order to test the

procurement model

The research has achieved Objective 4 by justifying the use of competition as a

valid and reliable indicator of VfM, in so far as avoiding market failure associated

with high and low EoI. This method of testing VfM is a novel and valid approach.

Furthermore, the empirical results from the survey of major road and health projects

have established a statistically significant relationship between EoI and the key

procurement dimensions of capital value of project, procurement mode and payment

terms. This illustrates that the key aspects of procurement mode or strategy can

influence the level of EoI. Given that these results support the hypothesis, it is

expected that a project with optimal EoI (between 5-8) – which represents a reliable,

valid, visible and early indicator of the optimal configuration of a project’s key


procurement dimensions (including a PPP or non-PPP approach) – is potentially on a

path to superior VfM. Conversely, projects with sub-optimal EoI are expected to

have room for VfM improvements in their procurement dimensions.

This research represents the operationalisation of the theoretical constructs;

that is, the measurement of TCE variables (asset specificity, frequency and

uncertainty) and RBT (value, rarity and imitability) variables in the context of

procurement of major infrastructure. In particular, the operationalisation of the

transaction cost and resource-based variables at an activity level includes a novel and

innovative approach to measuring the frequency variable. Frequency is

operationalised in size and typical terms, and this represents a more concrete method

of measurement than those used in previous studies.

More importantly, however, this research recognizes the different ways in

which to fully capture the potential to generate economies of scale and/or learning

economies, and to leverage naturally occurring network advantages across road and

health projects. Specifically, frequency in road projects is measured twice; namely, at

the project level to represent the frequency of all design and construction activities,

and at the network level to represent the geographical scope of all operations and

maintenance activities. Given that the network advantages lay within the boundaries

of health sector buildings, frequency in health projects is measured once only, and

across all design, construction, operations and maintenance at the project level.

As well as contributing to the TCE literature by its novel approach to the

measurement of the frequency variable, and its considerations of natural network

advantages, this research has developed a new approach to identifying the activity as

the unit of analysis. It has done this by reference to market specialisation, and

through the iterative process of answering the suite of questions in the Stage 1/Task

B: Make-or-buy Analysis (as discussed in Section 3.2.3).

6.2.3 Procurement selection literature

The first-order procurement decision-making model addresses the weaknesses

of implementation, tautological charge and risk assessment related to MAUA-based

approaches (as discussed in Section 1.3.1). It also suggests that we can expect to see

significant room for VfM improvement arising from a more reliable and

sophisticated approach to the procurement and delivery of major public sector


projects. The model is an entirely new and scientific method of arriving at a

procurement strategy by starting the project analysis from the activity or first-order

level, rather than from the second-order level associated with the management and

organisation of activities.

The activity analysis is followed by the exclusion and separate contracting of

potentially troublesome activities, the determination of a suitable level of bundling of

activities (that is, separate DCOM, DCO, DCOM, or PPP (DCOM with private

finance), and a decision on the nature of exchange relationship for each key activity,

main activity, or major activity. The management and organisation is then

considered, subsequent to establishing the nature of exchange between government

and contractor. This approach represents a fundamental difference in the process of

selection of a suitable procurement mode for a project. In summary, the first-order

procurement decision-making model:

• zzRepresents the first attempt to examine the market of each activity, and to

determine the differential capability and competencies of government and

market at the project activity level

• Provides more confidence that the procurement method is efficacious (at a time

when research has not yet met the challenge of fully assessing all production and

transaction costs and all benefits – including the relationship of the facility with

the front-line service across PPP and non-PPP, and across the whole-life and

operations of the facility)

• Addresses the weaknesses of current practice by providing a scientific and

economic approach to decision-making

• Generates more sophistication in procurement decision-making and advances

VfM in delivering major public sector projects

• Saves time and money for both government and industry by much more reliably

identifying the most suitable projects to be procured as PPPs, and by ensuring

that the extra-over work involved in a PPP (including PPP procedure beyond

performance specification and reference design up to financial close) is justified

• Provides a transparent and public interest document that can be fully disclosed,

that supplements the PSC (in projects in which parts of PSC are not published

due to commercial-in-confidence concerns), or entirely replaces the PSC as the


published justification for the choice of a PPP path (However, a full PSC is still

envisaged for early budgeting or approval purposes).

In summary, the procurement model – a scientifically (a priori) developed

procurement decision-making model deploys a novel integration of dominant

microeconomics theory (including two Nobel Prize-winning theories) in the specific

context of social and economic infrastructure. It is empirically tested using EoI as a

valid and a reliable indicator that the configuration of the key procurement

dimensions (represented by either a PPP or non-PPP) selected for the project

increases competition (by avoiding pre-contract market failure) and enhances

flexibility (by avoiding post-contract market failure). This indication increases the

chance that the procurement approach selected – either PPP or non-PPP mode – can

deliver superior utility and whole-life cost (or VfM) relative to the competing

alternative modes. In so doing, the research has achieved its aim.

6.3 Implications and recommendations for practice

Beyond developing an innovative model to determine procurement selection,

this research addresses several procurement policy issues (as raised in Chapter 1),

namely: challenges and difficulties in delivering VfM in relative terms in public

infrastructure procurement; the weaknesses of current procurement practice

(including transparency of procurement selection in relation to the disconnect

between risk analysis and procurement selection); and the vexed issue of when or

when not to select a PPP procurement method. The findings from the hypothesis

testing provide very strong evidence that these key issues have been addressed.

More importantly, the findings highlight the merits of the model’s approach to

transaction cost economising at the first-order level in activity analysis. The first-

order approach represents a key departure from current practice, which often starts at

second-order economising, which includes the organisation and management of the

activities at a project level. This indicates the potential for incorporating the model in

procurement policies, and thus displacing the MAUA approach; for example, in

Infrastructure Australia’s National Public Private Partnership Guidelines, Volume 1:

Procurement Options Analysis (2008) (See Section 4.3 – “The decision making

process”); and in corresponding section(s) in state PPP guidelines, along with similar

guidelines used throughout the world.


The possible implications of deploying the model are given in Section 5.8, and

are summarised as follows:

• The creation of a greater proportion of different sized projects (mindful of local

supply of firms)

• Greater use of multiple contracts and multiple exchange relationships to reduce

hold-up or costly variations in time and/or cost

• Increasing rationalisation of procurement across sectors in terms of capturing the

potential to generate economies of scale and/or learning economies, and

leveraging natural or network advantages. That is, in the health sector, a greater

scope for bundling operations and maintenance with design and construct; in the

road sector, a lesser scope to bundle operations and maintenance with design and

construction, in relation to conventional on-grade road and straightforward

elevated structures

• Greater opportunities for PPP approaches in the health sector

• Support for the internalisation of operations and maintenance activities in road

government agencies. Less reliance on stereotypical procurement that minimises

capital costs and/or time, such as Design and Construct, Managing Contractor,

Early Contractor Involvement

• Less likelihood of a reliance on Alliancing for an entire project, and a greater

likelihood of more discretion in deploying Alliancing contracts in parts of the

project where it can be efficient

• More time for planning and design development prior to the award of contract

These patterns are expected to be appreciably different than the procurement

patterns observed in the sample in this research. This expected difference is based on

the theory and logic underlying the model, which stands in stark contrast to MAUA.

Before deployment of the procurement model in its forward or predictive mode, it is

recommended that users first run the model in its backwards or review mode (as per

the case study approach) on several existing projects. This is to ensure familiarity

with the model before using it to guide the procurement on new projects.

6.4 Limitations of the model

As the model is a purely economic analysis, it is likely to deliver outcomes that

do not sit well with short term imperatives that are driven by political and/or


financial concerns. For instance, projects where time to completion is absolutely

critical and substantially shorter than, say, that required to complete a process

equivalent to reaching financial close in a PPP. The model’s role is to highlight the

differences between its approach and that being proposed. This role is useful in

forming the basis for productive debate among stakeholders in the project concerned.

For example, due to the model’s long term orientation towards VfM, stereotypical

procurement approaches that favour short term advantages of minimum capital costs

and/or minimum time to completion, such as, DC, ECI or MC, are not promoted.

Similarly, the model is limited in the private sector where the scope of works cannot

be clearly defined and which evolves over a period of time; such as in the resource

sector, EPCM (engineering, procurement, and construction management) contracting

is the preferred procurement approach. EPCM is similar to ECI and MC, where the

client employs an EPCM contractor to manage the project on its behalf.

In terms of the scope for bundling, the model is also less incline to encourage a

high level of bundling of operations and maintenance with design and construction in

the road sector; except for highly complex road structures to justify the level of

bundling. Furthermore, the model is limited in promoting single large size contracts

mindful of the local supply and maintaining an optimal level of competition. Finally

the model is static and will not, for example, support all instances where the state

government is seeking to nurture the market.

6.5 Recommendations for future research

There is much scope for future research. For example, the first-order

procurement decision-making model could be tested in a more direct fashion if all

costs (production costs and transaction costs across the life cycle of the facility) and

benefits (the relationship between the facility and core service outcomes) were

available and could be effectively measured on a sizeable sample of PPPs and non-

PPPs. In this case, it is expected that the model would predict a procurement mode

with empirically superior ratio of benefits and costs, or VfM. Moreover, research

could test the model using EoI in other types of infrastructure, such as schools and

prisons. Depending on the nature of the sector, future research could also increase the

scope of analysis to include more upstream and core service activities.

References 267

References

Acharya, V. V., Cooley, T. F., Richardson, M. P., & Walter, I. (2011). Market failures and regulatory failures: Lessons from past and present financial crises. ADBI Working Paper No. 264. doi:10.2139/ssrn.1757711

Akintoye, A. S., & MacLeod, M. J. (1997). Risk analysis and management in

construction. International Journal of Project Management, 15(1), 31-38.

Al-Tabtabai, H. M. (2002). Construction procurement selection strategy using

analytical hierarchy process. Journal of Construction Procurement, 8(2), 117-132.

Alchian, A. A. (1965). Some economics of property rights. II Politico, 30, 816-829.

Alchian, A. A., & Demsetz, H. (1972). Production, information costs, and economic

organization. American Economic Review, 62(5), 777-795.

Alchian, A. A., & Demsetz, H. (1973). The property right paradigm. Journal of

Economic History, 33(1), 16-27.

Alchian, A. A., & Woodward, S. (1988). The firm is dead; long live the firm a

review of Oliver E. Williamson's the economic institutions of capitalism. Journal of Economic Literature, 26(1), 65-79. doi:10.2307/2726609

Alhazmi, T., & McCaffer, R. (2000). Project procurement system selection model.

Journal of Construction Engineering and Management, 126(3), 176-184.

Allen Consulting Group, Infrastructure Partnerships Australia, & The University of

Melbourne. (2007). Performance of PPPs and traditional procurement in Australia. Sydney, Australia: Infrastructure Partnerships Australia.

Ambrose, M. D., & Tucker, S. N. (2000). Procurement system evaluation for the

construction industry. Journal of Construction Procurement, 6(2), 121-133.

Anderson, E., & Schmittlein, D. C. (1984). Integration of the sales force: An

empirical examination. The RAND Journal of Economics, 15(3), 385-395.

Argyres, N. (1996). Evidence on the role of firm capabilities in vertical integration

decision. Strategic Management Journal, 17(2), 129-150.

268 References

Armstrong, C. E., & Shimizu, K. (2007). A review of approaches to empirical

research on the resource-based view of the firm Journal of Management, 33(6), 959-986. doi:10.1177/0149206307307645

Arrow, K. (1971). Essays in the theory of risk bearing. Chicago: Markham.

Arrow, K. J. (1969). The organization of economic activity: Issues pertinent to the

choice of market versus non-market allocation. Paper presented at The analysis and evaluation of public expenditure: The PPB system, U.S. Joint Economic Committee, 91st Congress, 1st session, Washington, D.C.

Arthur Andersen, & LSE Enterprise. (2000). Value for money drivers in the private

finance initiative. London: UK Treasury Task Force.

Audit Commission. (2003). PFI in schools. Northampton: Belmont Press.

Bajari, P., & Tadelis, S. (2001). Incentives versus transaction costs: A theory of

procurement contracts. The RAND Journal of Economics, 32(3), 387-407.

Ball, M. (2007). Markets and institutions in real estate and construction. Chichester:

Wiley-Blackwell.

Barlow, J., & Köberle-Gaiser, M. (2008). The private finance initiative, project form

and design innovation: The UK's hospitals programme. Research Policy, 37(8), 1392-1402.

Barney, J. B. (1986). Strategic factor markets: Expectations, luck and business

strategy. Management Science, 32(10), 1231-1241.

Barney, J. B. (1991). Firm resources and sustained competitive advantage. Journal of

Management in Engineering, 17(1), 99–120.

Barney, J. B. (1995). Looking inside for competitive advantage. The Academy of

Management Executive, 9(4), 49-61.

Barney, J. B. (1999). How a firm's capabilities affect boundary decisions. Sloan

Management Review, 40(3), 137-145.

Barney, J. B. (2001a). Is the resource-based "view" a useful perspective for strategic

management research? yes Academy of Managment Review, 26(1), 41-56.

References 269

Barney, J. B. (2001b). Resource-based theories of competitive advantage: A ten-year retrospective on the resource-based view. Journal of Management, 27(6), 643-650.

Barney, J. B. (2002). Gaining and sustaining competitive advantage (2nd ed.). Upper

Saddle River, N.J.: Prentice Hall.

Barney, J. B. (2007). Gaining and sustaining competitive advantage (2nd ed.). Upper

Saddle River, N.J.: Prentice Hall.

Baruch, Y. (1999). Response rate in academic studies - A comparative analysis.

Human relations, 52(4), 421-437.

Barzel, Y. (1989). Economic analysis of property rights. Cambridge: Cambridge

University Press.

Bator, F. M. (1958). The anatomy of market failure. The Quarterly Journal of

Economics, 72(3), 351-379.

Baxter, P., & Jack, S. (2008). Qualitative case study methodology: Study design and

implementation for novice researchers The Qualitative Report, 13(4), 544-559.

Beattie, V., Goodacre, A., & Fearnley, S. (2003). And then there were four: A study

of UK audit market concentration — causes, consequences and the scope for market adjustment Journal of Financial Regulation and Compliance, 11(3), 250-265.

Bennett, J., & Grice, A. (1990). Procurement systems for buildings. In Quantity

surveying techniques: New directions. Oxford: BSP Professional Books.

Besanko, D., Dranove, D., Shanley, M., & Schaefer, S. (2007). Economics of

strategy (4th ed.). Hoboken, N.J.: John Wiley & Sons.

Bethlehem, J. G. (2009). Applied survey methods: A statistical perspective. Hoboken,

N.J.: Wiley.

Bhaskar, R. (1978). A realist theory of science (2nd ed.). Hassocks: Harvester Press.

Bingham, R. D., & Felbinger, C. L. (2002). Evaluation in practice: A methodological

approach (2nd ed.). New York: Chatham House.

270 References

Blaikie, N. W. H. (2009). Designing social research: The logic of anticipation (2nd ed.). Cambridge, UK; Malden, M.A.: Polity.

Blanc-Brude, F., Goldsmith, H., & Välilä, T. (2006). Ex ante construction costs in

the European road sector: A comparison of public-private partnerships and traditional public procurement. Economic and Financial Report 2006/01. Retrieved from

Bloomfield, P., Westerling, D., & Carey, R. (1998). Innovation and risks in a public-

private partnership: Financing and construction of a capital project in Massachusetts. Public Productivity and Review, 21(4), 460-471.

Boardman, A. E., Poschmann, F., & Vining, A. R. (2005). North American

infrastructure P3s: Examples and lessons learned. In The Challenge of Public-Private Partnership: Learning from International Experience (pp. 162-189). Cheltenham: Edward Elgar.

Boerner, C. S., & Macher, J. T. (2002). Transaction cost economics: An assessment

of empirical research in the social sciences. Business and Public Policy Working Paper BPP-85. Walter A. Haas School of Business, University of California. Berkeley. rom URL

Bon, R. (1991). Why are there construction firms and why is there subcontracting in

construction? Some lessons for Soviet-type economies in transition. Paper presented at Anglo-Soviet Symposium on the Management of the Construction Firm in the Market Economy, Moscow.

Boneta, F. J. P., Peris-Ortizb, M., & Gil-Pechuanb, I. (2010). Integrating transaction

cost economics and the resource-based view in services and innovation. The Service Industries Journal, 30(5), 701-712.

Brahm, F., & Tarziján, J. (2013). Transactional hazards, institutional change, and

capabilities: Integrating the theories of the firm. Strategic Management Journal. doi:10.1002/smj.2094

Brandon, P. S. (1990). The development of an expert system for the strategic

planning of construction projects. Construction Management and Economics, 8(3), 285-300.

Brannman, L., Klein, J. D., & Weiss, L. W. (1987). The price effects of increased

competition in auction markets. The Review of Economics and Statistics, 69(1), 24-32.

References 271

Bridge, A. (1999). The procurement of facilities services: A transaction costs approach and beyond. Paper presented at Proceedings of RICS COBRA 1999, University of Salford, Salford.

Bridge, A. (2007, 16-18 July 2007). Exchange relationships in building services

maintenance activity: a test of transaction cost economics and the resource dependency view. In Proceedings of the 25th Conference of Construction Management and Economics: Past, Present and Future.

Bridge, A. (2008). The determinants of the governance of air conditioning

maintenance in Australian retail centres PhD Thesis. Queensland University of Technology.

Bridge, A. (2014a). The determinants of the vertical boundaries of the construction

firm: Testing an integrative framework of vertical integration. Unpublished manuscript, Under review.

Bridge, A. (2014b). The nature of the exchange relationship in construction activity:

Testing alternative Transaction Cost Economics and Resource-Dependency View. Unpublished manuscript, Under review.

Bridge, A., Tiong, R. L., & Wang, S. Q. (2010). Value for money: Procuring

infrastructure. In Sustainable urban and regional infrastructure development: Technologies, applications and management (pp. 295-309). Hershey, PA: Information Science Reference.

Bridge, A., & Tisdell, C. (2004). The determinants of the vertical boundaries of the

construction firm. Construction Management and Economics, 22(8), 807-825.

Bridge, A., & Tisdell, C. (2006). The determinants of the vertical boundaries of the

construction firm: response. Construction Management and Economics, 24(3), 233-236.

Bröchner, J. (2008). Construction contractors integrating into facilities management.

Facilities, 26(1/2), 6-15.

Bröchner, J. (2011). Developing construction economics as industry economics. In

Modern construction economics: Theory and application (pp. 14-28). New York, London: Spon Press.

Brown, C. (2007). Situating critical realism. Journal of International Studies 35(2),

409-416. doi:10.1177/03058298070350020201

272 References

Cánez, L. E., Platts, K. W., & Probert, D. R. (2000). Developing a framework for make-or-buy decisions. International Journal of Operations & Production Management, 20(11), 1313-1330.

Centre for Excellence and Innovation in Infrastructure Delivery (CEIID). (2010).

Infrastructure Procurement Options Guide. Perth, Australia: Building Management and Works.

Chambers, J. (2007). Infrastructure research report. n.p.: Pension Consulting

Alliance, Inc. Retrieved from http://www.pensionconsulting.com/research_general_research.htm

Chan, A. P. C. (1995). Towards an expert system on project system on project

procurement. Journal of Construction Procurement, 1(2), 111-123.

Chan, A. P. C., Yung, E. H. K., Lam, P. T. I., Tam, C. M., & Cheung, S. O. (2001).

Application of Delphi method in selection of procurement systems for construction projects. Construction Management and Economics, 19(7), 699 - 718.

Chang, C.-Y. (2013). A critical review of the application of TCE in the interpretation

of risk allocation in PPP contracts. Construction Management & Economics, 31(2), 99-103.

Chang, C.-Y., & Ive, G. (2002a). On the economic characteristics of construction

procurement systems. Paper presented at Proceedings of CIB Symposium: Procurement Systems and Technology Transfer, Trinidad & Tobago.

Chang, C.-Y., & Ive, G. (2002b). Rethinking the multi-attribute utility approach

based procurement route selection technique. Construction Management & Economics, 20(3), 275-284.

Chang, C.-Y., & Ive, G. (2007a). The hold-up problem in the management of

construction projects: A case study of the Channel Tunnel. International Journal of Project Management, 25(4), 394-404. doi:10.1016/j.ijproman.2007.01.001

Chang, C.-Y., & Ive, G. (2007b). Reversal of bargaining power in construction

projects: meaning, existence and implications. Construction Management & Economics, 25(8), 845-855.

Cheung, E., Chan, A. P. C., & Kajewski, S. (2009). Enhancing value for money in

public private partnership projects: Findings from a survey conducted in Hong Kong and Australia compared to findings from previous research in the

References 273

UK. Journal of Financial Management of Property and Construction, 14(1), 7-20.

Cheung, S.-O., Lam, T.-I., Leung, M.-Y., & Wan, Y.-W. (2001). An analytical

hierarchy process based procurement selection method. Construction Management & Economics, 19(4), 427-437.

Coase, R. H. (1937). The nature of the firm. Economica, 4(16), 386-405. Retrieved

from reprinted (1990). pp. 33-55 In Coase, the firm, the market, and the law. Chicago: University of Chicago Press.

Coase, R. H. (1960). Problem of social cost. The Journal of Law and Economics,

3(1), 1-44.

Coase, R. H. (1988a). The nature of the firm: Influence. Journal of Law, Economics

and Organization, 4(1), 33-47.

Coase, R. H. (1988b). The nature of the firm: Meaning. Journal of Law, Economics

and Organization, 4(1), 19-32.

Coase, R. H. (1988c). The nature of the firm: Origin. Journal of Law, Economics and

Organization, 4(1), 3-17.

Coase, R. H. (1991). The nature of the firm: Influence. In O. E. Williamson & S. G.

Winter (Eds.), The nature of the firm: origins, evolution, and development (pp. 244). New York: Oxford University Press.

Coase, R. H. (2000). The acquisition of Fisher Body by General Motors. Journal of

Law and Economics, 43(1), 15-31.

Connor, K. (1991). A historical comparison of resource-based theory and five

schools of thought within industrial organization economics: Do we have a new theory of the firm? Journal of Management(17), 121-154.

Construction Industry Institute. (2001). Project delivery and contract strategy.

Austin, TX: Author.

Cooperative Research Centre for Construction Innovation. (2008). Capital works

procurement: The selection of a building procurement method. Brisbane, Australia: Author.

274 References

Costantino, N., & Pietroforte, R. (2004). Production arrangements by US building and nonbuilding contractors: an update. Construction Management and Economics, 22(3), 231- 235.

Costantino, N., Pietroforte, R., & Hamill, P. (2001). Subcontracting in commercial

and residential construction: an empirical investigation. Construction Management and Economics, 19(4), 439-447.

Cox, A. (1997). Business Success. Boston: Earlsgate Press.

Crocker, K. J., & Masten, S. E. (1988). Mitigating contractual hazards: Unilateral

options and contract length. Rand Journal of Economics, 19(3), 327-343.

Dainty, A. (2008). Methodological pluralism in construction management research.

In Advanced research methods in the built environment (pp. 1–13). Chichester: Wiley Blackwell.

Dell'Isola, M. D., Licameli, J. P., & Arnold, C. (1998). How to form a decision

matrix for selecting a project. Design-build Strategies, 4(2).

Demsetz, H. (1967). Toward a theory of property rights. American Economic

Review, 57(2), 347-359.

Demsetz, H. (1969). Information and efficiency: Another viewpoint. Journal of Law

and Economics, 12(1), 1-22.

Demsetz, H. (1988). Theory of the firm revisited. Journal of Law, Economics &

Organisation, 4(1), 141-161.

Department of Treasury and Finance Victoria. (2006). Public Sector Asset

Investment Program 2006–07: Budget Information Paper No. 1. Melbourne, Australia: Author Retrieved from http://www.budget.vic.gov.au/CA25713B0016B131/WebObj/BIP1-wholedocument/$File/BIP1%20-%20whole%20document.pdf.


Investment Program 2007–08: Budget Information Paper No. 1. Melbourne, Australia: Author Retrieved from http://www.budget.vic.gov.au/CA2572B00081B35D/WebObj/BudgetInformationPaperNo/$File/Budget%20Information%20Paper%20No.1.pdf.


Investment Program 2008–09: Budget Information Paper No. 1. Melbourne,

References 275

Australia: Author Retrieved from http://www.budget.vic.gov.au/CA257401000ED28B/WebObj/BudgetInformationPaperNo/$File/Budget%20Information%20Paper%20No.1_200809.pdf.


Investment Program 2009–10: Budget Information Paper No. 1. Melbourne, Australia: Author Retrieved from http://www.budget.vic.gov.au/CA25755B0004CE3B/WebObj/2009-10_BIP1/$File/2009-10_BIP1.pdf.

Department of Treasury and Finance Victoria. (2012). Investment Lifecycle and High

Value/High Risk Guidelines. Melbourne, Australia: Author.

Dillman, D. A. (1991). The design and administration of mail surveys. Annual

Review of Sociology, 17, 225-249.

Dillman, D. A. (2000). Mail and Iiternet surveys: The tailored design method (2nd

ed.). New York: Wiley.

Dillman, D. A. (2008). Total Design Method (TDM). In Encyclopedia of Survey

Research Methods (pp. 893-897). Thousand Oaks, CA: Sage Publications, Inc.

Dixon, T., Pottinger, G., & Jordan, A. (2005). Lessons from the private finance

initiative in the UK benefits, problems and critical success factors. Journal of Property Investment & Finance, 23(5), 412-423.

Domberger, S., & Rimmer, S. (1994). Competitive tendering and contracting in the

public sector: A survey. International Journal of the Economics of Business, 1(3), 439-453. doi:10.1080/758536232

Doree, A. G. (1997). Construction procurement by Dutch municipalities.

Construction Engineering and Management, 3(3), 78-88.

Duffield, C. (2009). Driving public infrastructure benefits through the choice of

procurement strategy. Paper presented at Proceedings of Global Innovation in Construction Conference, Loughborough University, Loughborough, UK.

Eccles, R. G. (1981a). Bureaucratic versus craft administration: The relationship of

market structure to the construction firm. Administrative Science Quarterly, 26(3), 449-469.

276 References

Eccles, R. G. (1981b). The quasifirm in the construction industry. Journal of Economic Behavior & Organization, 2(4), 335-357. Retrieved from http://www.sciencedirect.com/science/article/B6V8F-45MFS2J-3/2/878f13f74691559f019474dec28d3832. doi:10.1016/0167-2681(81)90013-5

Edwards, P., Shaoul, J., Stafford, A., & Arblaster, L. (2004). Evaluating the

operation of PFI in roads and hospitals. London: Certified Accountants Education Trust.

Eisenhardt, K. M. (1989). Agency theory: An assessment and review. The Academy

of Management Review, 14(1), 57-74.

Eriksson, P. E. (2006). Procurement and governance management - Development of

a conceptual procurement model based on different types of control. Management Revue, 17(1), 30-49.

Fama, E. F., & Jensen, M. C. (1983). Separation of ownership and control. [

Corporations and Private Property: A Conference Sponsored by the Hoover Institution (Jun., 1983)]. Journal of Law and Economics, 26(2), 301-325.

Farrell, L. M. (2003). Principal-agency risk in project finance. International Journal

of Project Management, 21(8), 547-561.

Fitzgerald, P. (2004). Review of Partnerships Victoria Provided Infrastructure.

Melbourne: Author.

Flanagan, R., Lu, W., Shen, L., & Jewell, C. (2007). Competitiveness in

construction: A critical review of research. Construction Management and Economics, 25(9), 989-1000.

Floricel, S., & Lampel, J. (1998). Innovative contractual structures for

interorganizational systems. International Journal of Technology Management, 16(1-3), 193-207.

Flyvbjerg, B., Holm, M. S., & Buhl, S. (2002). Underestimating costs in public

works projects: Error or lie? Journal of the American Planning Association, 68(3), 279-295.

Foss, N. J. (1993). Theories of the firm: Contractual and competence perspectives.

Journal of Evolutionary Economics, 3(2), 127-144.

References 277

Foss, N. J. (1997a). The classical theory of production and the capabilities view of the firm. Journal of Economic Studies, 24(5), 307-323.

Foss, N. J. (1997b). Resources, firms, and strategies: A reader in the resource-based

perspective. Oxford, New York: Oxford University Press.

Foss, N. J., & Knudsen, T. (2003). The resource-based tangle: Towards a sustainable

explanation of competitive advantage. Managerial and Decision Economics, 24(4), 291-307.

Franks, J. (1990). Building procurement systems. Ascot: Chartered Institute of

Building.

Friedman, M. (1962). Capitalism and freedom. Chicago: University of Chicago.

Furubotn, E. G., & Pejovich, S. (1972). Property rights and economic theory: A

survey of recent literature. Journal of Economic Literature, 10(4), 1137-1162.

Geyskens, I., Steenkamp, J.-B., & kumar, N. (2006). Make, buy, or ally: A

transaction cost theory meta-analysis. Academy of Managment Journal, 49(3), 519-543.

Ghobadian, A., Gallear, D., O’Regan, N., & Viney, H. (2004). Public-private

partnerships: Policy and experience. London: Palgrave.

Godfrey, P. C., & Hill, C. W. L. (1995). The problem of unobservables in strategic

management research. Strategic Management Journal, 16(7), 519-533.

González-Dı́az, M., Arruñada, B., & Fernández, A. (1998). Regulation as a cause of

firm fragmentation: The case of the Spanish construction industry. International Review of Law and Economics, 18(4), 433-450.

González-Dı́az, M., Arruñada, B., & Fernández, A. (2000). Causes of subcontracting:

Evidence from panel data on construction firms. Journal of Economic Behavior & Organization, 42(2), 167-187.

Government of South Australia. (2006). 2006-07 Capital investment statement:

Budget Paper 5. Adelaide, Australia: Author Retrieved from http://www.treasury.sa.gov.au/__data/assets/pdf_file/0004/1003/200607-bp5-capital-statement.pdf.

278 References

Government of South Australia. (2007). 2007-08 Capital investment statement: Budget Paper 5. Adelaide, Australia: Author Retrieved from http://www.treasury.sa.gov.au/__data/assets/pdf_file/0020/983/200708-bp5-capital-statement.pdf.





Government of Western Australia. (2006). 2006-07 Budget: Budget Paper No. 2

Volume 2. Perth, Australia: Author Retrieved from http://www.treasury.wa.gov.au/cms/uploadedFiles/State_Budget/Budget_2006_-_2007/bp2_vol2.pdf?n=2279.


Volume 2. Perth, Australia: Author Retrieved from http://www.treasury.wa.gov.au/cms/uploadedFiles/State_Budget/Budget_2007_-_2008/bp2_vol2.pdf.


Volume 2. Perth, Australia: Author Retrieved from http://www.treasury.wa.gov.au/cms/uploadedFiles/State_Budget/Budget_2008_-_2009/bp2_v2.pdf?n=2721.


Volume 2. Perth, Australia: Author Retrieved from http://www.treasury.wa.gov.au/cms/uploadedFiles/State_Budget/Budget_2009_2010/2009-10_bp2_vol2.pdf.

Grant, R. M. (1996). Toward a knowledge-based theory of the firm. Strategic

Management Journal, 17(Winter special issue), 109-122.

Greve, C. (2003). When public-private partnerships fail: the extreme case of the

NPM-inspired local government of farum in Denmark. Paper presented at Proceedings of European Group of Public Administration (EGPA), Oeiras, Portugal.

References 279

Griffith, A., & Headley, J. D. (1997). Using a weighted score model as an aid to selecting procurement methods for small building works. Construction Management & Economics, 15(4), 341-348.

Grimsey, D., & Lewis, M. (2009). Developing a framework for procurement options

analysis. In Policy, finance & management for public-private partnerships. Oxford, England: Wiley-Blackwell.

Grimsey, D., & Lewis, M. K. (2004). Public Private Partnerships: The worldwide

revolution in infrastructure provision and project finance. Cheltenham, UK: Edward Elgar.

Grimsey, D., & Lewis, M. K. (2007). Public Private Partnerships and Public

Procurement. Agenda, 14(2), 171-188.

Groenewegen, J., & Vromen, J. (1996). A case for theoretical pluralism. In

Transaction cost economics and beyond. Boston, MA: Kluwer Academic.

Grossman, S. J., & Hart, O. D. (1986). The costs and benefits of ownership: A theory

of vertical and lateral integration. The Journal of Political Economy, 94(4), 691-719.

Gulbrandsen, B., Sandvik, K., & Haugland, S. A. (2009). Antecedents of vertical

integration: Transaction cost economics and resource-based explanations. Journal of Purchasing and Supply Management, 15(2), 89-102. doi:10.1016/j.pursup.2008.12.003

Gunnarson, S., & Levitt, R. E. (1982). Is a building construction project a hierarchy

or a market? Paper presented at Proceedings of The seventh world congress on project management, Copenhagen.

Gupta, S. (2002). Competition and collusion in a government procurement auction

market. Atlantic Economic Journal, 30(1), 13-25.

H M Treasury. (2008). Infrastructure procurement: Delivering long-term value.

London: Office of Public Sector Information.

Hart, O. (1995). Firms, contracts and financial structure. Oxford: Clarendon Press.

Hart, O. (2003). Incomplete contracts and public ownership: Remarks, and an

application to public-private partnerships. Economic Journal, 113(486), C69-C76.

280 References

Hart, O., & Moore, J. (1990). Property rights and the nature of the firm. Journal of

Political Economy, 98(6), 1119-1158.

Heide, J. B., & John, G. (1988). The role of dependence balancing in safeguarding

transaction-specific assets in conventional channels. Journal of Marketing, 52(1), 20-35.

Henderson, D. A. (1990). The influence of corporate strategy, structure and

technology on location of procurement and sales. Unpublished PhD thesis. University of Michigan. Michigan. rom URL

Henderson, R., & Cockburn, I. (1994). Measuring competence? exploring firm

effects in pharmaceutical research. Strategic Management Journal, 15(Special Issue: Competitive Organizational Behavior), 63-84.

Hodge, G. A. (2005). Public-private partnerships: The Australian experience with

physical infrastructure. In The Challenge of Public-Private Partnership: Learning from International Experience. Cheltenham: Edward Elgar.

Hodge, G. A., & Greve, C. (2009). PPPs: The passage of time permits a sober

reflection. Economic Affairs, 29(1), 33-39.

Holmstrom, B., & Milgrom, P. (1991). Multitask principal-agent analyses: Incentive

contracts, asset ownership, and job design. Journal of Law, Economics & Organization, 7(Special Issue: Papers from theConference on the New Science of Organization), 24-52.

Holmstrom, B., & Milgrom, P. (1994). The firm as an incentive system. American

Economic Review, 84(4), 972-991.

Holmstrom, B., & Roberts, J. (1998). The boundaries of the firm revisited. The

Journal of Economic Perspectives, 12(4), 73-94.

House of Lords. (2010). Private finance projects and off-balance sheet debt: 1st

report of Session 2009-10. Great Britain Parliament, London: Select Committee on Economic Affairs.

Ibbs, W., & Chih, Y.-Y. (2011). Alternative methods for choosing an appropriate

project delivery system (PDS). Facilities, 29(13/14), 527-541.

Infrastructure. (2013). Oxford University Press.

References 281

Infrastructure Australia. (2008). A report to the council of Australian Government.

n.p.: Author.

Infrastructure Partnerships Australia. (2008). National Public Private Partnership

Guidelines Volume 1: Procurement options analysis. ACT, Australia: Author.

Infrastructure partnerships Australia. (2010). The role of superannuation in building

Australia's future. Sydney, Australia: Author.

Ive, G., & Chang, C.-Y. (2007). The principle of inconsistent trinity in the selection

of procurement systems. Construction Management & Economics, 25(7), 677-690.

Jacobides, M. G., & Winter, S. G. (2005). The co-evolution of capabilities and

transaction costs: Explaining the institutional structure of production. Strategic Management Journal, 26(5), 395-413.

Jaspers, F., & Ende, J. v. d. (2006). The organizational form of vertical relationships:

Dimensions of integration. Industrial Marketing Management, 35(7), 819-828.

Jayaratne, T. E., & Stewart, A. J. (1991). Quantitative and qualitative methods in the

social sciences: Current feminist issues and practical strategies. In Beyond methodology: feminist scolarship as lived research (pp. 85-106). Bloomington, I.N.: Indiana University Press.

Jensen, M. C., & Meckling, W. H. (1976). Theory of the firm: Managerial behavior,

agency costs and ownership structure. Journal of financial Economics 3(4), 305-360.

Jin, X.-H. (2009). Allocating risks on Public-Private Partnerships using a Transaction

Cost Economics Approach: A case study. The Australian Journal of Construction Economics and Building, 9(1), 19-26.

Jin, X.-H. (2010). Determinants of efficient risk allocation in privately financed

public infrastructure projects in Australia. Journal of Construction Engineering and Management, 136(2), 138-150.

Jin, X.-H., & Doloi, H. (2008). Interpreting risk allocation mechanism in public-

private partnership projects: An empirical study in a transaction cost economics perspective. Construction Management and Economics, 26(7), 707 - 721.

282 References

Jin, X.-H., & Doloi, H. (2009). Modeling risk allocation in privately financed

infrastructure projects using fuzzy logic. Computer-Aided Civil and Infrastructure Engineering, 24(7), 509-524.

Jin, X.-H., & Zhang, G. (2011). Modelling optimal risk allocation in PPP projects

using artificial neural networks. International Journal of Project Management, 29(5), 591-603.

John, G., & Weitz, B. A. (1988). Forward integration into distribution: An empirical

test of transaction cost analysis. Journal of Law, Economics, & Organization, 4(2), 337-355.

Johnson, R. (1990). The economics of building: A practical guide for the design

professional. New York: Wiley.

Joskow, P. L. (1985). Vertical integration and long-term contracts: The case of coal-

burning electric generating plants. Journal of Law, Economics, and Organization 1(1), 33-80.

Joskow, P. L. (1988). Asset specificity and the structure of vertical relationships:

Empirical evidence. Journal of Law, Economics, & Organization, 4(1), 95-117.

Kale, S., & Arditi, D. (2001). General contractors’ relationships with subcontractors:

A strategic asset. construction Management & Economics, 19(5), 541-549.

Kalnins, A., & Mayer, K. J. (2004). Relationships and hybrid contracts: An analysis

of contract choice in information technology. The Journal of Law, Economics, & Organization, 20(1), 207-229.

Kaplan, A. (1964). The conduct of inquiry. San Francisco, C.A.: Chandler.

Kaplowitz, M. D., Hadlock, T. D., & Levine, R. (2004). A comparison of web and

mail survey response rates. The Public Opinion Quarterly, 68(1), 94-101.

Kaufmann, P. J., & Stern, L. W. (1988). Relational exchange norms, perceptions of

unfairness, and retained hostility in commercial litigation. The Journal of Conflict Resolution, 32(3), 534-552.

Khalil, M. I. A. (2002). Selecting the appropriate project delivery method using

AHP. International Journal of Project Management, 20(6), 469-474.

References 283

Kim, J., & Mahoney, J. T. (2005). Property rights theory, transaction costs theory,

and agency theory: an organizational economics approach to strategic management. Managerial and Decision Economics, 26(4), 223-242.

Klein, B. (1988). Vertical integration as organizational ownership: The Fisher Body-

General Motors relationship revisited related papers. Journal of Law, Economics, and Organization, 4(1), 199-213.

Klein, B. (2005). The make-or-buy decision: Lessons from empirical studies. In

Handbook of New Institutional Economics (Vol. 4, pp. 435–464). Netherlands: Springer US.

Klein, B., Crawford, R. G., & Alchian, A. A. (1978). Vertical integration,

appropriable rents, and the competitive contracting process. Journal of Law and Economics, 21(2), 297-326.

Kogut, B., & Zander, U. (1993). Knowledge of the firm and the evolutionary theory

of the multinational corporation. Journal of International Business Studies, 24(4), 625-645.

Kogut, B., & Zander, U. (1996). What firms do? Coordination, identity, and learning.

Organization Science, 7(5), 502-518.

KPMG, & Infrastructure Australia. (2010). PPP Procurement: Review of barriers to

competition and efficiency in the procurement of PPP projects. n.p.: Author. Retrieved from http://www.kpmg.com/nz/en/issuesandinsights/articlespublications/smarterprocurement/pages/review-of-barriers-to-competition.aspx

KPMG, & University College London. (2010). Operating healthcare infrastructure.

n.p.: KPMG International.

Kumaraswamy, M., & Dissanayaka, S. (2001). Developing a decision support system

for building project procurement. Building and Environment, 36(3), 337-349. doi:10.1016/S0360-1323(00)00011-1

Kumaraswamy, M. M., Anvuur, A., & Mahesh, G. (2008). Contractual frameworks

and cooperative relationships. In Collaborative relationships in construction: Developing frameworks and networks (pp. 78–104). New York: Wiley-Blackwell.

Kumaraswamy, M. M., & Dissanayaka, S. (1998). Linking procurement systems to

project priorities. Building Research & Information, 26(4), 223-238.

284 References

Laffont, J.-J., & Martimort, D. (2002). The theory of incentives. Princeton, N.J.:

Princeton University Press.

Laffont, J.-J., & Tirole, J. (1987). Auctioning Incentive Contracts. Journal of

Political Economy, 95(5), 921-937.

Lai, W. C. (2000). The Coasian market-firm dichotomy and subcontracting in the

construction industry. Construction Management & Economics, 18(3), 355-362.

Langlois, R. N. (1992). Transaction-cost Economics in real time. Industrial and

Corporate Change, 1(1), 99-127.

Langlois, R. N., & Robertson, P. L. (1989). Explaining vertical integration: Lessons

from the American automobile industry. The journal of Economic History, 49(2), 361-375.

Lee, D. R., & Clark, J. R. (2013). Market failures, government solutions, and moral

perceptions Cato Journal, 33(2), 287-297.

Leiblein, M., & Miller, D. (2003). An empirical examination of transaction and firm-

level influences on the vertical boundaries of the firm. Strategic Management Journal, 24, 839-859.

Leviäkangas, P. (2007). Private finance of transport infrastructure projects: Value

and risk analysis of a Finnish shadow toll road project. Finland: VTT Technical Research Centre of Finland.

Libecap, G. D. (1989). Contracting for property rights. New York: Cambridge

University Press.

Lippman, S. A., & Rumelt, R. P. (1982). Uncertain imitability: An analysis of

interfirm differences in efficiency under competition. The Bell Journal of Economics,, 13(2), 418-438.

Liu, A. M. M. (1994). From act to outcome: A cognitive model of construction

procurement. In Proceedings of CIB W92 "East Meets West” Procurement Systems Symposium (pp. 169-178).

Love, P. E. D., Davis, P. R., Edwards, D. J., & Baccarini, D. (2008). Uncertainty

avoidance: Public sector clients and procurement selection. International

References 285

Journal of Public Sector Management, 21(7), 753-776. doi:10.1108/09513550810904550

Love, P. E. D., Edwards, D. J., Irani, Z., & Sharif, A. (2011). Participatory action

research approach to public sector procurement selection. Journal of Construction Engineering and Management, 138(3), 311-322.

Love, P. E. D., Skitmore, M., & Earl, G. (1998). Selecting a suitable procurement

method for a building project. Construction Management & Economics, 16(2), 221-233.

Love, P. E. D., Tse, R. Y. C., Holt, G. D., & Proverbs, D. G. (2002). Transaction

costs, learning and alliances. Journal of Construction Research, 3(2), 193-207.

Luu, D. T., Ng, S. T., & Chen, S. E. (2003). Parameters governing the selection of

procurement system – an empirical survey. Engineering, Construction and Architectural Management, 10(3), 209-218.

Luu, D. T., Ng, S. T., & Chen, S. E. (2005). Formulating procurement selection

criteria through case-based reasoning approach. Journal of Computing in Civil Engineering, 19(3), 269-276.

Lyons, B., & Mehta, J. (1997). Contracts, opportunism and trust: self-interest and

social orientation. Cambridge Journal of Economics, 21(2), 239-257.

Lyons, B. R. (1996). Empirical relevance of efficient contract theory: inter-firm

contracts. Oxf Rev Econ Policy, 12(4), 27-52.

Lyons, T., & Skitmore, M. (2004). Project risk management in the Queensland

engineering construction industry: A survey. International Journal of Project Management, 22(1), 51-61.

Macdonald, M. (2002). Review of large public procurement in the UK. London:

Author.

Macher, J., & Richman, B. D. (2008). Transaction cost economics: An assessment of

empirical research in the social sciences. Business and Politics, 10(1). doi:10.2202/1469-3569.1210

MacNeil, I. R. (1978). Contracts: Adjustment of long-term economic relations under

classical, neoclassical and relational contract law. Northwestern University Law Review, 72(6), 854-905.

286 References

Madhok, A. (1996). The organization of economic activity: Transaction costs, firm

capabilities, and the nature of governance. Organization Science, 7(5), 577-590.

Madhok, A. (2002). Reassessing the fundamentals and beyond: Ronald Coase, the

transaction cost and resource-based theories of the firm and the institutional structure of production. Strategic Management Journal, 23(6), 535-550.

Maher, M. E. (1997). Transaction Cost Economics and Contractual Relations.

Cambridge Journal of Economics 21(2), 147-170.

Mahoney, J. T. (1995). The management of resources and the resource of

management. Journal of Business Research, 33(2), 91-101.

Mahoney, J. T. (2005). Economic foundations of strategy. Thousand Oaks, C.A.:

Sage.

Manoliadis, O. G., Pantouvakis, J.-P., & Christodoulou, S. E. (2009). Improving

qualifications-based selection by use of the fuzzy Delphi method. Construction Management & Economics, 27(4), 373-384.

Masten, S. E., Meehan, J. W., & Snyder, E. A. (1991). The costs of organization.

Journal of Law, Economics, and Organization, 7(1), 1-25.

Masten, S. E., & Saussier, S. (2000). Econometrics of contracts: An assessment of

developments in the empirical literature on contracting. Revue d’Economie Industrielle, 92(92), 215-236.

McAfee, R. P., & McMillan, J. (1986). Bidding for contracts: A principal-agent

analysis. The RAND Journal of Economics, 17(3), 326-338.

McCaffer, R. (1979). Bidding behaviour. Quantity surveying (New Zealand), 6-13.

McIvor, R. T., Humphreys, P. K., & McAleer, W. E. (1997). A strategic model for

the formulation of an effective make-or-buy decision. Management Decision, 32(2), 169-178.

Milgrom, P. (1979). A convergence theorem for competitive bidding with differential

information. Econometrica, 47(3), 679-688.

References 287

Milgrom, P., & Roberts, J. (1988). Economic theories of the firm: Past, present, and future. Canadian Journal of Economics, 21(3), 444-458.

Miller, C. J. M., Packham, G. A., & Thomas, B. C. (2002). Harmonisation between

main contractors and subcontractors: A prerequisite for lean construction? Journal of construction Research, 3(1), 67-82.

Mohsini, R. A., & Botros, A. F. (1990). Pascon: An expert system to evaluate

alternate project procurement processes. Paper presented at Proceedings CIB W90 International Symposium on Building Economics and Construction Management, Sydney, Australia.

Molenaar, K. R., & Songer, A. D. (2001). Web-based decision support systems: Case

study in project delivery. Journal of Computing in Civil Engineering, 15(4), 259-267.

Molloy, J. C., Chadwick, C., Ployhart, R. E., & Golden, S. J. (2011). Making

intangibles "tangible" in tests of resource-based theory: A multidisciplinary construct validation approach. Journal of Management, 37(5), 1496-1518. doi:10.1177/0149206310394185

Moser, C., & Kalton, G. (1971). Survey methods in social investigation (2nd ed.).

Aldershot, England: Dartmouth.

Müller, R., & Turner, J. R. (2005). The impact of principal-agent relationship and

contract type on communication between project owner and manager. International Journal of Project Management, 23(5), 398-403.

National Audit Office. (2000). Examining the value for money of deals under the

private finance initative. London: The Stationary Office.

National Audit Office. (2003). The operational performance of PFI prisons (HC 700

Session 2002-2003). London: The Stationary Office.

National Audit Office. (2004). Improving procurement: Progress by the office of

government commerce in improving departments capability to procure cost-effectively. London: The Stationary Office.

National Audit Office. (2010). The performance and management of hospitals PFI

contracts (HC 68 Session 2010-2011). London: The Stationary Office.

National Audit Office. (2011). Lessons from PFI and other projects (HC 920 Session

2010-2012). London: The Stationary Office.

288 References

National Economic Development Office (NEDO). (1985). Think about building: A

successful business customer's guide to using the construction industry. London: National Economic Development Agency.

Nelson, R. R. (1987). Roles of government in a mixed economy. Journal of Policy

Analysis and Management, 6(4), 541-557.

Nelson, R. R., & Winter, S. G. (1982). An evolutionary theory of economic change.

Cambridge, M.A.: Belknap Press.

Newbert, S. L. (2007). Empirical research on the resource-based view of the firm: an

assessment and suggestions for future research. Strategic Management Journal, 28(2), 121-146. doi:10.1002/smj.573

Ngai, S., Drew, D., Lo, H. P., & Skitmore, M. (2002). A theoretical framework for

determining the minimum number of bidders in construction bidding competitions. Construction Management & Economics, 20(6), 473-482.

North, D. C. (1990). Institutions, institutional change and economic performance.

New York: Cambridge University Press.

NSW Department of Commerce. (2006). Procurement method selection guidelines.

Sydney, Australia: Author.

NSW Treasury. (2006). Infrastructure Statement 2006-07: Budget Paper No.4.

Sydney, Australia: Author Retrieved from http://www.treasury.nsw.gov.au/budget_papers/Budget_Papes_2006-07/bpapers.


Sydney, Australia: Author Retrieved from http://www.treasury.nsw.gov.au/budget_papers/Budget_Papers_2007-08/bp4/bp5.


Sydney, Australia: Author Retrieved from http://www.treasury.nsw.gov.au/budget_papers/Budget_Papers_2008-09/bp4/2008-09_budget_paper_4.


Sydney, Australia: Author Retrieved from http://www.treasury.nsw.gov.au/budget_papers/budget_2009-10/bp4/2009-10_budget_paper_4.

References 289

NSW Treasury. (2012). NSW Public Private Partnerships Guidelines. Sydney,

Australia: Author.

OECD. (2006). Infrastructure to 2030: Telecom Land, Transport, Water and

Electricity. Paris, France: Author. Retrieved from http://www.oecd.org/futures/infrastructureto2030/infrastructureto2030telecomlandtransportwaterandelectricity.htm

Parkhe, A. (1993). Strategic alliance structuring: A game theoretic and transaction

cost examination of interfirm cooperation. Academy of Managment Journal, 36(4), 794-829.

Penrose, E. T. (1959). The theory of growth of the firm. London: Basil Blackwell.

Peteraf, M. A. (1993). The cornerstones of competitive advantage: A resource-based

view. Strategic Management Journal, 14(3), 179-191.

Pollitt, M. G. (2002). The declining role of the state in infrastructure investment in

the UK. In Private initiatives in infrastructure: priorities, incentives and performance. Aldershot: Edward Elgar.

Pollitt, M. G. (2005). Learning from the UK private finance initiative experience. In

The challenge of public-private partnership: Learning from international experience. Cheltenham: Edward Elgar.

Pollock, A. M., Prince, D. W., & Player, S. (2007). An examination of the UK

Treasury's evidence base for cost and time overrun data in UK value-for-money policy and appraisal. Public Money & Management, 27(2), 127-134.

Pollock, A. M., Shaoul, J., & Vickers, N. (2002). Private finance and “value for

money” in NHS hospitals: A policy in search of a rationale? British Medical Journal, 324(7347), 1205-1208.

Popper, K. (1959). The logic of scientific discovery. London: Hutchinson

Poppo, L., & Zenger, T. (1998). Testing altermative theories of the firm: Transaction

cost, knowledge-based, and measurement explanations for make-or-buy decisions in information services. Strategic Management Journal, 19(9), 853-877.

Porter, M. E. (1980). Competitive Strategy: Techniques for Analyzing Industries and

Competitors. New York: Free Press.

290 References

Prahalad, C. K., & Hamel, G. (1990). The core competence of the corporation.

[Article]. Harvard Business Review, 68(3), 79-91. Retrieved from bsh.

Priem, R. L., & Butler, J. E. (2001). Is the esource-based "view" a useful perspective

for strategic management research? The Academy of Management Review, 26(1), 22-40.

Queensland Government - Department of Infrastructure and Planning. (2008a).

Public Private Partnerships - Policy. Brisbane, Australia: Author Retrieved from http://www.treasury.qld.gov.au/office/knowledge/docs/public-private-partnership-guidance/index.shtml.

Queensland Government - Department of Infrastructure and Planning. (2008b).

Public Private Partnerships guidance material, Supporting document - Business case development. Brisbane, Australia: Author Retrieved from http://www.treasury.qld.gov.au/office/knowledge/docs/public-private-partnership-guidance/index.shtml viewed on 14 June 2013.

Queensland Government. (2006). State Budget 2006-07: Capital Statement, Budget

Paper No. 3. Brisbane, Australia: Author Retrieved from http://www.budget.qld.gov.au/current-budget/budget-papers/bp3.php.


Paper No. 3. Brisbane, Australia: Author Retrieved from http://www.budget.qld.gov.au/budget-papers/2007-08/bp3-2007-08.pdf.





Rahman, M. M., & Kumaraswamy, M. M. (2004). Contracting relationship trends

and transitions. Journal of Management in Engineering, 20(4), 147-161.

Raisbeck, P., Duffield, C., & Xu, M. (2010). Comparative performance of PPPs and

traditional procurement in Australia. Construction Management and Economics, 28(4), 345-359.

Remler, D. K., & Van Ryzin, G. G. (2011). Research methods in practice : strategies

for description and causation. Thousand Oaks, California: Sage Publications.

References 291

Reve, T., & Levitt, R. E. (1984). Organization and governance in construction.

International Journal of Project Management, 2(1), 17-25.

Rindfleisch, A., & Heide, J. B. (1997). Transaction cost analysis: Past, present, and

future applications. Journal of Marketing, 61(4), 30-54.

Ross, S. A. (1973). The economic theory of agency: The principal's problem. The

American Economic Review, 63(2), 134-139.

Rumelt, R. P. (1984). Toward a strategic theory of the firm. In Competitive strategic

management (pp. 556-570). Englewood Cliffs, N.J.: Prentice Hall.

Runeson, G. (1997). The role of theory in construction management research:

comment. Construction Management and Economics, 15(3), 299-302.

Savas, E. S. (2000). Privatization and public-private partnerships. New York:

Chatham House Publishers and Seven Bridges Press.

Selten, B. (1973). A simple model of imperfect competition, where 4 are few and 6

are many. International journal of Game Theory, 2(1), 141-201.

Seymour, D., Crook, D., & Rooke, J. (1997). The role of theory in construction

management: A call for debate. Construction Management and Economics, 15(1), 117-119.

Sha, K. (2011). Vertical governance of construction projects: An information cost

perspective. Construction Management and Economics, 29(11), 1137-1147.

Shaoul, J. (2005). The private finance initiative or the public funding of private

profit. In The Challenge of Public-Private Partnership: Learning from International Experience. Cheltenham: Edward Elgar.

Shelanski, H. A., & Klein, P. G. (1995). Empirical research in transaction cost

economics: A review and assessment. Journal of Law, Economics, & Organization, 11(2), 335-361.

Shen, G. (1993). A knowledge based structure for implementing Value Management

in the design of office buildings PhD Thesis. University of Salford, Salford.

292 References

Shen, G. Q., & Yu, A. T. W. (2012). Value management: recent developments and way forward. Construction Innovation: Information, Process, Management, 12(3), 264-271. doi:10.1108/14714171211244631

Shepherd, W. G. (1982). Causes of Increased Competition in the U.S. Economy,

1939-1980. The Review of Economics and Statistics, 64(4), 613-626.

Sidwell, T., Kennedy, R., Bennet, J., & Chan, A. P.-C. (2003). A value driven

procurement decision tool. Retrieved from http://eprints.qut.edu.au/9963/

Singh, S. (1990). Selection of appropriate project delivery system for construction

projects. Paper presented at Proceedings CIB W90 International Symposium on Building Economics and Construction Management, Sydney, Australia.

Skitmore, M. (2002). Raftery curve construction for tender price forecasts.

Construction Management and Economics, 20(1), 83-89.

Skitmore, R. M., & Marsden, D. E. (1988). Which procurement system? Towards a

universal procurement selection technique. Construction Management & Economics, 6(1), 71-89.

Smith, A. (1776). An inquiry into the nature and causes of the wealth of nations.

Oxford: Oxford University Press.

SPSS. (1998). Survey research using SPSS. Illinois: SPSS Inc.

SRD Consulting. (2000). Achieving oustanding performance. Paper presented at

Value Management Workshop, Relation based contracting Queensland Department of Main Roads, Quest on North Quay Brisbane.

State Procurement Board. (2013). Procurement Policy Framework. Adelaide,

Australia: Government of South Australia Retrieved from http://www.spb.sa.gov.au/site/policies_guides.aspx

Stiglitz, J. E. (1985). Information and economic analysis: A perspective. The

Economic Journal, 95, 21-41.

Sweeney, S. (2009). Addressing market failure: Using transaction cost economics to

improve the construction industry's performance PhD Thesis. University of Melbourne, Department of Civil and Environmental Engineering.

References 293

Sweeney, S., & Duffield, C. F. (2006). Why is the selection of an appropriate project delivery methodology so hard? Paper presented at 1st International Construction Specialty Conference, Calgary, Alberta, Canada.

Tadelis, S. (2002). Complexity, flexibility, and the make-or-buy decision. The

American Economic Review, 92(2), 433-437.

Tadelis, S., & Williamson, O. (2010). Transaction Cost Economics. Berkeley

University of California.

Tang, L., Shen, Q., Skitmore, M., & Cheng, E. W. L. (2013). Ranked Critical Factors

in PPP Briefings. Journal of Management in Engineering, 29(2), 164-171.

Teece, D. J. (1980). Economies of scope and the scope of the enterprise. Journal of

Economic Behavior and Organization, 1(3), 223-247.

Teece, D. J., Pisano, G., & Shuen, A. (1997). Dynamic capabilities and strategic

management. Strategic Management Journal, 18(7), 509-533.

Teece, D. J., Rumelt, R., Dosi, G., & Winter, S. (1994). Understanding corporate

coherence: Theory and evidence. Journal of Economic Behavior & Organization, 23(1), 1-30.

Teisman, G. T., & Klijn, E.-H. (2001). Public-Private Partnerships in the European

Union: Official Suspect, Embraced in Daily Practice. In Public-Private Partnerships: Theory and Practice in International Perspective. New York: Routledge.

Teo, P., Bridge, A., & Gray, J. (2013, 5-9 May 2013). A new first-order decision-

making model for the procurement of public sector infrastructure: Procedures and Testing. In Proceedings of the 19th CIB World Building Congress, Brisbane 2013: Construction and Society.

Teo, P., Bridge, A., Gray, J., & Jefferies, M. C. (2011, 5-7 September 2011).

Developing a research method to test a new first-order decision making model for the procurement of public sector major infrastructure. In Proceedings of Association of Researchers in Construction Management Annual Conference (ARCOM 2011).

Teo, P., Bridge, A., Gray, J., & Rowlinson, S. (2012, 23 – 25 January 2012).

Towards testing a new first-order decision making model for the procurement of public sector major infrastructure. In Proceedings of International Conference on Facilities Management, Procurement Systems and Public Private Partnership (CIB W070, W092 and TG72).

294 References

Teo, P., Bridge, A., & Jefferies, M. C. (2010). Delivering value for money in the

procurement of public sector major infrastructure: A new first-order decision making model. In Proceedings of 2010 International Conference On Construction & Real Estate Management (pp. 306-311).

The Audit Office of New South Wales. (2006). The new schools privately financed

project. Sydney: Author.

Thomas, P. (2011). The evidence for and against PFI. Infrastructure Journal.

Trochim, W. (1989). Outcome pattern matching and program theory. Evaluation and

Program Planning, 12(4), 355-366.

Turner, J. R. (2004). Farsighted project contract management: incomplete in its

entirety. Construction Management and Economics, 22(1), 75-83.

Turner, J. R., & Simister, S. J. (2001). Project contract management and a theory of

organization. International Journal of Project Management, 19(8), 457-464.

Venkatesan, R. (1992). Strategic sourcing: To make or not to make. Harvard

Business Review(November-December), 98-107.

Walker, A. (2007). Project management in construction (5th ed.). Oxford, UK,

Malden, MA, USA: Blackwell Publishing.

Walker, A., & Chau, K. W. (1999). The relationship between construction project

management theory and transaction cost economics. Engineering, Construction and Architectural Management, 6(2), 166-176.

Walker, B., & Walker, B. C. (2000). Privatisation: Sell off or sell-out? The

Australian experience. Sydney: ABC Books.

Walker, G., & Weber, D. (1984). A transaction cost approach to make-or-buy

decisions. Administrative Science Quarterly, 29(3), 373-391.

Walliman, N. (2011). Research methods: The basics. London, New York: Routledge.

Weimer, D. L., & Vining, A. R. (1999). Policy analysis - Concepts and practice (3rd

ed.). Upper Saddle River, N.J.: Prentice Hall.

References 295

Wernerfelt, B. (1984). A resource-based view of the firm. Strategic Management Journal, 5(2), 171-180.

Williamson, O. E. (1971). The vertical integration of production: Market failure

considerations. The American Economic Review, 61(2), 112-123.

Williamson, O. E. (1975). Markets and hierarchies, analysis and antitrust

implications: A study in the economics of internal organization. New York: Free Press.

Williamson, O. E. (1981). The economics of organization: The transaction cost

approach. The American Journal of Sociology, 87(3), 548-577.

Williamson, O. E. (1985). The economic institutions of capitalism: Firms, markets,

relational contracting. New York, London: Free Press, Collier Macmillan.

Williamson, O. E. (1991). Comparative economic organization: The analysis of

discrete structural alternatives. Administrative Science Quarterly, 36(2), 269-296.

Williamson, O. E. (1996). The mechanisms of governance. N.Y.: Oxford University

Press.

Williamson, O. E. (1999). Strategy research: Governance and competence

perspective. Strategic Management Journal, 20(12), 1087-1108.

Williamson, O. E. (2000). The new institutional economics: Taking stock, looking

ahead. Journal of Economic Literature, 38(3), 595-613.

Williamson, O. E. (2005a). The economics of governance. The American Economic

Review, 95(2), 1-18.

Williamson, O. E. (2005b). Transaction cost economics and business administration.

Scandinavian Journal of Management, 21(1), 19-40.

Williamson, O. E. (2008). Outsourcing: Transaction cost economics and supply

chain management. Journal of Supply Chain Management, 44(2), 5-16.

Williamson, O. E. (2010). Transaction cost economics: The natural progression.

Journal of Retailing, 86(3), 215-226.

296 References

Wilson, R. (1977). A bidding model of perfect competition. The Review of Economic Studies, 44(3), 511-518.

Winch, G. (1989). The construction firm and the construction project: a transaction

cost approach. Construction Management and Economics, 7(4), 331-345.

Winch, G. (2001). Governing the project process: a conceptual framework.

Construction Management & Economics, 19(8), 799-808.

Winch, G. (2002). Managing construction projects: An information processing

approach. Malden, M.A.: Blackwell Science.

Winston, C. (2006). Government failure versus market failure: Microeconomics

policy research and government performance. Washington, D.C.: The Brookings Institution Press and AEI.

Winter, S. G. (1988). On Coase, competence, and the corporation conference papers

to celebrate the fiftieth anniversary of the "Nature of the firm". The Journal of Law, Economics, & Organization, 4(1), 163-180.

Yin, R. K. (2009). Case study research: Design and methods (4th ed.). Thousand

Oaks, Calif.: Sage Publications.

Zander, U., & Kogut, B. (1995). Knowledge and the Speed of the Transfer and

Imitation of Organizational Capabilities: An Empirical Test. Organization Science, 6(1), 76-92.

Zerbe, R. O., & McCurdy, H. E. (1999). The failure of market failure. Journal of

Policy Analysis and Management, 18(4), 558-578.

Appendix A – List of MAUA-based procurement approaches 297

Appendix A – List of MAUA-based procurement approaches

Adapted from (Ambrose & Tucker, 2000; Love, Davis, Edwards, & Baccarini, 2008)

No Category Author Year Description

Simple scoring/weighting 1 Simple NEDO 1985 Procurement path decision chart. Use of a rating system

using client’s priorities for nine criteria

2 Simple Franks 1990 Simple rating system of criteria against a limited number of procurement options

3 Simple Griffith and Headley

1997 Use of weightings to assess criteria and procurement options for small building works. Simple and easy to use

4 Simple NSW Department of Commerce

2006 Weighting of client priorities and procurement method to achieve the priorities. Simple to use but too many criteria

5 Simple weighted scoring

Love et al. 2011 6 step procurement selection process which involves participation of stakeholders in focus groups or workshops to determine a suitable procurement mode based on quantitative scoring and weighting of NEDO’s project objectives on time, certainty, complexities , time competition, quality, risks and responsibility; against various procurement options; and qualitative review process.

Multi-attribute utility approach (MAUA)

6 MAUA Skitmore and Marsden

1988 Use of multi-attribute utility analysis based on NEDO with a rating system and weighting of client priorities, furthermore discriminant and concordance analysis is also applied.

7 MAUA Bennett and Grice 1990 Based on NEDO and Skitmore and Marsden’s model using MAUA. Enables client’s to weight specific criteria multiplied by a set of utility ratings for various procurement options

8 MAUA Singh 1990 Furthered MAUA using more experts to validate the model 9 Organisation

roles incl. MAUA

Liu 1994 Organisational behaviour-based model utilising an act-to-outcome process governed by organisational goals, which are subject to moderators and determine performance relationships, using conjoint analysis to model decision maker’s judgement profile

10 Based on Bennett & Grice

Chan 1995 Utilises the Bennett and Grice model, but uses a different procurement category developed for the Australian construction industry

11 Skitmore Love et al. 1998 Based on Skitmore and Marsden’s model, and tested widely throughout Australia

Delphi Approach

12 Delphi +MAUA

Chan et al. 2001 Application of Delphi method in selection of procurement systems for construction projects

13 QBS + Delphi+Fuzzy

Manoliadis et al. 2009 Improving QBS by use of fuzzy Delphi method

Expert System (Computer-based system) 14 Knowledge

system Brandon et al. 1988 ELSIE – a computer expert system based on project

characteristics and client requirements. Subjective and contained a limited number of procurement options

298 Appendix A – List of MAUA-based procurement approaches

No Category Author Year Description 15 Expert system Moshini and

Botros 1990 PASCON – an expert system similar to ELSIE, which

evaluates and generates alternate procurement strategies based on the client's requirements, posture towards project control, risk taking & constraints of time and costs.

16 Decision support system

Kumaraswamy and Dissanayaka

2001 This was developed into a computerised knowledge-based expert system. Quite comprehensive in arriving at procurement. Not able to update system database.

17 CBR Luu et al. 2005 Case-based reasoning – capture and reuse of experiential knowledge from previous projects for procurement decisionmaking. Project characteristics, client characteristics and external environment taken into account.

18 DSS- web base Molenaar & Songer

2001 Web-based DSS that assist public sector owners in selection of projects suitable for design/build projects- simple to use. It involves filling in project characteristics, success criteria, project & personal data and the DSS will generate advice and graphs.

Analytical hierarchy process (AHP) - mathematical model 19 Weighting +

AHP Alhazmi and McCaffer

2000 Allows users to choose from a reduced number of prescribed strategies and alternative contract types. Sue of weighting/ranking systems juxtaposed with AHP. Very complex system to arrive a procurement option.

20 MAUT + AHP Cheung et al. 2001 Use of MAUT and analytical hierarchy process (AHP). NEDO criteria used. Utility factors corresponding to various procurement strategies established. To cater for individual project characteristics, the relative weightings of the selection criteria are assessed using AHP (which improves the objectivity of the weightings).

21 AHP Altabtabai 2002 Select project delivery system using AHP. 22 AHP Al Khalill 2002 Selecting project delivery method using AHP, against

limited procurement options. Matrix based 23 Matrix based Dell’Isola et al. 1999 Decision matrix-based model that rates the performance

of each procurement system for selected issues and their relative importance on a client/project.

24 Matrix based SRD Consulting 2000 Suitability matrices developed for Qld Dept of Main Roads. Scoring and rating to pre-determine optimum project deliver system.

25 Matrix based Construction Industry Institute, CII

2001 Project delivery selection workbook. Suitability matrix. Rates critical project goals by level of importance, scores each goal and ranks the most critical metrics. Limited options optimum project delivery system and prescribes.

26 Matrix based DST

Sidwell et al. 2003 Development of best practice decision matrix, includes a computerised Decision support tool (DST) which helps to derive a Project delivery method, by answering a series of questions regarding 10 key project attributes.

27 3D model Ambrose and Tucker

2000

3D model that includes three dimensions of client needs, project characteristics and procurement process features.

Procurement selection framework 28 Procurement

Method Toolkit Cooperative Research Centre for Construction Innovation, CRC

2008 Simple weighting of client priorities and scoring procurement methods to achieve the priorities.

29 5 stage Procurement options analysis

Grimsey & Lewis 2009 Comprises of 5 steps, namely data gathering, bundling analysis, procurement validation, procurement options analysis and lastly structuring and design of preferred procurement option.

Appendix B – Survey of major road and health projects (final version) 299

Appendix B – Survey of major road and health projects (final version)

Queensland University of Technology Faculty of Built Environment and Engineering

School of Urban Development

Australian Research Council Linkage Grant

Reforming the procurement of construction and financing of Australian infrastructure: Advancing capacity, competition and investment

Project 1: Construction capacity and competition in the

Australian major infrastructure market

Data collection instrument: Actual procurement and competition - information required schedule

300 Appendix B – Survey of major road and health projects (final version)

Dear Participant The purpose of the data being collected in this schedule is to identify relevant projects that fall within this study’s parameters (either a road or hospital/health care related project with a capital value over $50million and in which the construction contract was executed and work commenced between the period July 2006 and June 2010). Our goal is to collect a completed schedule for all projects falling within these parameters in five states - NSW; QLD; SA; VIC; and in WA. Beyond identifying projects falling within the above parameters, the data in this schedule will allow the research team to begin to establish a feel for the actual procurement approach taken and the actual level of construction competition observed in relation to each project represented by a completed schedule. In the next step of this research project this year, the actual level of construction competition will then be reconciled with potential competition surrounding the project concerned and which is based on construction capacity derived from a nationwide survey of contractors. In 2011, a number of projects will then be case studied to compare the actual approach taken to procurement with that predicted from theory. This research project aims to allow an improved understanding of the effects of procurement on construction capacity and competition and one of the outcomes of this project is the development of a new and early / first-order procurement decision making model for major infrastructure. The PhD researcher in this project (Pauline Teo) plans to visit you later in this year to discuss the project(s) identified in your completed schedule(s). In the meantime, if you have any queries relating to any aspect of this schedule, then please don’t hesitate to contact Pauline: Email: [email protected] Phone: 07 3138 8487 Mobile: 0430 609191 The data being collected in this schedule contributes towards the first of five projects in an Australian Research Council grant. This schedule begins after the following participant information that includes details of the overarching research grant within which this Project 1 and schedule sits. On behalf of the entire research team and partner organisations, thank you very much indeed for your help in completing and submitting this on-line schedule. Kind regards


PARTICIPANT INFORMATION for QUT RESEARCH PROJECT


infrastructure: Advancing capacity, competition and investment



Research Team Contacts

Pauline Teo – PhD Candidate Dr Adrian Bridge

School of Urban Development School of Urban Development

Phone 07 3138 8487 Phone 3138 1543

Email [email protected] Email [email protected]

Description This project is being undertaken as part of ARC Linkage Grant. The grant is funded by: • Australian Research Council

• Partnerships Victoria, Department of Treasury and Finance, Victoria

• New South Wales Treasury

• Queensland Treasury

• Western Australia Department of Treasury and Finance

• South Australia Department of Treasury and Finance

• Queensland Department of Infrastructure and Planning

• Construction Industry Institute Australia

• Infrastructure Association Queensland

• Infrastructure Partnerships Australia

• Coffey Commercial Advisory

• Aurecon

None of the funding bodies will have access to any of the data obtained during the project - outside of a funding body’s own data. The purpose of this grant is to develop a new and integrated procurement knowledge base to help address construction and finance constraints. The project outcomes will form the basis of procurement reform to increase construction capacity, competition and private sector investment (particularly by superannuation funds) to infrastructure. The purpose of Project 1 in this grant is to develop theory to explain the effect of procurement on the extent of competition and capacity amongst contractors in Australia’s infrastructure market, and in doing so develop a new early / first-order procurement decision making model. The research team requests your assistance because the data being collected in this schedule is held by government departments and agencies responsible for delivering road and hospital infrastructure. Participation

Your participation in this project is voluntary. If you do agree to participate, you can withdraw from participation at any time during the project without comment or penalty. Your decision to participate will in no way impact upon your current or future relationship with QUT. Your participation will involve the completion of this schedule and a short meeting with the PhD Candidate to discuss the


project and clarify details provided in the schedule, and which will take approximately 30 minutes for the schedule and a further 20 minutes to participate in the meeting. Expected benefits Government and private sector will benefit from the development and availability of new decision making models and financial models that are designed to advance Australia's position in the delivery of infrastructure in terms of increasing construction capacity and investment to infrastructure and improving value for money tax payers. The Australian Research Council and general academic community will benefit in terms of advancing knowledge and in training researchers. Risks

There are no out-of-the ordinary individual or business risks associated with your participation in this project. Confidentiality All comments and responses will be treated confidentially. None of the funding bodies will have access to any of the data obtained during the project - outside of a funding body’s own data. Only aggregated results and other reports with non-identifiable data will be made available to funding bodies and in any publications. Consent to participate The return of the completed questionnaire is accepted as an indication of your consent to participate in this project. That is, you are indicating that you:

• have read and understood the information document regarding this project

• have had any questions answered to your satisfaction

• understand that if you have any additional questions you can contact the research team

• understand that you are free to withdraw at any time, without comment or penalty

• understand that you can contact the Research Ethics Officer on +61 7 3138 5123 or [email protected] if you have concerns about the ethical conduct of the project

• agree to participate in the project

Questions / further information about the project Please contact the researcher team members named above to have any questions answered or if you require further information about the project. Concerns / complaints regarding the conduct of the project QUT is committed to researcher integrity and the ethical conduct of research projects. However, if you do have any concerns or complaints about the ethical conduct of the project you may contact the QUT Research Ethics Officer on +61 7 3138 5123 or email [email protected]. The Research Ethics Officer is not connected with the research project and can facilitate a resolution to your concern in an impartial manner.

Thank you again for helping with this research project. Please print-out a copy of this Participant Information for your information.


INFORMATION REQUIRED SCHEDULE Section A: Project identification Please identify the project by inserting answers to Questions 1-6.

1. The project’s name and contract/job number:__________________________________________

2. Government agency or department responsible for the delivery of project:___________________

3. The date project commenced or letter of acceptance (month / year):_______/_______

4. The date the project was practically completed/approximate date the project is expected to be

completed (month / year):_______/_______

5. The project’s location (state and region within state): ___________________________________

6. The approximate overall capital value of the project (contract sum/estimate - to the nearest $1

million): $_______million

Section B: Project scope and procurement Please indicate the project’s scope and procurement by answering either Question 7

or Question 8 - depending whether the project was procured as one contract or more

than one contract.

7. If the project was/is being procured as a single private sector contract, then please state method of

delivery used/being used (please use term most familiar to the government agency/department

responsible for the project):________________________________________________________

a. In this single contract, please briefly indicate within this contract the nature and

scope of the

i. New construction / installation:___________________________________

b. Where applicable, please also briefly indicate within this contract the nature and

scope of any

i. Design:______________________________________________________

ii. Operations and maintenance:_____________________________________

iii. Private finance:________________________________________________

c. Please describe the type of contract (e.g. name of standard contract):_____________

d. Please also describe the nature of the payment terms (e.g. cost plus including

pain/gain share; remeasurement; fixed price; target cost):_____________________


8. If the project was/is being procured using more than one private sector contract, then in relation to the largest contracts in terms of capital value (up to the largest 8 contracts), please briefly indicate:

a. The scope of the service(s) being provided in each contract (including any design, construction, operations and maintenance and private finance) and the nature of the new installation works that is the subject of the service(s) being provided in each contract:

i. Contract 1:___________________________________________________

____________________________________________________________

ii. Contract 2: ___________________________________________________

____________________________________________________________

iii. Contract 3:___________________________________________________

____________________________________________________________

iv. Contract 4: ___________________________________________________

____________________________________________________________

v. Contract 5: ___________________________________________________

____________________________________________________________

vi. Contract 6: ___________________________________________________

____________________________________________________________

vii. Contract 7: ___________________________________________________

____________________________________________________________

viii. Contract 8:___________________________________________________

b. Which of the above contracts in Question 8a contains the largest civil or building construction component?:_______________________________________________

c. With regard to the contract entered at Question 8b. and which contains the largest civil or building construction component:

i. Please state the method of delivery used/being used (please use term most familiar to the government agency/department responsible for the project):____________________________________________________

ii. Please also describe the type of contract (e.g. name of standard contract):___________________________________________________

iii. And please describe the payment terms (e.g. cost plus including pain/gain share; remeasurement; fixed price; target cost):_____________


Section C: Tendering procedure – concerning contract with largest civil or building construction component Please insert a brief description concerning the tendering procedure to select the contractor / consortium for the project, or in the case that the project comprises more than one contract, the contractor / consortium undertaking the contract with the largest civil or building construction component (inserted at Question 8b). 9. This description might include where applicable brief details concerning:

• Market sounding (particularly for a PPP) – including scope of private sector contacted and use of any industry representative bodies

• Expressions of interest

• Early Contractor Involvement; Negotiation with preferred contractor (following expressions of interest)

• Invitation to tender / request for proposals

• Brief reference to published codes of tendering practice or probity principles used

• Brief summary of any innovations or beyond business as usual practice used in the tendering procedure

____________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________ Please give some further details concerning expressions of interest in bidding for the project, or in the case that the project comprises more than one contract, for the contract with the largest civil or building construction component - inserted at Question 8b. 10. Please insert names of contractors or consortia that expressed interest in the following table; If

more than 30 contractors expressed an interest, then please insert total number:____________

a. g. m. s. y.

b. h. n. t. z.

c. i. o. u. aa.

d. j. p. v. ab.

e. k. q. w. ac.

f. l. r. x. ad.


11. If a prequalification listing was used to seek expressions of interest, then were all the contractors in the relevant category notified? Please circle: Yes / No / Prequalification listing not used

12. If a public notification was used to seek expressions of interest, then did all parties indicting their intention to express interest receive full expression of interest documentation or did provision of expression of interest documentation follow assessment of some initial information from the contractor / consortium? Please circle:

a. Expression of interest documentation to all

b. Expression of interest documentation upon satisfactory provision of initial information from potential bidder

c. Public notification not used

Please give some further details concerning the invitation to bid / request for proposals for the project, or in the case that the project comprises more than one contract, for the contract with the largest civil or building construction component - inserted at Question 8b.

13. Please insert names of contractors or consortia invited to bid /submit a proposal:

a. e.

b. f.

c. g.

d. h.

14. What was the rationale for the number of contractors/consortia invited to

bid?:________________________________________________________________________________________________________________________________________________________

15. Would your agency/department liked to have been able to invite more bids/proposals than that shown in Question 13? Please circle: Yes / No

Please give some further information concerning final bids for the project, or in the case that the project comprises more than one contract, for the contract with the largest civil or building construction component - inserted at Question 8b. 16. Please insert names of contractors or consortia that submitted a bid / proposal:

a. e.

b. f.

c. g.

d. i.


17. Was the accepted bid appreciably different (greater than 15% difference) than the budget - any pre-tender estimate or target/benchmark cost? Please circle: Significantly higher / About the same / Significantly lower

18. Where the bid included a significant design component and correspondingly one or more of the bid evaluation criteria concerned design quality/innovation, then how much did the accepted bidder’s score(s) in terms of design quality/innovation reflect expectations? Please circle: Expectations exceeded / expectations met / below expectations / not applicable

19. Where the bid allowed for the time period for construction to be determined by the contractor / maximum time to be improved upon by the contractor, and correspondingly one or more of the bid evaluation criteria concerned the time for construction, then how much did the accepted bidder’s score(s) in terms of time reflect expectations? Please circle: Expectations exceeded / expectations met / below expectations / not applicable

Section D: Comments and feedback Any comments or feedback you wish to make are very much welcomed. Please insert: ________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________________

Appendix C – Coding of survey of major road and health projects 309

Appendix C – Coding of survey of major road and health projects

Question coding

Coding

Section A project id number entered coding 0= road, 1= hospital q1 'project name' entered q2 'department involved' entered q3 date in month and year q4 date in month and year q51 coding 1=NSW, 2=QLD, 3=VIC, 4=WA, 5=SA q52 region entered q6 value in millions Section B q7i coding 1= single contract, 2= multiple contracts q7ii coding 1=C, 2=DC, 3=DCM, 4=DCOM, 5=CM, 6=MC, 7=ECI, 8=AC, 9=PPP q7a Enter description q7b1 Enter description q7b2 Enter description q7b3 Enter description q7c coding 1=AS 2124 q7d coding 1=fixed periodic payments (PPP), 2=lump Sum (Fixed price), 3 =

GMP/GCS/Target out-turn cost, pain/gain share, 4=Remeasurement/ Schedule of rates and prices/ Cost plus including pain/gain share

q8a1 Enter description q8a2 Enter description q8a3 Enter description q8a4 Enter description q8a5 Enter description q8a6 Enter description q8a7 Enter description q8a8 Enter description q8b Enter description q8c1 Enter description; coding 1=C, 2=DC, 3=DCM, 4=DCOM, 5=CM, 6=MC, 7=ECI, 8=AC,

9=PPP q8c2 Enter description; coding 1=AS 2124 q8c3 Enter description; coding 1=fixed periodic payments (PPP), 2=lump Sum (Fixed price), 3

= GMP/GCS/Target out-turn cost, pain/gain share, 4=Remeasurement/ Schedule of rates and prices/ Cost plus including pain/gain share

Section C q9 Enter description q10 Expression of Interest: Number of contractors entered q11 Coding 0=prequal not used, 1=no, 2=yes q12 Coding 1=EOI documentation to all 2=EOI documentation from potential bidder 3=public notification not used q13 Request for proposal: Number of contractors entered q14 Enter description

310 Appendix C – Coding of survey of major road and health projects

q15 Coding 1=no; 2=yes q16 Submit bids: Number of contractors entered q17 coding 0=appreciably lower; 1=about the same; 2= appreciably higher q18 coding 0=na; 1=below expectations; 2=expectations met; 3=expectations exceeded q19 coding 0=na; 1=below expectations; 2=expectations met; 3=expectations exceeded Section D q20 Enter description

Appendix D – Survey of civil and building contractors (final version) 311

Appendix D – Survey of civil and building contractors (final version)

Queensland University of Technology Faculty of Built Environment and Engineering

School of Urban Development

Australian Research Council Linkage Grant

Reforming the procurement of construction and financing of Australian infrastructure: Advancing capacity, competition and investment



Data collection instrument: Potential construction competition (actual construction capacity) –

Contractors’ questionnaire (Version for Building Contractors operating in New South Wales)

312 Appendix D – Survey of civil and building contractors (final version)

Dear Participant The purpose of the data being collected in this version of the contractors’ questionnaire is to allow the research team to form a view concerning potential competition amongst building contractors operating in New South Wales (based on actual capacity) for the construction of public sector major building infrastructure projects (including hospital/health care projects) with a capital value over $50million and in which at least expressions of interest were established between the period July 2005 to June 2010. Please note, you are asked to give answers in this questionnaire concerning your firm’s building construction operations only in New South Wales. If your firm also offers civil construction services in your state, then your Statewide Manager is being asked to arrange the submission of a separate version of this questionnaire concerning Civil Contractors and if your firm operates in more than one state, then your colleagues in this(ese) other state(s) are also being invited to complete this questionnaire and/or the Civil Contractors’ version of this questionnaire with respect to this(ese) other state(s). This research project aims to allow an improved understanding of the effects of procurement on construction capacity and competition in public sector major infrastructure and one of the key outcomes of this project is the development of a new procurement decision making model for major infrastructure. Our goal is to collect a completed/substantially completed questionnaire from 95 percent or more of all civil and building contractors in Australia that are capable of delivering a $50million project. If you have any queries relating to any aspect of this questionnaire, then please don’t hesitate to contact the PhD researcher (Ms Pauline Teo) undertaking this project. Pauline’s contact details are as follows: Email: [email protected] Phone: 07 3138 7419 Mobile: 0430 609191 If you tick the box at Question 1, then we will send you a summary of the analysis of aggregated data from this questionnaire and from the civil contractors’ questionnaire. The data being collected in this questionnaire contributes towards the first of five projects in an Australian Research Council grant. This questionnaire begins after the following participant information that gives details of the overarching research grant - within which this questionnaire and first project sit. On behalf of the entire research team and partner organisations, thank you very much indeed for your help in completing this questionnaire and your important contribution. Kind regards


PARTICIPANT INFORMATION for QUT RESEARCH PROJECT


infrastructure: Advancing capacity, competition and investment



Research Team Contacts

Pauline Teo – PhD Candidate Dr Adrian Bridge

School of Urban Development School of Urban Development

Phone 07 3138 7419 Phone 3138 1543

Email [email protected] Email [email protected]

Description This project is being undertaken as part of ARC Linkage Grant. The grant is funded by:

• Australian Research Council

• Partnerships Victoria, Department of Treasury and Finance, Victoria

• New South Wales Treasury

• Queensland Treasury

• Western Australia Department of Treasury and Finance

• South Australia Department of Treasury and Finance

• Queensland Department of Infrastructure and Planning

• Construction Industry Institute Australia (including Queensland Department of Transport and Main Roads and Queensland Department of Public Works)

• Infrastructure Association Queensland

• Infrastructure Partnerships Australia

• Coffey Commercial Advisory

• Aurecon

None of the funding bodies will have access to the data obtained in this questionnaire. The purpose of the overall grant is to develop a new and integrated procurement knowledge base to help address construction and finance constraints. The project outcomes will be considered by government as the basis of possible procurement reform to increase construction capacity, competition and private sector investment (particularly by superannuation funds) to public sector infrastructure. The purpose of Project 1 in this grant is to develop theory to explain the effect of procurement on the extent of competition and capacity amongst civil and building contractors in Australia’s infrastructure market, and in doing so develop a new procurement decision making model. The research team requests your assistance because the data being collected in this questionnaire is held by contractors bidding for state proposed road and hospital/health care related infrastructure. Participation Your participation in this project is voluntary. If you do agree to participate, you can withdraw from participation at any time during the project without comment or penalty. Your decision whether to participate will in no way impact upon your current or future relationship with QUT or any of the funding


bodies/organisations supporting the research. Your participation will involve the completion of this questionnaire and which will take approximately 30 minutes - once your data has been assembled.

Expected benefits Government and private sector will benefit from the development and availability of new decision making models and financial models that are designed to be considered by government to advance Australia's position in the delivery of public sector infrastructure in terms of increasing construction capacity and investment to infrastructure and improving value for money for tax payers. The Australian Research Council and general academic community will benefit in terms of advancing knowledge and in training researchers.

Risks There are no out-of-the ordinary individual or business risks associated with your participation in this project. Confidentiality All comments and responses will be treated confidentially. None of the funding bodies will have access to any of the data obtained in this questionnaire. Only aggregated results and other reports with non-identifiable data will be made available to funding bodies and in any publications. Consent to participate The return of the completed questionnaire is accepted as an indication of your consent to participate in this project. That is, you are indicating that you:

• have read and understood the information document regarding this project

• have had any questions answered to your satisfaction

• understand that if you have any additional questions you can contact the research team

• understand that you are free to withdraw at any time, without comment or penalty

• understand that you can contact the Research Ethics Officer on +61 7 3138 5123 or [email protected] if you have concerns about the ethical conduct of the project

• agree to participate in the project

Questions / further information about the project Please contact the research team members named above if you have any questions or if you require further information about the project.

Concerns / complaints regarding the conduct of the project QUT is committed to researcher integrity and the ethical conduct of research projects. However, if you do have any concerns or complaints about the ethical conduct of the project you may contact the QUT Research Ethics Officer on +61 7 3138 5123 or email [email protected]. The Research Ethics Officer is not connected with the research project and can facilitate a resolution to your concern in an impartial manner.

Thank you again for helping with this research project. Please print-out a copy of this Participant Information for your information.


QUESTIONNAIRE

Section A: General information Please note that unless otherwise indicated all questions concern your firm’s building construction services (including hospital/health care projects) and as head/main contractor in relation to all building projects located in New South Wales and excludes any civil projects located in New South Wales and excludes any domestic and international projects administered in New South Wales but located outside of New South Wales. If your firm offers civil construction services in New South Wales, then your Statewide Manager is being asked to arrange the submission of a separate version of this questionnaire concerning Civil Contractors and if your firm operates in more than one state, then your colleagues in this(ese) other state(s) are also being invited to complete this questionnaire and/or the Civil Contractors’ version of this questionnaire with respect to this(ese) other state(s). 1. If you would like to receive a summary of the analysis of aggregated data from this questionnaire

and from the Civil Contractors’ questionnaire, then please tick the following box: During the process of completing this questionnaire, if you would like to make any further comments, then these are welcomed in the final Section E and which comprises a free-text section. Section B: Sector analysis This section seeks to generate data that will allow the research team to divide-up the building construction market in New South Wales into sectors, or pools of firms, whose pricing decisions may affect each other. To do this, a project type (public sector hospital/health care) and project size (over $50million capital value), two time periods (pre-GFC and amidst/post GFC), the geographical scope of your firm’s civil construction operations within New South Wales and the scope of your firm’s building construction services (procurement modes offered) are deployed. In terms of procurement mode, the following abbreviations are used: • C = Construct only (lump sum or fixed price contract)

• D&C = Design and Construct

• DC&M = Design and Construct and Maintain

• DCO&M = Design and Construct and Operate and Maintain

• CM = Construction Management

• MC = Managing Contractor

• ECI = Early Contractor Involvement

• AC = Alliance Contracting

• PF =Any procurement mode that includes private finance e.g. a Public Private Partnership

Please note in the above procurement modes: • Design - could be delivered by your firm’s in-house designers and / or by independent design

firms appointed directly by your firm, or represented by your firm’s capability to form a consortium that includes an independent design firm bidding for a new project; and/or

• Operations & Maintenance (O&M) (or Facility Management – FM) - could be delivered by your firm’s in-house department / subsidiary and / or by independent O&M/FM firm(s) appointed directly by your firm, or represented by your firm’s capability to form a consortium that includes an independent O&M/FM firm(s) bidding for a new project; and/or

• Private finance - concerns your firm’s capability to form a consortium that includes an independent private financier and which generates debt and / or equity finance to bid for a new project e.g. a conventional Public-Private Partnership.


2. Please insert in each cell an upper bid limit value offered (to the nearest $25 million) as both a single entity bidder (in competition with other bidders) and as part of a joint bid (e.g Joint Venture or Consortium, again in competition with other bidders) to indicate that your firm was capable and generally willing to bid for public sector (federal/state/local government proposed) hospital/health care projects over $50million (capital value) in New South Wales and in terms of the procurement mode and time period represented by that cell. If your firm was not capable or not willing to bid for projects using a particular procurement mode and in the time period represented by a cell, then please tick the box at the bottom of the cell concerned.

ProcurementMode

Offered (see above

key)

Public sector hospital/health care projects – each project over $50million

Pre-Global Financial Crisis July 2005-June 2008

Amidst/ Post- Global Financial Crisis July 2008-June 2010

C Upper value offered (to nearest $25million), as - As single entity bid: $__________& as joint bid: $________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �

Upper value offered (to nearest $25million), as - As single entity bid: $_________& as joint bid: $_________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �

D&C Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �

Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $______________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �

DC&M Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


DCO&M Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


CM Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


MC Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


ECI Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


AC Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �


PF Upper value offered (to nearest $25million), as - As single entity bid: $______________& as joint bid: $_____________ Not capable or willing to bid using the procurement mode in this row and in the time period in this column �



3. Across the procurement modes offered by your firm – as you noted in the previous question, were there any locations in New South Wales in which your firm was not willing to bid for public sector (federal/state/local government proposed) hospital/health care projects over $50million (capital value) and in the two time periods below:

a. Pre-Global Financial Crisis (July 2005-June 2008): Please insert any locations in New South Wales in which your firm was unwilling to bid:____________________________

b. Amidst/Post-Global Financial Crisis (July 2008-June 2010): Please insert any locations in New South Wales in which your firm was unwilling to bid:_____________________


Section C: Analysis of structure-conduct-performance of firms in each sector Having used the previous data to establish broad sectors of firms in the market for constructing major public sector hospital/health care projects in New South Wales, this section seeks to generate data concerning your firm’s actual and projected capacity, as well as your firm’s actual performance and which will enable the research team to more accurately assess the potential level of competition in each sector in New South Wales. 4. Please insert answers in the following table concerning your firm’s actual and projected total

building construction capacity in terms of all building construction projects in New South Wales; that is, all values and all building project types - in both public sector (federal; state local government funded projects) and private sector (any other sources of client funding beyond public sector) and in each of the last five years (actual capacity), as well as in the financial year to June 2011 (projected capacity):

Your firm’s actual/projected total building construction capacity - all building construction projects in NSW

July 2005 -

June 2006

July 2006 -

June 2007

July 2007 -

June 2008

July 2008 -

June 2009

July 2009 -

June 2010

July 2010 -

June 2011 a. Approximate total building construction

turnover in NSW (to the nearest $25million)? $_____ (actual)

$_____ (actual)

$_____ (actual)

$_____ (actual)

$_____ (actual)

$_____ (projected)

b. Approximate percentage contribution made by all public sector projects to your firm’s total building construction turnover in NSW?

______% (actual)

______% (actual)

______% (actual)

______% (actual)

______% (actual)

______% (projected)

c. Approximate percentage contribution made by public sector hospital/health care projects to your firm’s total building construction turnover in NSW?

______% (actual)

______% (actual)

______% (actual)

______% (actual)

______% (actual)

______% (projected)

d. Approximate number of employees directly employed by your firm in NSW, including all salaried employees, permanent or temporary staff but excluding waged/project-based workforce, subcontractors and agency and labour hire workers?

______no (actual)

______no (actual)

______no (actual)

______no (actual)

______no (actual)

______no (projected)

e. Approximate percentage of direct employees in your firm in NSW employed as professional engineers?

_______% (actual)

_______% (actual)

_______% (actual)

_______% (actual)

_______% (actual)

_______% (projected)

f. Using one of the 4 descriptors in the dot-points below, how would you best describe your firm’s total building construction capacity in NSW?:

1 Significant spare capacity (operating at 85% or less capacity); or

2 Spare capacity (operating between 85-95% capacity); or

3 Near full capacity (operating between 95-100% capacity); or

4 Full/overstretched capacity

______ ______ ______ describe capacity (actual)





______ ______ ______ describe capacity (projected)

g. At any point in each of the financial years and due to capacity constraints, was your firm unwilling to express interest or bid for public sector building construction projects over $50million in NSW?

______ yes/no (actual)

_____ yes/no (actual)




______ yes/no (projected)

h. At any point in each of the financial years and due to capacity constraints, was your firm unwilling to express interest or bid for private sector building construction projects over $50million in NSW?

_______ yes/no (actual)





_______ yes/no (projected)


5. Please insert answers in each cell concerning your firm’s services or procurement modes delivered in New South Wales just prior to the Global Financial Crisis (July 2007-June 2008). If your firm did not offer one of the procurement modes in this period, then please insert “Not Applicable” in the column under the procurement mode concerned. Please note, in terms of procurement mode, the following abbreviations are used again:

• C = Construct only (lump sum or fixed price contract)








• PF = Any procurement mode that includes private finance e.g. a Public Private Partnership

Your firm’s procurement modes delivered in NSW just prior to the Global Financial Crisis (July 2007-June 2008):

C D&C DC&M DCO&M CM MC ECI AC PF

Approximate percentage each procurement mode contributed to all public sector building construction projects and turnover in NSW between July 2007 and June 2008? (percentages in this row should not exceed 100%)

___%

____%

____%

____%

____%

____%

____%

__%

___%

Approximate percentage each procurement mode contributed to all private sector building construction projects and turnover in NSW between July 2007 and June 2008? (percentages in this row should not exceed 100%)

___%

____%

____%

____%

____%

____%

____%

__%

___%


6. Please insert answers in the following table concerning your firm’s services or procurement modes delivered in New South Wales amidst/post the Global Financial Crisis (July 2009-June 2010). If your firm did not offer one of the procurement modes in this period, then please insert “Not Applicable” in the column under the procurement mode concerned. Please note, in terms of procurement mode, the following abbreviations are used once again:

• C = Construct only (lump sum or fixed price contract)








• PF = Any procurement mode that includes private finance e.g. a Public Private Partnership

Your firm’s procurement modes delivered in NSW amidst/post the Global Financial Crisis (July 2009-June 2010):

C D&C DC&M DCO&M CM MC ECI AC PF

Approximate percentage each procurement mode contributed to all public sector building construction projects and turnover in NSW between July 2009 and June 2010? (percentages in this row should not exceed 100%)

____%

____%

____%

____%

____%

____%

____%

____%

____%

Approximate percentage each procurement mode contributed to all private sector building construction projects and turnover in NSW between July 2009 and June 2010? (percentages in this row should not exceed 100%)

____%

____%

____%

____%

____%

____%

____%

____%

____%


7. What was the approximate percentage of all building contracts awarded to your firm in New South Wales and which tended to rely on more price competition as the basis of selection in the following sectors and in the following periods:

a. Public sector i. Just prior to the Global Financial Crisis (July 2007-June 2008). Please insert

percentage:_____% ii. Amidst/post the Global Financial Crisis (July 2009-June 2010): Please insert

percentage:_____% b. Private sector

i. Just prior to the Global Financial Crisis (July 2007-June 2008). Please insert percentage:_____%

ii. Amidst/post the Global Financial Crisis (July 2009-June 2010): Please insert percentage:_____%

8. What was the approximate percentage of all building contracts awarded to your firm in New South Wales in terms of standard contacts (e.g. Australian Standard) and what was the nature of non-standard contracts (i.e. contract terms authored by your clients and/or by your firm) in the following sectors and in the following periods:

a. Public sector i. Just prior to the Global Financial Crisis (July 2007-June 2008). Please insert

percentage 1. Standard contract terms and conditions.

a. Please insert percentage:_____% 2. Were the terms and conditions in non-standard contracts tending to be

authored by clients; your firm; or approximate same proportion? Please circle: Client / Your Firm / Same proportion


1. Standard contract terms and conditions. a. Please insert percentage:_____%

2. Were the terms and conditions in non-standard contracts tending to be authored by clients; your firm; or approximate same proportion? Please circle: Client / Your Firm / Same proportion

b. Private sector i. Just prior to the Global Financial Crisis (July 2007-June 2008). Please insert

percentage:_____% 1. Standard contract terms and conditions.

a. Please insert percentage:_____% 2. Were the terms and conditions in non-standard contracts tending to be

authored by clients; your firm; or approximate same proportion? Please circle: Client / Your Firm / Same proportion


1. Standard contract terms and conditions. a. Please insert percentage:_____%

2. Were the terms and conditions in non-standard contracts tending to be authored by clients; your firm; or approximate same proportion? Please circle: Client / Your Firm / Same proportion


9. Was your portfolio of your firm’s private sector building construction contracts concentrated in New South Wales; that is, did a small number (up to say six clients) account for the majority of the combined value of all your private sector building construction projects in New South Wales?

a. Please insert pre- Global Financial Crisis (July 2007-June 2008): Yes or No:________ b. Please insert amidst/post- Global Financial Crisis (July 2009-June 2010): Yes or No___

10. In what ways does your firm consider it differentiates itself - at the current time, from its main rivals in terms of your firm’s building construction services in New South Wales? Please give brief details:_____________________

11. If applicable, has the way in which your firm currently differentiates itself - as noted in the

previous Question 10, changed from the way(s) your firm may have differentiated itself from its main rivals pre-Global Financial Crisis? If so, then please give brief details:______________________________

12. What were your firm’s approximate earnings before interest and taxes (EBIT) as a percentage of

turnover / net sales in the last five years? If possible, please insert these details in respect of your firm’s total building construction activities (all public and all private sector) in New South Wales only:

a. 2005 / 6 Please circle: Less than 2% / between 2% and 5% / More than 5% b. 2006 / 7 Please circle: Less than 2% / between 2% and 5% / More than 5% c. 2007 / 8 Please circle: Less than 2% / between 2% and 5% / More than 5% d. 2008 / 9 Please circle: Less than 2% / between 2% and 5% / More than 5% e. 2009 / 10 Please circle: Less than 2% / between 2% and 5% / More than 5%

f. Please confirm what part of your firm’s activities the EBIT figures above

represent:_______


Section D: Effects of tendering and other procurement related factors on bidding and construction capacity This section focuses on your feedback on tendering policy and practice used by New South Wales government’s department(s)/agencies responsible for delivering major public sector building construction projects including hospital/health care projects (over $50million capital value). This section also seeks your feedback on other procurement related factors that may affect your firm’s near term construction capacity. Just a reminder, please don’t hesitate to provide any further comments concerning this section or any other part of this questionnaire in the next Section E. 13. In the period approximately from January 2010, how much do you feel the following aspects of

tendering policy and practice and other procurement related factors either encourages or discourages your firm’s willingness to bid for major public sector building construction projects including hospital/health care (over $50million capital value) in New South Wales:

a. The numbers of bidders for traditional price competition contracts: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

b. The numbers of bidders for more collaborative contracts: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

c. Tendering costs for traditional price competition contracts: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

d. Tendering costs for more collaborative contracts: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

e. Tendering costs for privately financed projects (e.g. PPPs): Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

f. The level of detail revealed from the Public Sector Comparator in PPPs: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

g. Tender evaluation criteria and weightings in more traditional price competition contracts: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages h. Tender evaluation criteria and weightings in more collaborative contracts: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages i. The time allowed for tendering: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages j. The sequencing and coordination of tender submission dates for different projects:

Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

k. The sequencing and coordination of projects (including extent of forward program of projects): Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages l. The proportion of traditional price competition contracts versus more collaborative

contracts: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

m. Sizing (value) and bundling of projects: Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages

n. Up to expressions of interest, level of confidence concerning government funding approval for proposed projects: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages o. Use and administration of performance guarantees and bonds: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages p. Allocation of risk in all contracts: Please circle:

Strongly encourages 1 2 3 4 5 Strongly discourages q. The time taken to award a project:

Strongly encourages 1 2 3 4 5 Strongly discourages r. Level of face-to-face contact with the government client during the bidding process:

Please circle: Strongly encourages 1 2 3 4 5 Strongly discourages


14. Again, in the period approximately from January 2010, what do you feel New South Wales state government department(s)/agency(ies) responsible for delivering major public sector building construction projects including hospital/health care (over $50million capital value) have done particularly well and better than business as usual in terms of procurement and tendering: Please give brief details:

a. Government department(s)/agency(ies) responsible for hospital/health care i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

b. Government department(s)/agency(ies) responsible for PPPs i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

c. Government department(s)/agency(ies) responsible for other building construction projects

i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

15. Once again, in the period approximately from January 2010, where do you feel there is some

scope for improvement amongst New South Wales government department(s)/agency(ies) responsible for delivering major public sector building construction projects including hospital/health care (over $50million capital value) in terms of procurement and tendering: Please give brief details:

a. Government department(s)/agency(ies) responsible for hospital/health care i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

b. Government department(s)/agency(ies) responsible for PPPs i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

c. Government department(s)/agency(ies) responsible for other building construction projects

i. ________________________________________________________________ ii. ________________________________________________________________ iii. ________________________________________________________________

16. In the period to June 2012, what would be your preferences in terms of the approach taken by

New South Wales government department(s)/agency(ies) responsible for delivering major public sector building construction projects including hospital/health care (over $50million capital value) in terms of procurement and tendering and which may increase your firm’s willingness to bid for these projects – and specifically:

a. Your preferred size / value range of project: Please insert $ (to nearest $10million):____ b. Your preferred procurement mode(s):________________________________________ c. Contract payment / risk terms (e.g. more fixed price with allocated risks or more shared

risk/rewards e.g. pain share/gain share regimes)_________________________________ d. Other preferences:________________________________________________________


17. Again in the period to June 2012, how much do you feel your firm’s building construction capacity / turnover in New South Wales might be constrained or adversely affected by the following factors:

a. Skilled trade resources: Please circle: Negligible constraint/affect 1 2 3 4 5 Severe constraint/affect

b. Professional resources: Please circle: Negligible constraint/affect 1 2 3 4 5 Severe constraint/affect

c. Financial resources if contemplating a PPP bid: Please circle: Negligible constraint/affect 1 2 3 4 5 Severe constraint/affect

d. Pricing rivalry amongst your firm’s current competitors: Please circle: Negligible constraint/affect 1 2 3 4 5 Severe constraint/affect

e. New overseas entrants into the market as additional rivals to your firm: Please circle: Negligible constraint/affect 1 2 3 4 5 Severe constraint/affect

18. The following questions concern Integrated Project Delivery (IPD) and based on design development using digital modelling software that incorporates intelligent objects (for example, Building Information Modelling – BIM) and concerns all your firm’s construction operations in New South Wales:

a. How much do you feel your firm is currently capable of using and benefiting from IPD and BIM? Please select:

Minimal capability 1 2 3 4 5 substantial capability b. In the period from July 2011 to June 2012, how much does your entire firm envisage

investing in IPD and BIM capabilities? Please select: Minimal investment 1 2 3 4 5 substantial investment

c. In what ways do you think procurement and tendering could maximise the benefits of IPD and BIM to positively impact all stakeholders and the construction industry: Please give brief details:_________________________________________________________ _______________________________________________________________________


Section E: Any other comments or feedback are very much welcomed __________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

__________________________________________________________________________________

Appendix E – Notification letter for survey of civil and building contractors 327

Appendix E – Notification letter for survey of civil and building contractors

4th May 2011

Mr ABC General Manager XYZ Pty Ltd Dear Mr ABC, Australian Research Council Grant: “Reforming the procurement of construction and financing of Australian infrastructure: Advancing capacity, competition and investment”

We are pleased to notify you of the above Australian Research Council grant awarded to Queensland University of Technology. As part of this grant, Queensland University of Technology entered into a collaborative research agreement with:

• NSW Treasury, Victorian Department of Treasury & Finance (Partnerships Victoria), Queensland Treasury, WA Department of Treasury and Finance, SA Department of Treasury and Finance and Queensland Department of Infrastructure and Planning; and

• The Construction Industry Institute Australia, Infrastructure Association Queensland, Infrastructure Partnerships Australia, Aurecon and Coffey Commercial Advisory.

The overall aim of the grant is to develop a new and integrated procurement knowledge base to help address current construction and finance constraints in delivering Australia’s infrastructure. The first project in the grant concerns state government policy and practice in the procurement of major infrastructure and aims to develop a new procurement decision making model for consideration by governments across Australia. As part of the data to be collected in this first project, a nationwide survey of civil and building contractors will be conducted.

We will shortly be writing to you and all civil and building contractors capable of delivering a $50million project in the five Australian jurisdictions represented by the government departments in the research agreement, to invite you and all these other contractors to participate in the nationwide survey.

Our goal is to receive a complete/substantially complete response from 95 percent or more of all civil and building contractors invited to participate.

The survey will provide you a significant opportunity to give feedback to government on procurement policy and practice. All responses will be treated confidentially and none of the funding bodies will have access to any of the data obtained in the survey questionnaire. Only aggregated results and other reports with non-identifiable data will be made available to funding bodies and in any publications in which Queensland University of Technology develop independent conclusions. If you indicate in the questionnaire that you would like to be informed of the results of the survey, then we will be very pleased to send you a summary of the analysis of aggregated and non-identifiable data in both the civil and building contractors’ versions of the questionnaire. Upon receipt of this information you may consider developing your own conclusions - beyond those reported by Queensland University of Technology and towards the widest possible debate of the matters arising from the survey. Thank you for your time in noting the above and we look forward to contacting you again in the near future to invite you to make a very important contribution to developing state government major infrastructure procurement policy and practice. Yours sincerely

Appendix F – Cover letter for survey of civil and building contractors 329

Appendix F – Cover letter for survey of civil and building contractors

14 December 2011 Mr ABC General Manager XYZ Pty Ltd Dear Mr ABC, Australian Research Council Grant: “Reforming the procurement of construction and financing of Australian infrastructure: Advancing capacity, competition and investment” As mentioned in the attached notification letter, we invite you to participate in a nationwide survey of civil and building contractors capable of delivering a $50million project and concerning state government policy and practice in the procurement of major infrastructure. More specifically, the survey contributes to the first project in the above Australian Research Council grant that aims to allow an improved understanding of the effects of procurement on construction capacity and competition and to develop a new procurement decision making model for consideration by governments across Australia. Further details of the research grant and this first project can be found from page 2 in the questionnaire and at https://wiki.qut.edu.au/display/arcmip. Please see enclosed a reference only hard copy of the version of the questionnaire for civil contractors operating in Victoria and the version of the questionnaire for building contractors operating in Victoria. Please also see attached details concerning accessing the questionnaire on-line. We invite you to submit on-line both the civil contractors’ version and the building contractors’ version of the questionnaire for Victoria and using the attached access details. Having developed and tested the questionnaire with the help of civil and building contractors in more than one state, we envisage that a general manager covering your firm’s building operations and a general manager covering your firm’s civil operations would be ideally placed to answer the majority of the questions based on their own knowledge of your firm’s operations. With a minority of the questions, these managers may need to consult with colleagues in other parts of your firm in order to provide answers. With regard to the substantial part of the questionnaire answered directly by a general manager, testing the questionnaire has shown that it can be completed within 30 minutes. My contact details are given below and contact details for the other researcher from Queensland University of Technology are given on the first page of the questionnaire and we will be very pleased to assist you in whatever way we can.

330 Appendix F – Cover letter for survey of civil and building contractors

Continued…

Our goal is to receive an on-line completed/substantially completed questionnaire from as many civil and building contractors in Australia as possible and which are capable of delivering a $50million project.

The questionnaire provides you with a significant opportunity to give feedback to government on procurement policy and practice and includes a free text section at the end of the questionnaire for any related comments you would like to make. All responses will be treated confidentially. Only the research team at Queensland University of Technology can access this data and none of the funding bodies will have access to any of the data obtained in this questionnaire. Only aggregated results and other reports with non-identifiable data will be made available to funding bodies and in any publications in which Queensland University of Technology develop independent conclusions. If you tick the box at Question 2, then we will be very pleased to send you a summary of an analysis of aggregated and non-identifiable data in both the civil and building contractors’ versions of the questionnaire. Upon receipt of this information, you may consider developing your own conclusions - beyond those that Queensland University of Technology report and towards the widest possible debate of the matters arising from the survey. It would be greatly appreciated if you or your delegate/general managers covering your firm’s civil and building operations complete and submit on-line, using the attached access details, the questionnaire for civil contractors in Victoria and the questionnaire for building contractors in Victoria and as soon as practicable within 20 working days of receipt of this invitation letter. In anticipation of your help and on behalf of the entire research team, all mainland state government partners and other partner organisations, thank you very much indeed for your firm’s time in completing these questionnaires and for your very important contribution. Yours sincerely

Appendix G – Coding of survey of civil and building contractors 331

Appendix G – Coding of survey of civil and building contractors

Questions Coding Civil/Building Coded 0=civil, 1=building State Coded 1=NSW, 2=QLD, 3=Vic, 4=WA, 5=SA q1 Coded 1=no; 2=yes q2a1 Enter value in $M

q2a11 Coded 0 = less than $50

1 = $50- not including 100M 2.5 = $100- not including 250M 5 = $250- not including 500M 10 = $500- not including 1B 100 = including and greater than $1B q2a2 Enter value in $M q2a21 Same as above q2a11 q2a3 Coded 1 = not applicable q2a4 Enter value in $M q2a41 Same as above q2a11 q2a5 Enter value in $M q2a51 Same as above q2a11 q2a6 Coded 1 = not applicable q2b1, q2c1, q2d1, q2e1, q2f1, q2g1, 2h1,q2i1 Enter value in $M q2b11, q2c11, q2d11, q2e11, q2f11, q2g11, 2h11,q2i11 Same as above q2a11 q2b2, q2c2, q2d2, q2e2, q2f2, q2g2, q2h2, q2i2 Enter value in $M q2b21, q2c21, q2d21, q2e21, q2f21, q2g21, q2h21, q2i21 Same as above q2a11 q2b3, q2c3,q2d3, q2e3, q2f3, q2g3, q2h3, q2i3 Coded 1 = not applicable q2b4, q2c4, q2d4, q2e4, q2f4, q2g4, q2h4, q2i4 Enter value in $M q2b41, q2c41, q2d41, q2e41, q2f41, q2g41, q2h41, q2i41 Same as above q2a11 q2b5, q2c5, q2d5, q2e5, q2f5, q2g5, q2h5, q2i5 Enter value in $M q2b51, q2c51, q2d51, q2e51, q2f51, q2g51, q2h51, q2i51 Same as above q2a11 q2b6, q2c6, q2d6, q2e6, q2f6, q2g6, q2h6, q2i6 Coded 1 = not applicable q3a & q3b Enter answer q4a1-q4a6 Enter value in $M q4b1-q4b6 Enter % q4c1-q4c6 Enter % q4d1-q4d6 Enter no

332 Appendix G – Coding of survey of civil and building contractors

Questions Coding q4e1-q4e6 Enter %

q4f1-q4f6 Coded 1=significant spare, 2=spare capacity, 3=near full, 4=full/overstretched capacity

q4g1-q4g6 Coded 1=no; 2=yes q4h1-q4h6 Coded 1=no; 2=yes q5a1-q5a9 Enter % q5b1-q5b9 Enter % q6a1-q6a9 Enter % q6b1-q6b9 Enter % q7a1, q7a2, q7b1, q7b2 Enter % q8a1, q8a3 Enter % q8a2, q8a4 Coded 1=Client, 2=Your firm, 3=same proportion q8b1, q8b3 Enter % q8b2, q8b4 Coded 1=Client, 2=Your firm, 3=same proportion q9a, q9b Coded 1=no; 2=yes q10 Enter description q11 Enter description q12a1, q12b1, q12c1, q12d1, q12e1 Coded 1=less than 2%, 2=bet 2%-5%, 3=more than 5% q12a2, q12b2, q12c2, q12d2, q12e2 Enter description

q13a- q13r Coded 1=strongly encourage, 2=encourage, 3=neutral, 4=discourage, 5=strongly discourage

q14a-q14c Enter description q15a-q15c Enter description

q16a Coded 0 = less than $50

1 = $50- not including 100M 2.5 = $100- not including 250M 5 = $250- not including 500M 10 = $500- not including 1B 100 = including and greater than $1B

q16b Coded 1=C, 2=DC, 3=DCM, 4=DCOM, 5=CM, 6=MC, 7=ECI, 8=AC, 9=PPP

q16c Coded 0=happy with either fixed price with allocated risk or shared risks/rewards, so long as risks is properly allocated. 1=Fixed periodic payments (PPP) 2= Lump sum (fixed price) 3 = GMP/GCS/Target out-turn cost, pain/gain share 4= Remeasurement/ Schedule of rates and prices/ Cost plus including pain/gain share

q16d Enter description q17a -q17e Coded 1=negligible, 2=minor, 3=moderate, 4=major, 5=severe

q18a Coded 1=minimal capability, 2= slight capability, 3=moderate capability, 4=sufficient capability, 5=substantial capability

q18b Coded 1=minimal investment, 2= slight investment, 3=moderate investment, 4=sufficient investment, 5=substantial investment

q18c Enter description q19 Enter description

Appendix H – Case study questionnaire instrument (initial version) for semi-structured interviews 333

Appendix H – Case study questionnaire instrument (initial version) for semi-structured interviews

Activity levels

Internalised activities

1 = Technical / or organisationally better market

2 = Technical better than market

3 � = Organisationally better than market

Externalised activities

5 � = Organisationally better than government

6 = Technical better than government

7 = Technical / or organisationally better government

External/ Internal activities

4 � = Potential for costly variations (time or money) in/arising from the

activity

334 Appendix H – Case study questionnaire instrument (initial version) for semi-structured interviews

Investment

1. Site • Plant and equipment that are site specific and costly to move to a far

location

2. Physical asset • Investments of tangible assets in specific technology for a single

purpose not usable for other activities, e.g. equipment, inventory, machinery, software, excluding leases/ hire/ agreements

3. Human resource / knowledge

• Highly specialized knowledge and skills

4. Dedicated asset specificity • Discrete investment to increase capacity, e.g. increase a plant, that

cannot readily be put to work for other purposes

5. Temporal asset specificity

• Costs of switching consultant/ contractor/ operator in the midst of carrying out works

6. Temporal asset specificity

• investments in building and protecting reputation, for example, regarding a special group of products or a particular market segment.


Activity Internal/ External

Level Uncertainty (Potentially greatest source of variation in time/ delay)

Investment (Any specific investments)

No. Pattern Design 1. 2. 3. 4. 5. 6. New installation/construction works 7. 8. 9. 10. 11. 12. Implementation of operations 13. 14. 15. 16. 17. 18. 19. 20. Implementation of maintenance 21. 22. 23. 24. Any other activities

336 Appendix H – Case study questionnaire instrument (initial version) for semi-structured interviews

Please note all answers mindful of circumstances at very approximate date that actual procurement decision made:_______year Key activity:

• Activity Number ________________

• Indication of distinct knowledge/skill set: ________________

• Indication of non-trivial: ________________

• Actual internal or external: ________________

• Estimated pattern/level: ________________

1. Value : a. How much is the activity of case study project peripheral to state government front-

line services? i. Please circle : extremely peripheral 1 2 3 4 5 6 7 extremely core

b. If state government were to perform all this activity of all new projects in-house/within an agency in the state government, then how much would this have tended to increase costs and put more pressure on the relevant budget?

i. Please circle: very greatly increase cost 1 2 3 4 5 6 7 very greatly reduce costs

2. Rarity :

1. How much is the knowledge and skills required to carry out aspects of the activity of the case study possessed by all the relevant firms capable of projects over $50million in the locality of the case study project?

i. Please circle: possessed by all firms 1 2 3 4 5 6 7 possessed by few firms

2. How much is there a sufficient supply of firms capable of delivering the activity to the case study?

i. Please circle: plentiful supply of firms 1 2 3 4 5 6 7 scarce supply of firms

3. Costly-to-imitate : a. How difficult is it for a manual to be written (that reflects knowledge, policies and

procedures) and followed in order to required to carry out the activity in terms of the aspects of the case study?

i. Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult b. How difficult is it for all related firms capable of projects over $50million in the

locality of the case study to develop same knowledge and skills required to carry out the activity in terms of the aspects of the case study?

i. Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult

4. Asset specificity :

1. How much time and investment would be needed to make any adaptations/undertake any specific additional education/research/training to knowledge and skills already possessed by the relevant firm with capability required to carry out the activity in terms of the aspects of the case study?

i. Please circle: minimal adaption time and investment 1 2 3 4 5 6 7 substantial adaption time and investment

2. At any stage (mid stage or end of a stage) in carrying out the activity of the case study project, how straightforward (in terms of effect on the project’s timeline) would it have been to have replaced the firm performing the activities?

i. Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult


5. Uncertainty : a. How much is the nature (range of tasks) of the activity required in the case study

straightforward to predict by someone with expertise in the activity? i. Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult

b. How much is the performance/outputs and standards of the activity in the case study road able to be specified clearly prior to appointment of the service of the related firm?

i. Please circle: precisely 1 2 3 4 5 6 7 vaguely c. How much is the nature (range of tasks) and extent (scale; resources and time) of the

activity subject to changes caused by unknown factors relating changes in physical conditions and materials and/or state government changes (e.g. scope) during the period allowed for design in the case study project?

i. Please circle: extremely small 1 2 3 4 5 6 7 extremely high

6. Frequency: a. How much demand was there for a workforce (one or more individuals) required to

perform the activity by the state government (or case study project only) and to keep this workforce fully employed beyond any one project and on an ongoing/indefinite basis (or during the case study project only?)

i. Please circle: Very much insufficient 1 2 3 4 5 6 7 substantially more than sufficient

Appendix I – Case study questionnaire instrument for health or road projects 339

Appendix I – Case study questionnaire instrument for health or road projects

Science and Engineering Faculty Civil Engineering and Built Environment School

Australian Research Council Grant

(further details of grant available at: https://wiki.qut.edu.au/display/arcmip)

CASE STUDIES OF MAJOR HEALTH OR ROAD PROJECTS

PARTICIPANT INFORMATION SHEET FOR A QUT RESEARCH PROJECT

(Full Title of Grant): Reforming the procurement of construction and financing of Australian infrastructure:

Advancing capacity, competition and investment (Full Title of Project in Grant):

Project 1: The effect of procurement on the extent of construction competition and capacity and new first-order procurement decision-making model

QUT Ethics Approval Number 1000000345

RESEARCH TEAM

Principal Researchers: Ms Pauline and Dr Adrian Bridge Associate Researchers: Professor Martin Skitmore and Mr Jason Gray and Mr Marcus Jefferies

DESCRIPTION

This research is being undertaken as part of Australian Research Council (ARC) Linkage Grant funded by:

ARC; Partnerships Victoria, Department of Treasury and Finance, Victoria; New South Wales Treasury; Queensland Treasury; Western Australia Department of Treasury and Finance; South Australia Department of Treasury and Finance; Queensland Department of Employment, Economic Development and Innovation; Construction Industry Institute Australia (including Queensland Department of Transport and Main Roads and Queensland Department of Public Works); Infrastructure Association Queensland; Infrastructure Partnerships Australia; Coffey Commercial Advisory; and Aurecon

None of the funding bodies will have access to the data obtained in this questionnaire.

The overall purpose of the research is to develop a new and integrated procurement knowledge base to help address construction and finance constraints in Australia. The project outcomes will be considered by government as the basis of possible procurement reform, to increase construction capacity, competition and private sector investment (particularly by superannuation funds) to public sector infrastructure. The purpose of this component of the research (Project 1 in the overall grant) is to develop an improved understanding of the effect of procurement on construction competition and capacity in road and health public sector major infrastructure and in doing so develop and test a new first-order procurement decision-making model.

The research team requests your assistance as you are a representative of a public agency with knowledge of the actual procurement of one of the four cases studies being used to develop and test the new procurement decision-making model.

340 Appendix I – Case study questionnaire instrument for health or road projects

PARTICIPATION

Your participation in this project is voluntary. If you do agree to participate, you can withdraw from participation at any time during the project without comment or penalty. Your decision to participate, or not participate, will in no way impact upon your current or future relationship with QUT.

Your requested participation in the research will involve discussion and identification of key design; construction; operations; and maintenance activities and completion of a questionnaire on each key activity. Your answers to the questions form one of the steps in the procedures in the new first-order procurement decision-making model that identifies a procurement strategy and which can be compared to the actual approach taken procuring the case study. The extent to which the outcomes from the model match actual procurement is then evaluated using competition (expressions of interest) as an external criterion and which is consistent with value-for-money in whole-life terms. The process of meeting to identify activities and seeking your answer to the questions on each activity may resemble more of a workshop and which should take approximately 3 hours to conduct.

In addition, with your permission, we would like to access/collect any archival documentation you may wish to offer that relates to or supports the information you provide in the meeting/ questionnaire.

EXPECTED BENEFITS

Government and industry will benefit from the development and availability of new decision making models and financial models that are designed to be considered by government to advance Australia's position in the delivery of public sector infrastructure in terms of increasing construction capacity and investment to infrastructure and improving value for money for tax payers. The ARC and general academic community will benefit in terms of advancing knowledge and in training researchers.

RISKS

There are no out-of-the ordinary individual or business risks associated with your participation in this project.

CONFIDENTIALITY

All comments, responses and any case study documents provided will be treated confidentially and stored securely in accordance with QUT’s Management of Research Data Policy. None of the funding bodies will have access to any of the data obtained in this questionnaire. Only aggregated results and other reports with non-identifiable data will be made available to funding bodies and in any publications.

CONSENT TO PARTICIPATE The return of the completed questionnaire is accepted as an indication of your consent to participate in this project. That is, you are indicating that you:

• Have read and understood this information document regarding this project.

• Have had any questions answered to your satisfaction.

• Understand that if you have any additional questions you can contact the research team.

• Understand that you are free to withdraw at any time, without comment or penalty.

• Understand that you can contact the Research Ethics Unit on 3138 5123 or email [email protected] if you have concerns about the ethical conduct of the project.

• Agree to participate in the project as outlined above.

QUESTIONS / FURTHER INFORMATION ABOUT THE PROJECT Please contact the research team members if you have any questions or if you require further information. Ms Pauline Teo Dr Adrian Bridge 0430 609 191 [email protected] 3138 1543 [email protected]

CONCERNS / COMPLAINTS REGARDING THE CONDUCT OF THE PROJECT QUT is committed to research integrity and the ethical conduct of research projects. If you have any concerns or complaints about the ethical conduct of the project please contact the QUT Research Ethics Unit on 3138 5123 or email [email protected]. The Research Ethics Unit is not connected with the research project and can facilitate a resolution to your concern in an impartial manner.


Final Version New first-order procurement decision-making model

Introduction Definition of procurement and model’s Value for Money philosophy The UK National Audit Office (2004) defines procurement as, “the whole-life process of the acquisition of goods, services and works..., beginning when a potential requirement is identified and ending with the conclusion of service contract or ultimate disposal of an asset”. The effective and efficient procurement of infrastructure is often translated as achieving value for money (VfM). HM Treasury (2008) defines VfM as, “securing the best mix of quality and effectiveness for the least outlay over the period of use of the goods or services bought. It is not about minimising upfront prices...”. The best mix can be interpreted as the best ratio between benefit (utility/return) and cost, or VfM = f (cost/benefit). More specifically, benefits comprise largely font-line/user utility given a whole-of-life concern and costs include both production costs (design; construction; operations; and maintenance) and transaction costs. In turn, transaction costs comprise both internal management costs to the buyer (in this case, state government) and more tangible external transaction costs. Research opportunity A severe lack of empirical and comprehensive comparisons of PPPs and non-PPPs facilities throughout whole-life (including operations) emphasizes the importance, in particular reliability and validity, of early decision-making guidelines and models used to inform the efficacy of either going down a PPP or non-PPP path in pursuance of value for money (VfM) There exists scope to improve the practice of deploying early indicators or filters and/or indirect (perceptual/semi-quantitative) approaches to determining whether a project takes a PPP path or a non-PPP path (as depicted in Figure 1). For example, the decision-making process envisaged in IA’s National Public Private Partnership Guidelines (2008) and in particular Step 2 (shortlist delivery model) and Step 4 (delivery model options analysis) that is a variation on Multi-Attribute Utility Approach (MAUA)

Figure 1. Research opportunity

Schematic of procedures in new first-order procurement decision-making model • The objective of this research is to develop and test a new first-order procurement decision-

making model that will identify a procurement strategy including whether a new project is better procured as a PPP or non-PPP, and in terms of which path is more likely to deliver superior VfM; In doing so, this model will

o Provide more confidence that procurement is efficacious - in the period in which all research catches-up to the challenge of fully assessing all production and transaction costs and all benefits including the relationship of the facility with the front-line service across PPP versus non-PPP and across the whole-life (including operations) of the facility

o Address the weaknesses of current practice in terms of the new model representing a scientific and wholly economic approach to decision-making

o Generate more sophistication in procurement decision-making and advance VfM in delivering major public sector projects

o Save time and costs to both government and industry by much more reliability identifying the most suitable projects to be procured using a PPP approach and ensuring

Procurement Mode

Risk Analysis Project

Attributes

New institutional economic

theories and resource-

based theory

Procurement Mode

Government Resources

Private Sector

Resources

Project Attributes


VfM Relative

Terms


VfM

Nominal


the extra-over work involved in a PPP, including PPP procedure beyond performance specification and reference design up to financial close, is justified.

o Provide a transparent and public interest document that can be fully disclosed and supplement the PSC (in projects in which parts of PSC not published due to commercial-in-confidence concerns) or entirely take the place of the PSC in terms of what is published as justification for going down a PPP path (a full PSC is still envisaged be for early budgeting/approval purposes)

• In contrast to current practice, the level of analysis in the new decision-making model commences with a focus on a first-order level of analysis (Stage 1) in terms of the analysis of key design, construction, operations and maintenance activities. These activities are measured in accordance with two Nobel prize winning theories from the New Institutional Economics (transaction costs theory and transaction cost economics that concern the key aspects of procurement) and the dominant theory from the field of Strategic Management relating to procurement (resource-based theory) and configured within a set of procedures in the new model that are falsifiable/testable (scientific); This set of procedures in summarized in the schematic below in Figure 2

Figure 2. Schematic of model (Teo et al., 2010; 2011; 2012; 2013) • In summary, the model informs the procurement strategy by delineating three key procurement

dimensions comprising size of project, extent bundling and the nature of the exchange relationship (including method of remuneration) between government and private sector firm at the head of each bundle/contract within the project.

• Experiential evidence supports the importance of these three key procurement dimensions. That is, in the UK the House of Lords Select Committee on Economic Affairs (2010) concludes project size and flexibility as two key drivers in determining the suitability of PPPs to deliver VfM:

o Size is the first key procurement dimension informed by the new model in this research and the second dimension studied, that is level of bundling, also affects project size

o The exchange relationship, as the third key procurement dimension informed by the new model, directly speaks to the ability of the buyer or government to exercise flexibility post contract

Structure of this report • This report proceeds to provide some details concerning the procedures in Figure 2 and which are

then presented in terms of being deployed in Health Case Study #1. Following this modeling of Health Case Study #1, a summary of the approach taken to testing the validity and reliability of the new first-order procurement decision-making model is given and which shows how the results from Health Case Study #1 contributes to the testing the new procurement model.

Procedures in the new first-order procurement decision-making model Stage 1 Task A Activity analysis (first-order analysis)

Stage 1 (1st-order)

Stage 2 (2nd-order)

Initial Schematic Design

Task A Activity Analysis

Task B Make-or-Buy Analysis

Procurement Strategy

Task C Market Analysis

Exchange Relationship Analysis

Bundling Analysis

Stage 3 (3rd-order)


In this task, the model prompts/guides the user to identify first-order or key activities (distinct technology with distinct knowledge/skill sets and non-trivial i.e. significant cost relative to cost of project): • Key activities have, as their source of adding value the actual technology associated with design;

construction – new installation; operations and maintenance – implementation)

• Not a key activity in this approach e.g. o A milestone in progress/on the program such as schematic design (covers many

consultant disciplines) o A building element (covers many trades) o Organisation and/or management (including planning/programming and supply chain

management including procurement) as the dominant source of adding value) – this is considered a second-order activity

• Delimited (limitations of scope determined by researcher as opposed to limitations out of researchers control)

o From and including key activities arising from schematic (hence excludes upstream activities like planning; surveying site; geotechnical survey and other activities that informed schematic)

o To and including maintenance and operation of built asset but excluding core/front-line activities (e.g. excludes downstream activities like clinical services and excludes activities around design; supply; installation/positioning of loose equipment and plug-in equipment, as well as excluding ancillary services that do not interact with the building/asset e.g. Linen and porterage services in hospitals)

o From the period of time starting with the conclusion of the schematic to the point the asset is deemed obsolete and requires rehabilitation/major and extensive refurbishment (hence rehabilitation or refurbishment is excluded)

• The model starts with a focus on key production activities of design, construction and operations and maintenance and is, therefore, a first-order analysis in contrast to current practice that starts at an organisational level e.g. a main contractor or a managing contractor that manages across key activities and which is second-order analysis (more on this later). This corresponds to a critique of ineffective risk analysis associated with current practice (Teo et al., 2010; 2011; 2012; 2013). That is, current practice is:

o Too focused, in terms of attempting to identify individual risks (guessing with information missing at schematic stage and then second-guessing when, for example, consultants attempt to assess size/value of risk of activities that are better understood by other parties, for example, contractors)

o At same time, this approach to risk analysis becomes overshadowed by a main concern coming other way that is on matching key opening day outcome(s) to a procurement mode - that is expressed at the oganisational or second-order level and undermines a full/effective risk analysis, including activities not being allocated to the party with best resources to manage risks

o In contrast, first-order analysis answers Chang’s (2013) call for a unit of analysis that is mediated by governance and, in doing so, leads into a make-or-buy analysis around the key activities i.e. this is the conventional approach in Transaction Cost Economics that sees transaction costs occurring across technological distinct boundaries and so the unit of analysis is mediated or affected by governance and lends itself to make-or-buy analysis and which is effectively an indirect approach to risk analysis (does not indentify individual risks, rather adopts a more appropriate focus, mindful of the information gap inherent in a schematic design, on who is best placed to deal with risks potentially arising out of an activity); More specifically the model deploys three theories. One of these theories is the dominant theory concerning procurement in the field of Strategic Management and two of these theories are Nobel prize winning. That is:

� Resource-Based Theory (RBT) that addresses technical/production risks – which party is better equipped to deliver the key activity and therefore better


able to manage potential risks arising out of the key activity; This also includes which party is better are able to secure sufficient demand/workload concerning the activity in order to generate economies of scale and/or learning economies in order to more efficiently deliver the activity (dominant Strategic Management theory);

� Coase’s transaction costs thesis that focuses on internal transaction costs and reveals which party is better placed to manage an activity and deliver the activity more efficiently (Nobel prize for economics in 1991); and

� Transaction Cost Economics (TCE) that focuses on external transaction costs and surfaces potential exchange risk arising out of hold-up as a function of a specific investment by one party (in this case, state government) in the key activity that creates a dependency of state government on the supplying party and makes the non-investing supplying party the less vulnerable party and therefore the more powerful party and able to behave potentially in a negative opportunistic way on the occurrence of a change/variation. Hence, the key variables in this theory are asset specificity and uncertainty (Nobel prize for economics in 2009).

Stage 1 Task B Make-or-buy analysis on each activity (first-order analysis) In this task, the model can be used in two ways, either backwards in terms of explaining/reviewing a previous procurement decision in a completed project or forwards in terms of guiding the identification of a procurement strategy for a new project in terms of a procurement strategy that is more likely to yield best ratio of benefits/front-line user utility and whole-life-costs for a new project. In this case of a major health project, the model is being used in review mode. In this mode, this task in the model prompts/guides the user to: • Note whether each of the key activities identified in the previous task were/are internalised or

externalised;

• Answer the standard questions on each activity and which are slightly different for internalized and externalized activity in order to generate a pattern of symbols for the 6 variables/economic dimensions representing the three theories being used; and

• Match this empirical pattern or actual pattern to the closest theoretical pattern in the following table in order to establish an internalisation or externalization level for the activity:

Table 1: Integrative framework of vertical integration (based on Bridge and Tisdell 2004) Level Logic Value

(Capacity) (RBT)

Rarity

(RBT)

Costly to Imitate

(RBT)

Asset Specificity

(TCE)

Uncertainty

(TCE)

Frequency

(TCE)

Mode of Governance

1 Capability (RBT)

+++ Score: 7






Internal

2 Competence (RBT)

++ Score: 6


0 Score: 1 to 4




Internal

3 Transaction Costs/

Competence (Coase)

+ Score: 5

0 Score: 1 to 4

0 Score: 1 to 4


0 Score: 1 to 4


Internal

4a TCE/Hold-up (Williamson)

-/+ Score: 4

0 Score: 1 to 4

0 Score: 1 to 4




Internal

4b TCE/Hold-up (Williamson)

-/+ Score: 4

0 Score: 1 to 4

0 Score: 1 to 4



0 Score: 1 to 4

External

5 Transaction Costs/

Competence (Coase)

-

Score: 3

0

Score: 1 to 4

0

Score: 1 to 4

0 to +

Score: 1 to 5

0

Score: 1 to 4

0

Score: 1 to 4

External

6 Competence (RBT)

-- Score: 2


0 Score: 1 to 4



0 Score: 1 to 4

External

7 Capability (RBT)

--- Score: 1





0 Score: 1 to 4

External

From this point, the above table is referred to as “the framework”. In plain terms, this framework in the model is explaining the rationale for internalisation or externalisation in terms of each of the above


levels as shown in Figure 3. In some cases in which identification of discrete activities is not clear, then upon attempting to answer the questions, if the user finds that there is conflict in their answers, for example, part of the activity would generate different answers to that of the other part of the activity, then the model is prompting the user to split the activity into two different activities. On the other hand, if the user is finding the same answers to all questions are being found across two or more activities, then this the model is prompting the user to consider combining the activities concerned. In other words, there may be some iteration between this task and the previous task.

Internalised activities 1 = Firm/government technically and/or organisationally better than market (not able to be

matched by market at decision date) 2 = Firm/government technically better than market (cheaper and/or better quality) arising

out of resources (e.g. access to resources; access to tacit knowledge in resources) 3 � = Firm/government organisationally better than market (cheaper and/or quicker response

but not better quality) activity may not be critical and some tolerance on response time) based on managing resources within the activity

• e.g. due to unpredictable nature of work government could employ a market firm to be available but this market firm may be idle and not contributing directly in activity e.g. prevention work reducing reactive work and/or elsewhere; plus market firm submitting overheads and profit and market firm’s operatives not able to be directly controlled/not as efficient in implementation of activity; plus higher external transaction costs associated with finding and appointment/delays if attempting to tender this work in ad hoc fashion/some minor variations and cost to contract management

External/ Internal activities 4 � = Potential for market to hold-up government in terms of securing more reward/costly

variations (time and/or money) in and/or arising from the activity (e.g. where activity is critical and emergency response required)

Externalised activities 5 � = Market organisationally better than government (cheaper and/or quicker response but not

better quality) activity may not be critical and some tolerance on response time) based on managing resources within the activity

• e.g. this work more predictable and less external transactions but importantly market have advantage in terms of accessing/aggregating higher levels of work to generate flow to allow more competence in organising and making less mistakes (planning; coordinating; procuring and management supply chain including bad behaviour of suppliers)

• So levels 3 and 5 concern supervision (make-or-buy is getting resources that are readily available and not dependent on critical workload and the organising these resources and so this seems to pick-up the issue of lack of capacity (budget and/or supervision) at margins of workload; if chose to internalise then may cost more if demand drops and/or see more mistakes through overstretched supervision

6 = Market technically better than government (cheaper and/or quicker and/or better quality)

arising out of resources (e.g. access to resources; access to tacit knowledge in resources) • So this derived from technical systems/procedures and firm specific tacit technical

knowledge that can’t be readily sourced in contrast to Pattern 5 in which resources that can be readily source to develop capability and may include increase capacity; developing this technical advantage requires critical threshold of work to justify investment

7 = Market technically and/or organisationally better than government (not able to be matched by government at decision date)

Figure 3. Explanation/rationale for internalisation or externalisation (in plain terms)

The model is able to explain either internalisation or externalisation based on the in-house capacity pertaining to the activity and which prevailed at the date the actual approach to procurement was


decided but, at the same time, the model also indicates whether this level of capacity was likely to have been economic (right level of capacity) or not economic (too much capacity or too little capacity). More specifically, the model is able to determine whether the level of capacity prevailing at the date the actual procurement approach was decided was economic or not by examining the answers and resultant symbols for the value/capacity variable and the frequency variable. That is, there is a close relationship between these variables as follows: • The value/capacity variable is accepted by its authors not to be a good economic predictor on the

basis that it is tautological (i.e. a firm may internalize an activity because this improves the firm’s profits and to improve profits the firm internalizes the activity; and this variable from RBT corresponds/co-varies exactly in proportion to each of the levels in the framework and so would be the only variable needed and which is unlikely given the very different logics behind the three theories being used and which each have something important to contribute

o This is then like the capacity the argument – a firm or in this case state government may internalize an activity because of current capacity or externalize an activity because of lack of current capacity; Capacity can then be misleading in pursuance of VfM and subject to and can be wield in a non-economic (e.g. political/financial influence) way.

o So the model makes the value variable capture capacity because it has explanatory merit but is mindful of its economic prediction weakness and hence the importance to read the result on the value/capacity variable in conjunction with the result on the frequency variable that is an objective economic indicator

• Measurement of frequency in accordance with TCE (buyer/firm side only)

• TCE envisages the frequency dimension in terms of 2 elements, that is the extent of a portfolio of “large and recurrent transactions” including but not exclusive to the focal transaction/activity; Such that more frequency justifies costs of internalization (internal transaction costs are spread). Frequency is then deployed in a classical production cost way to capture division of labour is limited by the extent of the market. That is, future demand/work is needed to see/justify benefits from investments in learning economies and /or technology to give economies of scale. This means frequency is measured on the firm’s side only (not both sides and in this case state government side only) and in terms of whether there are large transactions/activities, using same resources as the focal transaction/activity, and whether there are recurrent large transactions/activities, using same resources as the focal transaction/activity, being generated by state government and into the feasible future. The issue then becomes how do you measure “large” and “recurrent” in our context? ; Large is defined as a relative measure of projects greater than the majority of all new projects and recurrent is defined as being, at the decision date, a pipeline of large projects, that is a forward estimate of more than one year duration say up to point that state government can reasonably estimate (say to the next state election).

• Ideally, the value/capacity and frequency variables would positively correlate/match indicating about the right level of internalisation or externalization and if not and they negatively correlate/mismatch, then this indicates too much or too little capacity/internalisation or externalization.

• There may be, however, two exceptions to this: o First (proximity) : if state government agency/subsidiary is a subcontractor to a main

contractor, then it can said that this de facto internalised; This could happen when overall project too large for state agency/subsidiary but the key activity is within range of state government/subsidiary and the key activity is physically in the middle of the other key activities such that it is difficult to organise this activity in middle of other activities being organised by an external firm and to avoid contractor claiming being held-up by state agency/subsidiary

o Second (dynamic efficiency): there is also the possibility of state government giving work to the private sector to keep them turning over and this could explain situation where it appears best that government internalise activity but it has been externalised - the model would not support this because it’s static efficient approach but in this scenario neither the model/theory or practice can be criticised as being categorically


incorrect on economic grounds as the model is promoting short term economic efficiency whilst the practice is possibly promoting longer term efficiency.

• Given the inherent tautological weaknesses with the value/capacity variable, then frequency gives the appearance of the next best/powerful variable in terms of, on the surface, being able to singularly (without reference to the other variables) explain internalisation or externalization. However, it’s not this straightforward, as there is the issue that if the firm, or in this case state government, simply kept increasing its staff on activities that are large and recurrent (elements of the frequency variable) and/or kept adding activities that are large and recurrent to its portfolio of internalized activities, then it increases its bureaucracy costs and costs associated with low power incentives and which in total amount to increasing internal transaction costs. In other words, Nobel Laureate Coase (1932) explains that where internal transaction costs become too great, then externalization of all and/or part of large and recurrent activities (make and/or buy) or not internalizing the next large and recurrent activity at the margin may be efficient and explains why we don’t have one big firm/government in market economies. In summary then, a high frequency score on an activity (large and recurrent) begins to suggest that this activity could be efficiently internalized but is not the entire story. To help complete the picture, a high frequency score on a particular activity needs to be read in conjunction with at least a high score on the asset specificity variable and as high scores are also read on the other variables, then the economic case for internalisation increases in strength.

• On measuring frequency, it’s not tautological to ask what is the most frequency occurring project i.e. typical and size project - in terms of capital value, as this concerns the entire project discernable by its physical boundaries e.g. geography/length in roads or GFA/area in health projects; the model then determines whether the entire project is best sized as one contract or broken down into smaller sized contracts

o Note as model/questions are configured in terms of measuring theoretical variables at project level and so it’s limited to evaluating procurement strategy at level of project but could be configured to measure/address possibility of linked/bundled/program of projects that may change balance of advantage for example say from government to market in remote areas.

Stage 1 Task C market analysis (first-order analysis) The model prompts the user in this task to perform a check or Structure-Conduct-Performance (SCP) analysis on each activity to corroborate the levels identified though the framework. The model then prompts the user to consider any Level 7 activities arising out of the scale of the activity’s work in the project and guides the user to attempt to breakdown the activity concerned into two or more sub-activities to avoid a lack of competition/possibly more powerful sub-market firm (e.g. subcontractor) as part of supply chain managed by a different upstream market firm (e.g. main contractor). Stage 2 Bundling analysis (second-order analysis) From this point in Stage 2, the focus is now on the procurement of the externalised activities only and mindful that it could be that the model may include an activity that was actually internalized by treating it as externalized (i.e. a mismatch between the capacity and frequency scores in which capacity is positive but frequency is low) or exclude an activity that was actually externalized and treat this as internalised (again a mismatch between capacity and frequency but this time in which capacity is negative and frequency is high). In this task, the model prompts the user to consider any potentially troublesome activities that will lead to firm/state government becoming dependent/potentially vulnerable and market firms becoming too powerful i.e. a Level 4b activity (hold-up ex post/post contract arising out of high asset specificity and uncertainty) and the model also guides the user to consider any residual Level 7 activities arising from size (that can’t be broken-down because of physical/proximity issues), as well as any Level 7 activities arising out of rare technology leading to lack of competition and market power to set prices ex ante/pre-contract and again hold-up problems ex post). More specifically, the model prompts the user to consider:


• Can any of the potential opportunistic behavior by market firms associated in these troublesome activities be checked by the firm being engaged at the head of the supply chain/contract that includes these activities; That is, the firm e.g. main contractor at the head of the supply chain that is coordinating/subcontracting these troublesome using the threat of future work to address any potential negative opportunistic behavior

• If not, then consider either o A special relationship with the firms providing these troublesome activities through the

firm being engaged at the head of the supply chain that includes these troublesome activities (e.g. use of Prime Cost Sum and a nominated subcontractor/supplier arrangement) – this then drops down into an exchange relationship consideration; or

o A separate contract between the firm/state government and the market firm providing the troublesome activity i.e. exclude the troublesome activities from any subsequent bundle.

The model then guides the user to bundle the remaining key activities into main activities: • Design

• Construction

• Operations and

• Maintenance o In doing so, the user is now effectively undertaking a second-order analysis (point at

which current practice often starts) in terms of assessing the activities of organizing/managing each of these main activities e.g. main contractor on construction

� Again, the model prompts the user to attempt to breakdown each main activity if the size and/or complexity of a main activity leads to Level 7 (again, leading to lack of competition and market power to set prices ex ante and hold-up problems ex post)

From here, the user is able to Identify any potentially viable major activities i.e. bundle(s) of either DCO main activities or DCM main activities or DCOM main activities that can be market sounded as a PPP(s) or otherwise via a contract(s). If not, then the default is a separate D; separate C; separate O; and separate M main activities organized by government (either directly and/or using market firm (e.g. PM/consultant).

In other words and in terms of the key mechanism being deployed to realize benefits of the integration of Operations and Maintenance in with Design and Construction, in pursuance of increased benefits and lower whole-life costs, the model considers that private finance holds the greatest high power incentive and followed by contract. If the market is not responding positively to using these mechanisms then bureaucracy is used to perform the integrating role.

An exception to seeking to integrate operations and/or maintenance in with Design and Construction occurs when neither the buyer (government) nor supplier(s) (private sector have appreciably differential advantages concerning operations and/or maintenance and a design that is focused on minimises capital cost would not significantly undermine whole-life costs. In this case, a Design and Construct (or variation of this approach including Managing Contractor or Early Contractor Involvement) is prompted by the model for consideration.

In summary there are 3 levels of activities: • Key activities = about highest level of market specialization at the nature of key activity

• Main activities = Set of D or C or O or M key activities

• Major activities = Set of DCO or DCM or DCOM main and/or key activities

And organisation and management includes: • Planning and coordination of timing of tasks within activities; and key activities within main

activities; and main activities within major activities; Hence in one project there are many project managers operating at different levels

• Procurement and supply chain coordination/control e.g. including quantity surveyors etc

Stage 3 Exchange relationship analysis (third-order analysis)


The model finally guides the user to measure the TCE variables only in terms of determining the best value exchange between state government and the market firm that can deliver each major activity (bundle) and/or each main activity and/or each key activity. The exchange relationship ranges from:

• Efficient relational exchange o E.g. Alliancing (pain/gain share regime)

• Efficient discrete exchange o E.g. arms-length exchange, or standard neo-classical contracting such as lump-

sum/fixed-price contracts with mechanising to resolves dispute by third parties

• Short-term inefficient discrete exchange o E.g. non-standard contracts written with bespoke credible threats for non-performance

• This measurement now includes the task of planning and coordination across key activities and so this measure of main and major activities may produce a different result on the TCE variables than that recorded across the TCE results measured in the constituent key activities.

The TCE scores are then matched with the exchange relationship variables to determine the best exchange as shown below in Figure 4.

Figure 4 Exchange relationship (Bridge, 2008)

In the continuum between relational exchange and discrete exchange in Figure 4 connections can be made with three broad categories of contract in PAT as shown in Table 2.

Table 2. Summary of outcome, hybrid and behaviour based contracts in PAT

Behaviour-based Hybrid (Outcome-based and Behaviour-based)

Outcome-based

• Cost-plus • Low power incentive • Principal's risk to completion • Agency cost

o specifying behaviour o monitoring behaviour o outcome uncertainty o very complex project o less goal alignment

• Target outturn costs/guaranteed construction sum linked with gain share/pain share regime

• risks balanced between agent and principal

• suitable for outcome uncertainty


o specifying outcomes o verifying outcomes o risk premium o Suitable for information

asymmetry o outcome certainty


That is, there is point in relational exchange that is envisaged best governed by internalization and exchange within the firm and depicted by the dotted part of the line to the left in Figure 4. This part of the line can be connected to behavior-based contracts in PAT. The solid part of the line starting from the left in Figure 4 can be connected to contracts that are somewhere between behavior-based contracts and outcome-based contracts (or hybrid contracts) in PAT. Finally, the remaining part of the solid line in Figure 4 can be connected to outcome-based contracts in PAT. By degree, the further this is to the right of the solid line, then these contracts incorporate greater information asymmetry in favour of the supplier (and lack of knowledge held by government to collaborate in the exchange) and therefore power held by the supplier. Such that, a greater level of credible threats become necessary to pre-empt hold-up but at the same time are costly to write and are priced by the supplier. Consequently, this is not an efficient exchange at one level but nonetheless is the most overall efficacious approach to governance in terms of addressing the risk in the exchange. Stage 1 Task B Make-or-Buy Analysis (first-order analysis)

E ff ic ie n t E x c h a n g e

( n at u ral ly o c c u rrin g )

+ + + R e la tion a l +

In e ff ic ien t E x c h an g e

+ D iscr ete + + +

E ff ic ien t E x c h an g e

+ D iscr ete + + +

E x ter n a l

E x cha n g e

T C E

A sset S p ecif ic ity +

U n cer ta in ty + /+ + +

F r e q u e ncy +

0 0

0 /+ + + 0 /+ + +

+ /0 0

+ + + +

+ + /+ + +

0 0

cba

1 2 3


Standard questions on each activity and answered mindful of circumstances at very approximate date that actual procurement decision made. A key internalized activity: Activity description • Indication of distinct knowledge/skill set:

• Indication of non-trivial:

• Actual internal or external: • Draft answers = closest (pattern) and level: Refer to Table 1

1. Capacity:

a. In terms of staff employed by state department (including within a sub-agency/subsidiary) to work full-time on the activity in the case study, how much would this have been resisted by capacity constraints (would have exceeded relevant staff salary scales/budget and/or overstretched supervisory staff)?

Please circle: Significantly beyond capacity 1 2 3 4 5 6 7 Comfortably within capacity

2. Rarity: a. How much was the knowledge and skills required in this activity in the case study

possessed by all top-tier specialist local firms capable of delivering the activity? Please circle: possessed by all firms 1 2 3 4 5 6 7 possessed by few firms


Please circle: plentiful supply of firms 1 2 3 4 5 6 7 scarce supply of firms

3. Costly-to-imitate: a. How difficult would it have been for a performance brief or manual to be written


Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult b. How difficult would it have been for all top-tier specialist local firms capable of

delivering the activity to develop the same knowledge and skills required to deliver the activity in the case study?

Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult 4. Asset specificity:

a. How much investment and/or time would be needed to acquire new knowledge (including software/hardware) and/or adapt existing knowledge (including software/hardware) to deliver the activity in the case study; that is, beyond knowledge already possessed by a top-tier specialist local firm with capability to deliver the activity?

Please circle: minimal investment and/or adaption time 1 2 3 4 5 6 7 substantial investment and/or adaption time

b. At any stage in the delivery of the activity in the case study project, how much negative impact (in terms of effect on the project’s timeline and/or budget) would have been experienced by the state department, if the state department had decided to externalise the activity and replace the firm providing the activity to the case study?

Please circle: minimal impact 1 2 3 4 5 6 7 substantial impact 5. Uncertainty:

a. How much were the tasks (types and amount of time) in the activity required in the case study straightforward to predict by someone with expertise in the activity in major projects?

Please circle: extremely straightforward 1 2 3 4 5 6 7 extremely difficult


b. How much were the tasks (types and amount of time) in the activity required in the case study likely to be subject to changes caused by unknown factors relating changes in physical conditions and/or state department changes (e.g. scope) during the period allowed for the activity in the case study?

Please circle: extremely small 1 2 3 4 5 6 7 extremely high

6. Frequency (Version A at the project level please see pages 12-14):1 a. In the state electoral period surrounding the case study (from commencement of

state government term to end of term incorporating the case study procurement decision date), how did the case study compare with the state department’s most frequently occurring or typical new project in terms of capital value and what is the capital value size category of the case study? Please circle:

1. = Case study is not typical and a small project ($1 to $10 million) 2. = Case study is not typical and a moderately-sized project ($10 to

$100 million) 3. = Case study is not typical and a large project ($100 to $250 million) 4. = Case study is not typical and a very large/extremely large project

(greater than $250 million) 5. = Case study is typical and a small to moderately-sized project ($1 to

$100 million) 6. = Case study is typical and a large project ($100 million to

$250million) 7. = Case study is typical and a very large/extremely large project

(greater than $250million)

6. Frequency (Version B at the network level please see pages 12-14): a. In the state electoral period surrounding the case study (from commencement of

state government term to end of term incorporating the case study procurement decision date), how did the operations and maintenance to case study compare with the state department’s most frequently occurring or typical operations maintenance and what is the geographical scope of the state department’s operations and maintenance responsibility? Please circle:

1. = Case study was not typical and a small scope (main roads in state capital city)

2. = Case study was not typical and a moderately scope (main roads in state metropolitan area)

3. = Case study was not typical and a large scope (main roads statewide)

4. = Case study was not typical and a very large/extremely large scope (main roads statewide and suburban roads)

5. = Case study was typical and a small to moderate scope (main roads in state capital city to metropolitan area)

6. = Case study was typical and a large scope (main roads statewide) 7. = Case study was typical and a very large/extremely large scope

(main roads statewide and suburban roads)

1 Please see previous explanations on the use of different versions of frequency measurement and that in health projects only Version A is used and measured once for all activities, whilst in road projects version A is used and measured once for all design and construction activities and version B is also used and measured once for all operation and maintenance activities.


A key externalized activity: Activity description • Indication of distinct knowledge/skill set:

• Indication of non-trivial:

• Actual internal or external: • Draft answers = closest (pattern) and level: Refer to Table 1

1. Capacity: a. If state health department had employed additional/new staff (including within a

sub-agency/subsidiary) to work full-time hours on this activity in the case study, then how much would this have been resisted by capacity constraints (would have exceeded relevant staff salary scales/budget and/or would have overstretched supervisory staff)?

Please circle: Significantly beyond capacity 1 2 3 4 5 6 7 Comfortably within capacity

2. Rarity: a. How much was the knowledge and skills required in this activity in the case study

possessed by all top-tier specialist local firms capable of delivering the activity? Please circle: possessed by all firms 1 2 3 4 5 6 7 possessed by few firms


Please circle: plentiful supply of firms 1 2 3 4 5 6 7 scarce supply of firms

3. Costly-to-imitate: a. How difficult would it have been for a performance brief or manual to be written



b. How difficult would it have been for all top-tier specialist local firms capable of delivering the activity to develop the same knowledge and skills required to deliver the activity in the case study?


4. Asset specificity: a. How much investment and/or time would be needed to acquire new knowledge

(including software/hardware) and/or adapt existing knowledge (including software/hardware) to deliver the activity in the case study; that is, beyond knowledge already possessed by a top-tier specialist local firm with capability to deliver the activity?

Please circle: minimal investment and/or adaption time 1 2 3 4 5 6 7 substantial investment and/or adaption time

b. At any stage in the delivery of the activity in the case study project, how much negative impact (in terms of effect on the project’s timeline and/or budget) would have been experienced by the state department, if the state department had decided to replace the firm providing the activity to the case study?

Please circle: minimal impact 1 2 3 4 5 6 7 substantial impact 5. Uncertainty:

a. How much were the tasks (types and amount of time) in the activity required in the case study straightforward to predict by someone with expertise in the activity in more general in major projects?



b. How much were the tasks (types and amount of time) in the activity required in the

case study likely to be subject to changes caused by unknown factors relating changes in physical conditions and/or state department changes (e.g. scope) during the period allowed for the activity in the case study?

Please circle: extremely small 1 2 3 4 5 6 7 extremely high 6. Frequency (Version A at the project level please see pages 12-14):2

a. In the state electoral period surrounding the case study (from commencement of state government term to end of term incorporating the case study procurement decision date), how did the case study compare with the state department’s most frequently occurring or typical new project in terms of capital value and what is the capital value size category of the case study? Please circle:

1. = Case study is not typical and a small project ($1 to $10 million) 2. = Case study is not typical and a moderately-sized project ($10 to

$100 million) 3. = Case study is not typical and a large project ($100 to $250 million) 4. = Case study is not typical and a very large/extremely large project

(greater than $250 million) 5. = Case study is typical and a small to moderately-sized project ($1 to

$100 million) 6. = Case study is typical and a large project ($100 million to

$250million) 7. = Case study is typical and a very large/extremely large project

(greater than $250million)

6. Frequency (Version B at the network level please see pages 12-14): a. In the state electoral period surrounding the case study (from commencement of

state government term to end of term incorporating the case study procurement decision date), how did the operations and maintenance to case study compare with the state department’s most frequently occurring or typical operations maintenance and what is the geographical scope of the state department’s operations and maintenance responsibility? Please circle:

1. = Case study was not typical and a small scope (main roads in state capital city)

2. = Case study was not typical and a moderately scope (main roads in state metropolitan area)

3. = Case study was not typical and a large scope (main roads statewide)

4. = Case study was not typical and a very large/extremely large scope (main roads statewide and suburban roads)

5. = Case study was typical and a small to moderate scope (main roads in state capital city to metropolitan area)

6. = Case study was typical and a large scope (main roads statewide) 7. = Case study was typical and a very large/extremely large scope

(main roads statewide and suburban roads)

2 As per internalised activities, please see previous explanations that explain the use of different versions of frequency measurement and that in health projects only Version A is used and measured once for all activities, whilst in road projects version A is used and measured once for all design and construction activities and version B is also used and measured once for all operation and maintenance activities.

Appendix J – Case study database 355

Appendix J – Case study database

Road case #R1

List of documents and drawings 1 Project Contract document Volume 1, including letter of acceptance, contract document list, a set

of tender correspondence, schedule of rates, planned cashflow schedule, tender program, daywork schedule, environmental plan and traffic management plan.

2 Project Contract document Volume 2, comprises of tender schedule, general conditions of contract, supplementary conditions of contract, standard documents, project specific conditions of contract and list of drawings.

3 Project Contract document Volume 3, Entire Road construction contract, Volume 3, comprising of Information for tenderers, conditions for tendering, tender forms and tender schedules, general conditions of contract, supplementary conditions of contract, standard documents, project specific conditions of contract and list of drawings.

4 Project Contract document, Road construction contract, Volume 4, consist of contract drawings. 5 Contract award, sourced from contract database. 6 Responses to survey of major road and health projects for Road Case #1.

Road case #R2

List of documents and drawings

1 Letter, subject: Proposal for extension of consultancy services. 2 Project brief. 3 Memorandum, subject: Project – interim purchase order. 4 Functional brief. 5 Final project alliance agreement, comprising of, Alliance principles and fundamentals;

governance, management and organisation; project referral and development; project implementation phase; insurances, payment, variations and financial audits; disagreements, indemnities and liabilities; termination, default and withdrawal; confidentially and intellectual property; and general provisions and interpretation.

6 Project control procedure, final issue. 7 Breakdown of agreed target cost estimate (Stages 1, 2 and 3). 8 Independent estimator review revision A. 9 Project Stage 1 and 2, Design management plan, issue 3. 10 Project work package list, updated. 11 Risk opportunity action plan, updated. 12 Document register – Vendor data. 13 Site plan option B (marked up). 14 Project construction program (stages 1, 2 and 3), updated. 15 Estimate of project scope of work and ITS equipment count table. 16 Responses to survey of major road and health projects for Road Case #2.

356 Appendix J – Case study database

Health case #H1 List of documents and drawings 1 PPP investigation report, 51 pages. 2 Invitation for expressions of interest, 51 pages. 3 Request for detailed proposals, volume 1, including indicative phasing program, and site plan. 4 Project deed, 552 pages. 5 Memorandum: Subject: Functional layout options.. 6 Deed of amendment. 7 Project summary, 52 pages. 8 Development application, Title page. 9 Development application, Overall site plan. 10 Development application, Overall site plan (proposed). 11 Development application, Location plan. 12 Development application, Location plan (proposed). 13 Project plan application, Lower ground floor plan. 14 Project plan application, Ground floor plan 1 of 2. 15 Project plan application, Tertiary mental health unit, ground floor plan. 16 Project plan application, General hospital first floor plan. 17 Project plan application, General hospital second floor plan. 18 Project plan application, General hospital third floor plan. 19 Project plan application, Existing floor plan 1 of 2. 20 Project plan application, Refurbishment of administration and clinical services, proposed floor

plan. 21 Project plan application, Existing tertiary floor plan 1 of 2. 22 Project plan application, Existing tertiary floor plan 2 of 2. 23 Project plan application, General hospital elevations 1 of 2. 24 Project plan application, Mental health elevations and section, elevation 2 of 2. 25 Project plan application, Landscape plan 1 of 2. 26 Project plan application, Landscape plan 2 of 2. 27 Detail and contour plans.

Health case #2 List of documents and drawings 1 Tender evaluation and recommendation report for managing contractor. 2 Tender evaluation and recommendation report for managing contractor. 3 Memorandum, subject: recommendation to accept GCS offer from the managing contractor for

the project. 4 Project definition plan (PDP) volume 1 final, 445 pages, gives the scope of services designed by

consultants, including the basic construction program and PDP estimate. 5 Project definition plan volume 2 final, 608 pages, comprising functional briefs for Front of house,

ambulatory care, general inpatient, mental health inpatient, clinical support, therapies and support, administration, education and training and corporate services.

6 Project definition plan volume 3 final, comprising room data sheets for Front of house, ambulatory care, general inpatient, mental health inpatient, clinical support, therapies and support, administration, education and training and corporate services.