From Washington to ROME: A Road Less Traveled by Public Sector

Reformers

by

Matthew Andrews1 and Anwar Shah2

Second draft, September 2001Do not cite without authors’ permission.

Comments welcomed

1. Matthew Andrews can be contacted at mrandrew@amailbox.syr.edu2. Anwar Shah can be contacted at ashah@worldbank.orgThe views expressed in this paper are personal views of the authors’ alone and should not be attributed to the World Bank Group.

- 1 -

contentsINTRODUCTION:TOWARDS EFFECTIVE PUBLIC SECTOR REFORM IN THE DEVELOPING WORLD 3SECTION ONE: THE NEED FOR ROME 11CHAPTER ONE:THE NEED FOR PUBLIC SECTOR REFORM IN THE DEVELOPING WORLD 121.1. Administrative and governance weaknesses in the developing world 121.2. Chapter conclusion 18CHAPTER TWO:COMMON REFORM RESPONSES IN THE DEVELOPING WORLD 202.1. Elements of reform response to administrative failure 202.2. Chapter conclusion 28CHAPTER THREE:UNDERSTANDING REFORM EXPERIENCE IN THE DEVELOPING WORLD 293.1. Identifying why reforms ‘fail’ in the developing world 293.2. Chapter conclusion 40SECTION TWO: PRESENTING ROME 41CHAPTER FOUR:ROME: RESPONDING TO REFORM NEED 424.1. ROME: Making sense of the reform puzzle 424.2. ROME: completing the puzzle, tackling the problems 464.3. Chapter conclusion 48CHAPTER FIVE: PARTICIPATORY-DECENTRALIZATION IN THE ROME MODEL 505.1. Providing citizens with a voice, creating demand for results 515.2. Decentralizing governments to access and respond to citizen voice 725.3. Conclusion: The citizen-government contract 81CHAPTER SIX:RESULTS ORIENTED MANAGEMENT IN THE ROME MODEL 856.1. Results oriented relationships in the administration 876.2. Providing the tools required for a results-oriented response to citizen demand 996.3. Conclusion: Facilitating results-oriented and responsive administration 110CHAPTER SEVEN:RESULTS ORIENTED EVALUATIONS IN THE ROME MODEL 1147.1. Involving citizens in results-based evaluations 1167.2. Transforming formal evaluations processes 1207.3. Conclusion: Results oriented evaluation and incentives to govern well 133CHAPTER EIGHT:IMPLEMENTING ROME: ENSURING THE ROAD TO ROME IS NOT ONE OF WRACKS AND RUINS 1368.1. Institutionalizing ROME 1368.2. Implementing ROME 1388.3. ROME, the institutional environment and reform space 1418.4. Conclusion: Implementing ROME by increasing ‘reform space’ 153CONCLUSION:ROME: AN EFFECTIVE PUBLIC SECTOR REFORM FOR THE DEVELOPING WORLD 155SELECTED REFERENCES 156

- 2 -

INTRODUCTION:TOWARDS EFFECTIVE PUBLIC SECTOR REFORM IN THE DEVELOPING

WORLD

Public sector reforms are pervasive in the developing world. Unfortunately the failure of public sector reform is equally pervasive in the developing world. Research by organizations like the World Bank find that many interventions fail to make lasting and effective impressions on recipient countries. Unresponsive, unaccountable, inefficient and ineffective bureaucracies seem impossible to change.

The problem with reform in the developing world

Reform options ranging from macroeconomic stabilization approaches to capacity-building initiatives and decentralization have failed to deliver on their promises of change in the last two decades. Consider evidence and perspectives on the issue:

Public expenditure adjustment is a major focus of public sector reforms in the developing world. A 1997 World Bank report of public expenditure reforms under adjustment lending examined the degree to which reforms between 1979 and 1994 actually helped to shape spending policies in developing nations (Huther, Roberts and Shah, 1997). The paper concludes that reforms based on adjustment had “small positive effects on expenditure patterns” but also points out a number of reform shortcomings. In particular, the report suggests that reforms have not significantly improved expenditure quality, or changed the government incentives and mind-set.

As with many reform observers, Merilee Grindle writes that, “Public sector organizations perform poorly in many developing countries; in some places they barely function at all” (Grindle, 1997: 481). On the subject of dominant 1990s reforms focused on improving public performance through civil service adjustments, she is equally pessimistic, quoting De Merode and Thomas that “no conclusive evidence was found of better pay and leaner staffing…leading to major product9ivity gains.”

The Medium Term Expenditure Framework (MTEF) is a promising reform in the developing world. The South African example of MTEF is seen as a particularly ‘good’ application of the reform. The South African government introduced the MTEF in 1996 as part of a general policy package emphasizing fiscal discipline, deficit reduction, and development (Walker and Mengistu, 1999). The MTEF is part of a vibrant set of reform policy and legislation. The reform is considered best practice in the development community (World Bank, 1998) largely because it has ‘led’ to reduced deficits (a key budgetary outcome), with deficits (as a percentage of total expenditure) falling from an average 15.86 between 1981 and 1993 to 11.25 between 1994 and 1999. Unfortunately, the MTEF reform tenure has also seen decreases in key expenditure types: capital spending comprised 9% of spending between 1981 and 1993, but only comprised 4% between 1994 and 1999; Fixed assets

- 3 -

made up 2.94% of total expenditure up to 1993, but had a 1.49% share between 1994 and 1999 (South African Reserve Bank, July 2000). These results raise questions about MTEF’s ‘success’: Is the MTEF responsible for the improved fiscal discipline, or was it a factor related to economic upturn in the period? While fiscal discipline appears to have improved, other fiscal outcomes have not. Strategic spending has dropped in real terms and there is a concern critique that departments are underspending their budgets, leading to inefficiencies and low levels of budgetary responsiveness (Business Day, April 9 2001). Can a reform be successful if it improves one outcome but not others? Is the decline in capital spending a manifestation of Alesina and Perotti’s (1996) observation that medium term budgeting allows creative accounting and the postponement of ‘difficult’ projects?

Even recent reforms are problematic…

Recent solutions to the problems of governance and administration in the developing world are taken from the results and performance movement that seems to have proven effective in countries like New Zealand. ‘Results oriented management tools’ copied from such experience are seen in nations from South Africa to Malaysia to Uganda, Chile, Brazil, Indonesia, India and Mauritius. Frank discussion with development experts involved in the application of such tools, and with local reformers, indicate that these ‘tools’ are not taking shape in most settings. Reform evaluations and case studies of ‘results oriented management’ and ‘new public management’ tell similar stories. Consider evidence on the issue:

Desai and Imrie (1998) provide interesting observations of reform to the “bureaucratic and protectionist regime” of India. New public management and performance-based interventions in India date back a number of years. In these reforms, the authors find more rhetoric than results, however, especially pertaining to proposed participation in the initiative: governments remain insular even though policy and legislation requires that they open their systems to citizens. Desais and Imrie imbibe a generally negative tone, stating that: “The possibilities for enhanced democracy within the managerialist state are being circumscribed by a heightening of clientilism and co-option” and “Much of the new managerialism is….contradictory and flawed, characterized by de-democratizing tendencies and a fixation with procedural and technical processes” (1998: 645).

Problems with new public management reforms are also evident in Africa. In Malawi, a number of interventions were introduced between 1991 and 1997 (Adamolekun, Kulemeka and Laleye, 1997). These included pay reforms, the introduction of a Medium Term Expenditure Framework, and a privatization program. If one evaluates the reforms in terms of their implementation, they appear generally successful. Most of the mechanisms and management processes have indeed been implemented. If one evaluates the reforms in terms of their impact on behavior, they do not appear that successful. There is little evidence of actual change—the majority of civil servants still hold to old patterns of

- 4 -

interaction and behavior and the old rules of organization. Further, reform elements do not always support each other (decentralization of government and centralized financial planning, for example), and although some technical process changes have been praised, old processes remain in operation.

Larbi (1998: 378) provides an example of problematic new public management interventions in the Ghanaian health sector, where managerial reforms at central government level are blamed for the failure of “decentralization and other long-term institutional reforms” because they facilitated a tightening of central controls over the key managerial tasks of spending and staffing. Where the general basis of new public management in countries like New Zealand is results based flexibility, in this situation the reform is manifest as a strangling, controlling intervention used by central authorities to expend their interests.

Integrated Development Plans (IDPs) are intended to facilitate performance-based, participatory government in South African municipalities (Africa, 1999; Otzen et al., 1999; Wallis, 2000). All municipalities are required by law to introduce such plans into their management processes as the primary driver of their operating and capital budgets. The plans are intended as the main means by which local governments develop strategic policy capacity, mobilize resources, and target such resources to strategic activities. The national Department of Provincial and Local Government has recently begun to ask how well IDP implementation is progressing—whether the reform is proving a ‘success’. While the results of in-house surveys are reported under a heading that “most municipalities are now involved in integrated development planning” (DCD,1999) the surveys actually show that the reform is not taking off as planned: as of 31 March 2000 only 46% of the 483 municipalities responding to the quarterly Project Viability survey had formally approved IDPs in place. The same percentage answered that communities are involved in updating IDPs, while fewer answered that the IDP constituted the budget driver its was intended to (Department of Provincial and Local Government, 2000). Furthermore, case studies reveal that IDPs are having no effect on participation (with the participation element sorely limited). The IDP also lacks significant effect on failures associated with administrative information flows. In essence, observers question whether the reform has had any effect on incentives, behavior or outcomes. Even where it is in place, IDP has not countered ‘business as usual’.

Is there hope for a ‘better’ approach to reform?

Given this evidence, it seems as if another reform idea is proving an unsure ‘solution’ to public sector problems in the non-industrialized world. But there is hope, evident in isolated cases of effective results-oriented reform in developing countries. From the influence civic results demand has had on bureaucratic responsiveness in Rajasthan in India, to the successes with community-based evaluations in Uganda and Malaysia, it is possible to identify experiences where the focus on public sector outcomes is indeed

- 5 -

causing governance structures and incentives to change and outcomes to improve. Consider these cases:

The Malaysian public sector adopted local level, citizen-focused, results oriented government in the 1990s. The reform is described as having “encouraged a change in the mindset of public officials, who are now required to search for more efficient and effective methods for the delivery of public services that satisfy customers” (Chiu,1997: 175). The reform envisions the civil service and the government working closely with business and civic organizations, leading to mutually beneficial relationship (Abdul Karim, 1997). The changed mind-set has gone a long way in improving the performance of the Malaysian public sector (Mohamad,1997).

In Uganda decentralization and participatory-based performance budgeting reforms are cited as reasons why crucial sectors like education and health are seeing a turn-around and increased responsiveness(Kisubi,1996; Reinikka,1999; Brinkerhoff,2000). The turn-around is remarkable because the three biggest problems argued (by authors like Schick,1998) to limit reforms are evident: The traditional bureaucracy followed no rules and had little capacity, the service types resemble complex craft and coping products, and there is no private market from which competition and competitive forces can be felt. The results are clear, in a survey in 1998 the Ministry of Education “found major improvements in the flow of funds” (Reinikka,1999). The mix of decentralization, contract, and evaluation (enabled by improved information access) led to increased financial accountability.

Another example comes from Bolivia, where the Popular Participation and the creation of Territorial Base Organizations set up to monitor the provision of public services (community evaluation) have had a marked impact on accountability and responsiveness. “While the functioning of the system is still hampered by a lack of resources, illiteracy, difficult transport, unclear procedures and the absence of transparency and of a democratic culture, there are also indications of increased accountability” (Schneider, 1999, 526).

A final example comes from Rajasthan province in India. It points to the centrality of information in reform). It also shows that reform can emanate from within civil society. Essentially, the ‘reform’ in Rajasthan involves local communities demanding the rights to photocopy government documents and then performing informal, community-based evaluations of their governments . The community evaluation devices take a number of forms, including public hearings at which detailed accounts derived from public expenditure records and other documents are read aloud to assembled villagers. “Through this direct form of ‘social audit’” discrepancies have been identified and public officials (politicians and administrators) have been bought to account. Furthermore, there is now an active process of social negotiation regarding how government documents are structured so that they maximize the civil society window (evaluations) and facilitate government learning from civil interaction (ROM) (Jenkins and Goetz, 1999).

- 6 -

Learning from reform ‘success’ and charting a new reform approach, ROME

Beyond identifying these positive cases, it is also possible to identify the factors that separate reform experience in these ‘successful’ settings with the prevailing, ‘unsuccessful’ results-oriented reform applications. The dominant observation when comparing past and current ‘underachieving’ reform examples with the sub-set of successes, is that results oriented reform only appears to effectively change the incentives and focus of governments when it is sourced within representative, participatory structures at the local level. Essentially, ‘bottoms-up’, decentralized participatory structures are the basis of effective results oriented reform in developing countries. These structures provide the source of results identification (citizens communicate their demands) and the prime location for results evaluation (citizens are in the best place to identify if their demands are met or not).

This monograph builds the observation that localized participation is central to results-oriented reform in the developing world into a model of reform, Participatory, Results Oriented Management and Evaluation (ROME). The model is inherently bottoms-up, combining three elements: participatory decentralization, results-oriented management and results-oriented evaluation:

Participatory decentralization provides the foundation in which to locate results oriented management and evaluation ideas. Through this element, political and administrative entities have their incentives defined as they focus on citizens at the local level, are directed to the results that matter most (as identified by citizens), and are ideally located for capacity-enhancing, results-oriented civil society partnerships. This element counters the top-down, paternalistic model common in developing countries with bottoms-up, localized governance structures in which citizens drive, inform and capacitate their own governments.

Results-oriented management (ROM) takes place within the context of the participatory, decentralized governance system in ROME (unlike its top-down structure in countries like New Zealand). ROM provides tools like performance-based budgeting that are only effective in this setting, where citizens can hold governments accountable for their implementation and ensure their influence. ROM tools, and the implied delegation within the ROM approach, are the basis of a major public-sector mind-set change in the developing world: from a process and probity concentration in administration and management to an output and outcomes focus, in which government focuses on what it does for citizens (as measured in outputs and outcomes produced).

Results-oriented evaluation (ROE) is the final element of ROME. ROE solidifies the incentives created by results-oriented management techniques, providing an important formal and social ‘check’ on governance and an important feedback loop for public actors to improve their operations. Politicians and administrators are held accountable for the results they produce as these results are regularly evaluated—

- 7 -

through internal and external audits as well as through community evaluations (or social audits). These evaluations create a link between the participations and results oriented management elements of the reform model and help to ensure that governments are accountable and transparent.

These elements will all be recognized by development experts and government officials in countries seeking reform solutions, who might ask: “what’s the difference between ROME and reforms already in place?” Indeed, reform elements like these have been regularly mentioned in theories like new public administration (Marini, 1971; Frederickson, 1980), new public management (Osborne and Gaebler, 1982; Barzelay, 2000) and new institutionalism (North, 1990; Ostrom, Gardner and Walker, 1993; Picciotto, 2000). They are also commonly alluded to in the development literature, with Picciotto’s recent statement standing out: “To achieve poverty reduction, we need new concepts that are holistic, results based, and participatory” (Picciotto, 2000: 361).

But ROME is different to the established reform theories and practices.

The ROME differences

ROME differs on two dimensions: (1) The way in which elements are organized in ROME is significantly different from conventional reform approaches, and (2) The participatory decentralization element is more pronounced in ROME than it is in other reform approaches, constituting the proverbial reform ‘stumbling block’. In terms of the first of these, convention in the development community emphasizes the tools of results-oriented management in centralized reform initiatives. Participation, decentralization and evaluation are politically correct ‘add-ons’ downstream, with participation generally limited to ‘professionally appropriate’ groups with pre-existing access to government, decentralization occurring only when local governments have the capacity—which is never, and evaluation attached as a concept needed ‘when there is something to evaluate’ by central authorities (limiting the change effect evaluations can embody). In contrast to this conventional approach, ROME emphasizes participatory decentralization as the foundation of results oriented management and evaluation, with the latter pair implemented jointly in response to civic demand at the local level. The model so depends on citizen participation and influence that a non-reform is considered a more viable option than a non-participatory reform.

Institutional environment matters more than institutions

Theoretically, ROME presents a different approach to the popular new institutional thinking in reform. Whereas the 1990s saw the emphasis on getting institutions right in government, which translated into exporting ‘the right institutions’ from ‘successful’ industrialized nations to other nations, ROME emphasizes getting the ‘institutional environment right’ so that nations can develop their own appropriate institutions. This amounts to a new ‘new institutional emphasis’ on where institutions come from, as

- 8 -

reflected in Alesina and Perotti’s statement that “systematic research that tries to explain institutional building is overdue and is an excellent area of future work” (1999:15).

From the cases of successful reform, it is possible to identify that a specific foundational environment is necessary for effective reform in the developing world: a localized, participatory social and governance structure. This environment is the focus of ROME, which sees the major reform interventions as developing strong social institutions and legal institutions that bring government close to citizens. With these in place, evidence shows that governments automatically change their mindsets and move to adopt results oriented rules and tool—because citizen demand create an incentive for such behavior.

The structure of this monograph

Shah’s recent paper titled “Governing for Results in a Globalized and Localized World” (2000) provides the starting point for this monograph. Having identified the concept, Shah’s presentation emphasizes the importance of results-oriented management and results oriented evaluation in a highly participatory, decentralized governance setting. The current monograph formalizes Shah’s ROME model by analyzing the three concepts individually, focusing on how they combine for effective reform, and discussing important issues pertaining to reform implementation. The monograph is intended as a theoretical as well as practical document, which charts a road to ROME that may not be easy, but is certainly a fresh and effective way of considering public sector reform in the developing world. Beyond this introduction, the monograph comprises two sections: the first explaining why ROME is needed, the second presenting the ROME model and the third discussing ROME implementation.

The first section, explaining the need for ROME, comprises three chapters. The first discusses the need for public sector reform in the developing world, focusing on the need for ‘solutions’ to identified weaknesses in the typical development administration. The following chapter identifies common reform elements and themes introduced throughout the developing world to deal with these weaknesses: macroeconomic stabilization, capacity building, results-oriented management, decentralization and participation. The final chapter of the section argues that these reform elements offer disjointed, potentially ambiguous and incomplete solutions to governance weaknesses, requiring a more complete approach.

The second section presents ROME as a model offering such a complete reform approach. The first chapter of this section presents ROME in concise form. This chapter argues that, by combining participatory decentralization, results-oriented management, and results oriented evaluation, ROME constitutes a model that directly tackles common governance weaknesses in the developing world. The following three chapters discuss the three elements of ROME in more detail. In these chapters various ‘pieces’ of the public sector reform puzzle are contextualized and explained. ‘Pieces’ pertaining to participation and decentralization include, for example, the development of citizen ‘voice’ mechanisms and intergovernmental contracts that focus local entities on citizens.

- 9 -

‘Pieces’ pertaining to results oriented management include management tools like total quality management, performance-based budgeting and the balanced scorecard. ‘Pieces’ pertaining to results oriented evaluations include social and internal, results-oriented audits. ‘Pieces’ are fitted into larger reform elements of the reform, with the emphasis on their contribution to such (rather than their stand-alone value).1

The central principles of ROME-based reform are presented throughout the discussions and form the basis of the opening chapter in a third section, focused on questions of reform implementation and adoption. While identifying the salient principles of ROME individually, the eighth chapter focuses on implementing reforms to facilitate important synergies. The chapter emphasizes implementing results-oriented management reforms within the context of a participatory, decentralized government structure, and bolstered by results-oriented evaluations: start with participation, encourage social evaluations, and then give administrators opportunities to develop results-oriented management abilities. The administrators can then be expected to avail themselves of such opportunities (having incentives to produce the best results possible). The chapter also identifies factors understood to affect reform implementation and requiring attention from countries attempting to adopt ROME-type reforms. Referencing Shah’s (1998) model of the public sector institutional environment, and the broad literature on public sector reform implementation, a model is developed to argue that reformers should consider three important aspects when implementing reform: authority, acceptance and ability. The three factors combine to create the room for effective reform. Key lessons are provided to assist reformers in ensuring the reform space is as large as possible, facilitating a smooth road to ROME—effective and meaningful reform for the developing world.

1 Mindful of the fact that many of these ‘pieces’ emanate from industrialized countries or the private sector (and thus may not be ‘fitted’ to the developing country context), each discussion identifies and focuses on the salient principles of reform rather than the formal tools (that should always be shaped for particular settings).

- 10 -

SECTION ONE: THE NEED FOR ROME

Readers may wonder why it is necessary to develop ‘another’ model for public sector reform in the developing world. This section tackles such question. Drawing from literature, which mostly provides descriptive versions of reform experience in developing countries (rather than empirically valid evaluations of such), this section argues that:

1. Reform is needed to address weaknesses understood to pervade governance structures and processes in developing countries.

2. A string of reform elements have developed in response to such need.3. Reforms that combine such elements are generally shown to fail in addressing

governance weaknesses because individual elements are poorly conceived or implemented, some weaknesses are totally ‘untreated’ by reforms, and reform combinations result in different elements negating the effects of others.

The overall argument is that existing reforms are faulty. A new model is needed because governance weaknesses remain and extant reforms create more confusion than they provide solution.

- 11 -

CHAPTER ONE:THE NEED FOR PUBLIC SECTOR REFORM IN THE DEVELOPING WORLD

Public sector reform is central to development in transitional and lower income countries because, while public organizations dominate society, making it imperative that they function effectively, they generally have poor records of functioning in such a way, seldom providing effective service levels or acting as efficient agents of social development.2 Literature generally characterizes public organizations in the developing world in terms of the following: weak outcomes and low accountability, unresponsiveness, ineffectiveness, and inefficiency (Huther and Shah,1998; Kaufmann et al., 2000).

1.1. Administrative and governance weaknesses in the developing world

Identifying weaknesses in governance and administration in the developing world is not difficult. Casual and formal observers tell similar stories of self-serving governments with little or no organization, and governance structures that have either degenerated completely or are in the process of doing so. Peterson sums the situation up by saying (of Africa specifically): “the catalogue of organizational ills one finds in public bureaucracies…is daunting” (1998, 39). Table 1 lists a number of the ‘ills’ or weaknesses typically shown to prevail in the governance processes and bureaucracies of developing countries.

Table 1. Common governance and administrative weaknesses in developing countriesLow capacity: Low personnel capacity

Low systems capacityOrganizational centralization and top-down governance: Process orientation

Permanence and non-innovationService monopolies: Uncompetitive and unaccountable public production

processes Weak incentives for production efficiency

‘Social insulation’, low transparency and poor participation: Social exclusion based on process and ‘normal

professionalism’ Internally biased, non-responsive incentive structures

Poor organizational evaluation and accountability mechanisms: Weak internal evaluation mechanisms

Few external (social) evaluation mechanisms

2 In discussing capacity development Qualman and Bolger (1996:4) comment on the need for reform in public organizations of the developing world. They explain that reform “frequently targets government because (as noted in a 1994 UNDP report on CD) capable governments are central for development, and "governments are ultimately responsible for creating the framework for development." The onus rests on government to create the 'enabling environment' for organizational and institutional development.” For a discussion of ideas like this, see Turner and Hulme (1997).

- 12 -

1.1.a. Low capacity

The literature commonly relates problems of low public sector capacity. Weak operational capacity is often presented as a reason why governments fail to serve constituents, with abilities often being inconsistent “with the (governance) task at hand” (Shah, 1998: 7). Coston says that “governance problems” in general “may derive from a lack of capacity” (1998: 480). Governments are typically shown to suffer weak personnel and systems abilities, with the former entrenched because of insufficient skills bases in developing countries, low compensation in public organizations, and poor human resource policies resulting in bloated but under-skilled, under-achieving civil services.

Low personnel capacity

A traditional reform line argues that human resources are the key to successful governance and public service delivery (Bell, et al. 1999). Personnel capacity influences service delivery and performance in executing primary functions (Mentz, 1997). Studies from countries like Tanzania (Therkildsen, 2000), Malawi (Adamolekun, et al., 1997) and South Africa (Fölscher,2001) illustrate problems of personnel capacity, with the general consensus that such developing administrations lack effective personnel numbers, qualifications and policies. A particular policy weakness relates to personnel compensation, with literature arguing that governments pay too many employees too little money.3 This kind of problem is argued to underlie public sector maladies ranging from corruption to service inefficiency.

Low systems capacity

Literature also reveals the importance of process and systems capacity in governance and development. Processes and systems order the way in which governance operations are undertaken. Where processes and systems are lacking or are poor they frustrate such operations and yield unstructured administrative activities, which constitute impediments to effective governance. Studies particularly emphasize the importance of systems capacity related to budgeting, financial management and human resource management, arguing that deficient systems in these areas lead to constrained abilities and low accountability in service provision (Hilderbrand and Grindle, 1997; Schick, 1998).

1.1.b. Organizational centralization and ‘top down’ governance

The literature also commonly describes hierarchical, centralized and ‘top down’ administrative entities in the developing world.4 Such structures emphasize control in the governance process, and require role players to adhere to process above all else (as is typical to the Weberian model). This process orientation impacts negatively on governance, undermining democratic responsiveness and focusing public provision on 3 Adequate compensation is a must, and wage compression is to be avoided (Schiavo-Campo, de Tommaso and Mukherjee, 2000)4 Stoker (1997: 175) describes this kind of mindset as one where governance must “start and end with the government agencies.”

- 13 -

internal management issues rather than outward service concerns. In many developing countries the administrative arms are considered totally unresponsive and ‘top down’ structures are seen as devices used for insulating bureaucratic heads from political masters and constituencies (to continue the metaphor, administrative arms wrap around the organization to close it to access, rather than extending from the organization to increase access and reach). An allied problem in centralized systems relates to the common status-quo bias. Most administrations organized according to the classic bureaucratic model emphasize permanence (of process and position) and non-innovation, entrenching production processes and limiting change potential.

Process orientation

Administrative entities in the developing world reflect the traditional Weberian hierarchy, characterized by tight spans of control and specific lines of responsibility. Processes within the model drive interaction, incentives, and production. Organizational participants focus on how they produce rather than on what they produce. This kind of organization was considered suitable for the public sector in the earlier half of the twentieth century and in colonial structures (Gullick and Urwick,1940). It was argued at the time that tight, formal rules of process would limit opportunities for malfeasance and corruption, and enhance accountability. In colonial governments centralization ensured that ‘mother country’ interests prevailed over local interests.

This approach has recently been questioned, with the process orientation criticized for its potentially negative effect on public efficiency and results accountability (Osborne and Gaebler, 1992; Kaul, 1997). Governments are under scrutiny because of their concentration on questions of rule adherence rather than on questions of service and production responsiveness. This criticism is particularly relevant in developing country bureaucracies where formal processes centralizing governance are often poorly developed (as discussed under low capacity) and organizational influences are based on personality and bureaucratic politics. This results in the process orientation frustrating production efficiency and entrenching a politicized version of ‘administration by relationship’ where informal interaction processes drive behavior and governance, leading to administrative organizations that are neither efficient nor accountable.

Permanence and non-innovation

Governments have long been labeled as poor innovators. While there is evidence that this need not be true in all settings, there is also experience showing that strictly hierarchical, centralized organizations (like public bureaucracies in the developing world) are slow to change. These organizations are designed for stability, with production processes and rules of interaction devised to facilitate an efficient and effective production process (unsuited to change) (Peters, 1996: 7-12). Change in such a structure can only enter from the top, through the central organizing authority, minimizing the dynamic potential of street-level bureaucrats whose jobs require (by their very nature) a flexible and adaptive approach to governance and service provision. Such employees often receive high levels of job security within the centralized organization to compensate

- 14 -

for any foregone creativity or discretion (which further negates any pre-existing service orientation in public ‘servants’). Combined with the process-orientation, this kind of organizational characteristic entrenches a status-quo bias that limits innovation and organizational change (including organizational reform).

1.1.c. Service monopolies

Related to the problems of administrative centralization is a third weakness in most developing administrations: the prevalence of service monopolies. This problem is evident in all governments, where monopolies often arise for economically legitimate reasons (the production of a public good, for example). The problem is particularly evident in countries where monopolies emerge for non-economic reasons, however, and governments provide goods that could be better produced in the private sector (where competitive pressures ensure that incentives exist for efficient provision) (Rothenberg Pack, 1987; Savas,1987; Kettl,1993; Tanzi and Schuknecht,2000). Where monopolies produce services, theory suggest that production processes will be uncompetitive and unaccountable, and that incentives for production efficiency will be weak.

Uncompetitive and unaccountable public production processes

Monopolistic entities are common in the public sector, producing services that range from pure public goods like defense to goods usually produced privately, like tourism. These entities are often criticized for producing services using uncompetitive and inefficient production processes. Such processes yield low output levels and high output prices, common problems with public goods and services in the developing world (and problems foreseen by theorists like Musgrave, 1959). These monopolistic tendencies are often seen to manifest in the growth tendencies of public budgets (an argument synthesized by peters, 1996: 24025). Uncompetitive production processes are also generally represented as unaccountable, with well established trade, production and information barriers providing opportunities for corruption and self maximization (Egeberg, 1995). Situations of unconstrained inefficiency (like public service monopolies) facilitate rent seeking because barriers provide buffers to external scrutiny and because corrupt gains can easily be passed off as waste related to the poor production method. These two factors reduce the chance of corruption detection or prosecution.

Weak incentives for production efficiency

Service monopolies have weak incentives to adjust inefficient production processes. Competitors who would benefit from inefficient production by a government agency are not present, for example, and thus government agencies have no incentive to produce efficiently. Combined with the status quo bias, this incentive problem results in entrenched inefficiencies and indeed facilitates adverse incentive structures: inefficient bureaucrats do everything they can to protect their production processes and inefficient project and input choices from facing competitive pressures (from other producers or from comparative statistics). Government cost statistics exhibit this kind of behavior,

- 15 -

often developed to limit the potential for comparison with private sector cost statistics. As public administrators enjoy monopolies over this kind of information, they are unaccountable, unresponsive, and have no incentive to act efficiently (Niskanen, 1971; Moe, 1984; Kraan, 1996).

1.1.d. ‘Social insulation’, low transparency and poor participation

A fourth area of concern relevant to governance in the developing world is the lack of citizen access to governance processes. Largely because of the weaknesses already alluded to, governments are typically portrayed as being insular, non-transparent, and non-participatory (Blair, 1999; Brinkerhoff, 2000). The centralizing nature of public organizations encourages an internal focus and rules of process provide barriers to effective citizen access into governance decision-making processes, for example. Citizens also face great difficulty in accessing information regarding public processes (like budgeting) and in de-mystifying such when access is achieved (Kraan, 1996; DISHA 1999).5 These problems amount o poor participation, which yields most governance processes largely un-democratic (or at best indirectly democratic). Citizens are generally left out of decisions and even production processes, facilitating the development of a public sector culture that rules over, instead of serving, citizens (leading to government as, what Shah (2000) calls ‘the coldest of cold monsters’).

Social exclusion based on process and ‘normal professionalism’

There are a number of reasons why governments in the developing world exclude citizens from the governance arena. Two such reasons are directly related to the nature of the public bureaucracy model under discussion, process and ‘normal professionalism’. The issue of process refers to the fact that administrative entities commonly run according to rules of procedure that govern their interaction and production. These rules are internally developed (formally and informally) within the public administration, and act as a buffer between such organization and the outside world. This is a key element of the Weberian model, where organizational survival requires an effective protective ‘shield’ against external influences that might unbalance the balanced bureaucracy, and is described by Stillman (1991) as “stateless,” destructive to human and democratic values (see Peters, 1996: 50). Essentially, procedural rules create a barrier for social inclusion in the governance process (a simple example is the red tape required when citizens lodge complaints or make applications for services).

‘Normal professionalism’ is an associated problem in public administrative organizations. This relates to situations where professionals charged with administrative duties bias such duties to favor those with whom they have a professional or social affinity (who they see as able to contribute or with whom they have a professional desire to correspond) (Chambers, 1983; Andrews, 2001). This leads to observed experience in developing countries where administrative entities relate to and represent social elites, are more

5 The central importance of information in the development and governance process is most eloquently argued by Freire (1985).

- 16 -

active in urban than in rural areas (where ‘normal professionals’ are not found) and bias ‘participatory’ programs in favor of elite involvement (Pelling,1998; Andrews, 2001). Social participation in the governance process is limited in all such situations.

Internally biased, non-responsive incentive structures

The non-participatory nature of governance structures in developing countries reflects entrenched incentive patterns. Essentially, because governments have traditionally been non-participatory and centralized, it must be accepted that participation threatens established interests and the incentives structuring behavior. Neither political representatives nor administrators have an active incentive to open governance processes to citizen involvement, as this would threaten established lines of relationship, decision-making and influence. Public organizational incentives often center on process conformity rather than service responsiveness, which reinforces these personal interests and further limit the potential for effective participation.

1.1.e. Poor organizational evaluation and accountability mechanisms

A fifth area of administrative weakness in the developing world pertains to organizational evaluation and accountability. Effective evaluation mechanisms form the basis of institutional arrangements and incentives: organizations tend to concentrate on those aspects of their mandate that are actively evaluated (knowing that poor evaluations will result in negative consequences for the organization) (this is one of the central messages of a World Bank Conference on evaluations, Feinstein and Piccioto (2000)).

There are two types of evaluation weakness in the developing world: (1) evaluations are sometimes lacking completely, or (2) evaluation mechanisms are too narrow in focus. Where governments have no evaluation mechanisms, organizations have no incentive to concentrate on their performance and individuals attempting to comment on organizational performance have no information for doing so (Picciotto, 2000). Where evaluation mechanisms are narrowly focused, they create biased incentives and accountability structures (focused on the organizational aspects under investigation), and allow observers a limited perspective of the governance process (Andrews, 2001). Evaluation weaknesses exist with regards to evaluation mechanisms within the government organization (like audits) as well as evaluation mechanisms located outside the government organization (like social evaluations) (Shah, 2000; Feinstein and Puetz, 2000; Kauffman, 2000).

Weak and one-dimensional internal evaluation mechanisms

Most governments (even in the developing world) have some kind of internal evaluation mechanism. This could be located in a public protector’s office, an auditor general’s office, or even in a president’s office. It is usually centralized, and focused almost exclusively on issues of public financial probity (asking questions regarding how much is spent and whether spending follows legal rules and procedures). These kinds of

- 17 -

evaluations tend to carry limited influence in developing countries, and are often open to criticism of political interference (this sentiment is evident in Schick, 1998). Many audit reports are performed on special political request, for example, in situations where they can be openly used to further political agendas. Another criticism of these evaluations relates to their irregularity and inconsistent quality. Although many audit offices in the developing world are building capacity, for example, most still lack the resources to perform audits routinely for all administrative entities (Anyanwu (2001) provides an interesting discussion of capacity shortfalls in Nigeria, while the need for capacity building is a central theme of Feinstein and Picciotto (2000)). Furthermore, evaluations tend to be one-dimensional, concentrating only on fiscal probity and rule adherence. This institutionalizes the centralizing structures discussed earlier, and reinforces process-oriented incentives (at the expense of social responsiveness and efficiency) (Andrews, 2001).

Few external (social) evaluation mechanisms

‘External evaluations’ refer to critical reviews of government processes or performance from outside the government organization. Formal external evaluation sources include bond rating reports (in United States municipalities), and World Bank country evaluations (concentrated mostly on developing countries). Informal external evaluation sources include non-governmental and activist organizations emanating from civil society, and other channels of ‘social auditing’ (Jenkins and Goetz, 1999; Jackson, 2000; Kaufmann, 2000). These mechanisms create lines of accountability between the public sector and pockets of the citizenry. In the case of bond ratings, for example, evaluations provide a link between the business and investment community and governments in which they invest. In the case of social audits, evaluations link communities with governments to whom such communities pay taxes and from whom citizens expect service (Jackson (2000:121) says that, in such cases, citizens interest derive from the fact that they are “probably the most results-oriented stakeholders in any project network”). Most developing countries lack either of these external evaluations types, which results in no institutionalized connection between governance processes and outcomes and the citizenry affected by such. This limits public sector accountability and also undermines basic assumptions of democracy: governments have no incentive to consider citizen demands when they govern, as these citizens have no way of evaluating government performance.

1.2. Chapter conclusion

The idea that governments in the developing world require reform is not a new one. The list of public sector ills presented here is hardly exhaustive, but does present the argument that public organizations in the developing world are dysfunctional, inefficient and ineffective. These characteristics arise because of a mixture of social, organizational and institutional weaknesses that commonly plague governments in non-industrialized settings, causing such to fail even in providing basic services to their citizens, and to

- 18 -

constitute a major stumbling block to democratic and economic reforms. Grindle (1997:481) paints a vivid word picture of the situation in such situations:

“(In some cases) budget austerity and poverty have caused public offices to crumble into disreputable slums and public sector morale to decline to unprecedented levels. In some situations, public officials report to work sporadically to collect even more sporadic paychecks that, in any event, provide family subsistence for only a few days. In still other cases, public officials without meaningful work sit listlessly in offices bereft of pencils, telephones, and light bulbs. Incentives are often perverse; many officials spend their days shuffling papers and inventing ways to obstruct the public’s business. Pervasive corruption and rent-seeking characterize many public sectors around the world, but the crisis of the state is clear when even well-intentioned public officials find it extremely difficult to attend to the public good.”

The work of reform is to provide mechanisms to improve the state of governance in such situations, to redress the governance failures of the past and to treat identified organizational ills. The next section explores reform elements designed with this purpose in mind.

- 19 -

CHAPTER TWO:COMMON REFORM RESPONSES IN THE DEVELOPING WORLD

Many attempts have been made to develop governance interventions (reforms) that deal with the areas of administrative weakness discussed. In this light, Coston (1998) identifies ‘waves’ of reform, which tie closely to the chronological adoption of ideas regarding government improvement. Peters (1996) presents similar ‘models’ of governance in a more theoretical explanation of recent reform thinking. Turner and Hulme (1997) link the start of reform ‘waves’ in the developing world to the 1983 World Bank Development Report which, they argue, sparked a new perspective on the problems of development that laid the focus squarely on public organizations:

“Development failures and disappointments were now seen not simply as the result of inappropriate policy choices but also because state institutions were performing poorly. The public organizations that were encouraged to expand and multiply in earlier years were now perceived as obstacles to development—and expensive ones at that” (Turner and Hulme, 1997: 105-106).

2.1. Elements of reform response to administrative failure

Early reform elements focused on issues of poor capacity and public sector inefficiency, promoting macroeconomic stabilization, administrative downsizing and capacity building. In the 1990s interventions moved increasingly from a macro concentration (involving economy-wide reform) to a micro concentration (involving specific public sector intervention). This move reflected many lines of thought, from new public management to new public administration to new institutionalism. These theories spawned reform elements including results-oriented management, decentralization and participation. Major elements and themes are shown in table 2 and discussed thereafter.

Table 2. Major elements and themes of reform in the developing worldElement Themes

Macroeconomic stabilization and administrative downsizing: Fiscal discipline focus

Privatization Downsizing

Capacity building: Pay restructuring Skills development (usually centralized) Formal process development (usually centralizing)

Results-oriented management: Outputs (or outcomes) focus: customers matter Subsidiarity principle in management (intragovernmental) Government as a business needs business tools

Decentralization: Subsidiarity principle (intergovernmental) Competitive service production Customer service

- 20 -

Participation: Government responsive to citizens (as more than customers) Citizen involvement in decisions (direct democracy) Decentralization as devolution (local accountability)

2.1.a. Macroeconomic stabilization and administrative downsizing

The main reforms introduced into the developing world in the 1980s focused on macroeconomic stabilization and structural adjustment.6 Governments were targeted in these initiatives because of their penchant for over-expenditure and inefficiency. Because of their macro nature as economy-wide reforms, these macroeconomic interventions were focused more on the symptoms of government weaknesses (like over-expenditure) than they were on weaknesses themselves (as presented above). This said, a specific weakness that reformers from the period did emphasize was the prevalence of service monopolies (and the need for reform of such). Reforms in the genre were typically concentrated at the central government level and had three themes, a fiscal discipline focus, privatization and downsizing. All three themes continue to influence reforms in the developing world, especially in countries where development organizations are involved in governance improvement and development initiatives.

Fiscal discipline focus

The first theme of macroeconomic stabilization reform is fiscal discipline. This relates to the need for governments to control the fiscal burden arising from public interventions in society (in keeping with neo-liberal economic theory) (Campos and Pradhan, 1996). Early reforms in the late 1980s involved calls for fiscal austerity (decreased spending and lower fiscal deficits), often failing to provide tools for achieving such (apart from downsizing programs, discussed below). The mid to late 1990s saw greater detail in such policies, with stabilization tools built around elaborate public budgeting reforms designed to centralize spending authority and control (Schick, 1998; World Bank, 1998). Tools like the medium term expenditure framework are commonly used for this control purpose in countries from South Africa to Uganda, for example, while other institutional interventions that centralize budgeting and fiscal allocation and management activities are commonly emphasized in academic and policy literatures reflecting the need for fiscal discipline (Campos and Pradhan,1996; Alesina and Perotti,1996, 1999; Schick,1998; Matheson,1998;World Bank,1998).

Privatization

Privatization is the second theme of stabilization reform. Usually introduced to counter the service monopoly weaknesses of developing country governments and the fiscal burdens associated with state owned enterprises, these initiatives introduced a range of instruments designed to increase competition in service provision

6 In the 1980s development assistance shifted largely from financing investments (such as roads and dams) to promoting policy reform. This change came because of a growing awareness that developing countries were held back more by poor policies than by a lack of finance for investment. (Dollar and Svensson, 1998:2)

- 21 -

(Savas,1987;Kettl,1993). Approaches range from outright privatization (where state owned enterprises are sold to private entities) to partial privatization and service provision partnerships (where government still has partial or complete ownership rights) to outsourcing (where government makes provision arrangements but contracts production to private entities). In the mid-to-late 1990s reforms in this genre continue throughout the world, with the latest version being managed competition, where public service producers compete with private providers in planned competitions operators (Eggers, 1994; Goldsmith, 1997; Burgiel, 1998; Moore and Rose, 1998; Lisk, 1998).

Downsizing

Downsizing is a third theme of stabilization reform, directly related to the previous two. Focused on over-employment and under-effectiveness in governments, this theme involves reducing the size and responsibilities of the public sector (Grindle, 1997: 481; World Bank, 1999). The theme ties in closely with those of fiscal discipline and privatization, with all three themes emphasizing smaller, less burdensome government. In all three themes government emerges as the central problem of development (a message reflected in policy decisions in developing and developed countries at the time). Downsizing reduces the size of the problem. Downsizing initiatives start with analyses of the optimal size of the civil service and the kinds of skill areas required in civil service systems, and moves onto methods of cutting unwanted employees through attrition, special retrenchment packages and such. It continues to be a major issue in governance reforms throughout the world.

2.1.b. Capacity building

Development organizations and host governments typically introduce a second kind of reform, focused on capacity building, in the wake of (or alongside) stabilization initiatives. Capacity building reforms arise because of the general belief that governments in the developing world lack basic abilities required for governance, as discussed in the section on capacity weaknesses in the development administration. These reforms are typically biased to the supply side of capacity (Coston, 1998) and involve interventions to improve the quality of personnel and systems. Such interventions are generally undertaken at or through agencies in central governments and focus on three areas: pay restructuring, skills development and process improvement.

Pay restructuring

A common belief in the development community is that public organizations have poor staffing capacity because they pay their employees too little (Klitgaard, 1997). In the early 1990s this led to a reform drive focused on human resource capacity and the internal organization of the public bureaucracy, with emphasis on “improving salaries and conditions of employment through civil service reform (Grindle, 1997,481; World Bank, 1999). Conditions of service reforms are intended to improve the competitiveness of public organizations in hiring (paying more, governments expect to attract better civil

- 22 -

servants, for example) and to reduce public sector ills like corruption (argued to result from incentives prevalent in poor pay environments). These reforms typically involve improving salaries, better structuring compensation packages, and improving basic working conditions. Most of these policies are driven by central governments (often with the assistance of development organizations).

Skills development (usually centralized)

The supply side bias in capacity building reforms is most evident in the focus on skills development. The dominant position in development is that improved skills will lead to more developed countries and improved governance. Reforms thus focus on developing the ‘skills supply’ in development administrations, especially core competencies related to financial management (Aitken, 1997; Mentz, 1997; Bell, et al. 1999). As with reforms focused on pay restructuring, these capacity building initiatives generally emphasize the central government level, either concentrating only on capacity at this level or developing capacity centrally first with the expectation that central agencies will then develop capacities in deconcentrated or devolved agencies.

The centralizing emphasis in skills development is also intra-governmental, with a focus on building skills in “ managerial, professional and technical fields” of the public organization rather than at the level of street level service provision (Cohen 1993: 11 – 16; Vidlakova, 1993; Adomelekun, et al., 1997; Aitken, 1997; Watson, 2000). This focus is typical in top-down, bureaucratic structures where the management role is paramount. Mentz (1997) calls this the ultimate level of personnel capacity; skilled managers are expected to bring out the best in their staff, developing personnel abilities and facilitating successful reform (Sanders, 1998). Recent skills development initiatives have also focused on developing the operational capacities in entities acting as supports to political representatives (like legislative support agencies) (Hill and Shook, 1998; Peters, 1998; Foltin, 1999; Krafchick, 2001). In countries like South Africa and Tanzania poor political support capacity limits the ability of parliamentary representatives to engage in budgetary dialogue or to impact policies (Therkildsen, 2000: 65; Krafchick, 2001).

Formal process development (usually centralizing)

Recent capacity building initiatives also commonly involve the improvement of organizational processes. This reform theme underlies the emphasis on developing basic bureaucratic structures in developing countries before moving to New Zealand-type results-oriented reform (Schick,1998; World Bank, 1998). Formal bureaucratic processes are argued to form the foundation of public organizational stability, discipline and capacity. Administrations in the developing world are observed to lack these systems, and thus the stability, discipline and capacity to carry out basic governance functions. Reforms rumming according to this theme generally involve introducing processes of financial probity and control, driven by central budgeting or finance agencies. These interventions are intended to increase organizational stability and capacity by formalizing and centralizing processes and operations, thus providing a foundation for future reforms. Many developing countries are involved in implementing these kinds of capacity-

- 23 -

enhancing systems in conjunction with general stabilization initiatives (with a resulting emphasis on organizational centralization).

2.1.c. Results-oriented management

A third reform element arising in the 1990s emphasizes results oriented management in bureaucracies. This develops ideas from a wide variety of theories, particularly the New Public Management (Osborne and Gaebler, 1992; Kaul, 1997; Schick, 1998; Batley, 1999). Reforms in the genre focus on strengthening public sector through the promotion of a results-orientation in management, directly countering weaknesses associated with the centralized and monopolistic character of public organizations (process orientation, permanence and non-innovation, uncompetitive and unaccountable public production processes and weak incentives for production efficiency) (Peters,1996 and 1998). The most ‘fashionable’ reform element (especially in the United States and OECD countries), results-oriented management reforms develop along three themes: output (or outcome) focus, the subsidiarity principle in management, and the belief in generic management (and generic management tools).7 While evident in all levels of government in countries like the United States, these themes have had limited exposure in the developing world (especially at the sub-national level) where results-oriented interventions are typically implemented in conjunction with (or in the shadow of) capacity building reforms.

Outputs (or outcomes) focus: citizens as customers

Results, as measured through outputs and outcomes, are at the center of results-oriented management (Hood and Jackson,1991: 33-34). This emphasis develops from the idea that citizens are customers more interested in the nature of production outputs and outcomes than they are in production processes (Osborne and Gaebler,1992). The results focus is practically embodied in the kinds of administrative reforms being introduced in countries like the United States, New Zealand, Australia, Singapore and Malaysia. Results-oriented budgeting, results-oriented contracting, and service performance employment contracts are common in these governments, introduced to shift the management focus from process to performance. These reforms tend to be driven by central government agencies (or by central agencies in sub-national governments) who administer service contracts and agreements and who are meant to monitor outputs and outcomes to ensure that such agreements hold true.

Subsidiarity principle in management

The subsidiarity principle is a second theme of results-oriented management. The general idea of this principle is that activities in an organization or a government should be decentralized to the lowest level possible—where the role is most effectively played. This theme requires a significant amount of discretion to succeed, with delegated agents, whether intra-governmental (within a level of government) or intergovernmental (at a

7 These are the kinds of themes identified in work by authors like Peters (1996), Desai and Imrie (1998), Christensen and Laegreid (1999) and Hood (2000).

- 24 -

different level of government), given the room to develop their own lines of action (Likierman, 1998). This reform theme has been reflected in a number of ways, among them the separation of roles in the public sector (a re-inspired separation of ‘policy’ and ‘administration’ specialties) (Hood and Jackson,1991: 33-34). The idea is that, given the clear and appropriate service provision role, a government employee or agency can be held accountable for their performance in such role.

Government as a business needs business tools

The third theme of the results-oriented management movement is the belief in generic management—there is very little difference seen between public and private management (Osborne and Gaebler,1992). This theme manifests in the transfer of management ideas from the private to the public sector, with countries around the globe adopting the tools of private business to perform their governing roles (Kaul,1997). The rationale behind this is that if government is like a business, then a government that uses the ‘best practices’ from business is bound to succeed. Such rationale has been criticized in the academic and practitioner literature, with the general argument that government is different to business and that business tools do not necessarily suit the public environment (Batley,1999).8 This said, the dominant trend in public reform around the globe is to adopt private sector-type tools for process improvement.

2.1.d. Decentralization

Intergovernmental decentralization is another prominent element of reform in both the developed and developing worlds. The foundational economic arguments for decentralization center on efficiency gains, increased competition and customer access benefits evident in decentralized governments (Oates,1972). These arguments have been extended to application in the developing world, with an emerging argument that decentralization and development are mutually fuelling forces (Kee,1977; Bahl and Nath, 1986;Wasylenko,1987). The applications in developing countries are numerous, as this reform element is a chosen tool of organizations like the World Bank (Shah,1998, 2000) and the United Nations (United Nations,2000). Key themes of such reforms are the subsidiarity principle, competitive service production and the emphasis on government close to citizen customers.

Subsidiarity principle (intergovernmental)

The decentralization movement is focused on allocating governance responsibilities to the lowest possible level of government. This is the general message of Oates’ Decentralization Theorem, which states that:

“For a public good – the consumption of which is defined over geographical subsets of the total population, and for which the costs of providing each level of

8 The main difference most authors identify is the most basic one: governments lack the bottom line that business defines itself (and its tools) by.

- 25 -

output of the good in each jurisdiction are the same for the central or the respective local government – it will always be more efficient (or at least as efficient) for local governments to provide the Pareto-efficient levels of output for their respective jurisdictions than for the central government to provide any specified and uniform level of output across all jurisdictions” (Oates, 1972, p.35).

As with the subsidiarity principle in results-oriented management (where the issues relate to intra-governmental decentralization), intergovernmental decentralization reform involves allocating appropriate governance roles and functions to different levels of government. The nature of the decentralization differs according to the kinds of responsibilities and discretion local governments have in such roles, however, with some reforms favoring complete devolution of responsibilities (sub-national governments are responsible for all aspects of service provision, for example) and others opting for delegation or deconcentration (where sub-nationals are only responsible for production of services, for example).

Competitive service production

Market oriented theorists tend to favor decentralization as a reform tool because it promises smaller government entities operating in a competitive environment. Tiebout (1956) stimulated research on this aspect of decentralization, arguing that the existence of multiple jurisdictions allows for competition between such which leads to greater efficiency and responsiveness in society as a whole. This aspect of decentralization reform relates directly to the earlier discussion of monopolization: the more decentralized a government is, the less it is expected to monopolize service production.

Customer service

A final aspect of decentralization that is often emphasized in reform I initiatives relates to customer service. Decentralized governments are argued to be less top-down than centralized governments, facilitating greater focus on those receiving services. These people are called ‘customers’ in the popular vernacular. Moving government closer to such customers is intended as a way of making government more responsive and effective in serving such, hence improving the relevance and impact of government in the developing country setting. As Oates says:

“Shifting greater responsibility to local authorities is seen by many as a way to break the ‘grip’ of central planning and mismanagement that has bedeviled efforts to set poorer nations on a course of self-sustaining growth. Both political leaders within the developing countries and advisors from without have sounded the call for decentralization as a mechanism to make policy more responsive to local needs.” (1994: 238).

2.1.e. Participation

- 26 -

‘Participation’ is a buzz-word in the development community, especially linked to the many decentralization initiatives in place. The reform literature provides a strong argument in favor of participation in governance. Dimensions to this argument, include the various perspectives in favor of decentralization, which generally favor governance at the local level because citizens can actively participate in communicating demands and in evaluating and responding to performance at such level (Tiebout,1956; Oates,1972). Political science perspectives are also important, with many commentators referring back to Hirschmann’s (1970) famous concepts of “exit” and “voice” in discussing participation (if communities have effective exit options, and avenues for voicing their demands and impressions of government performance, governments have an incentive to meet such demands). Recent reforms (like PRSPs in the World Bank and more isolated initiatives by organizations like DISHA in India) are often developed with direct reference to these theoretical perspectives, and focus on overcoming the ‘social insulation’ problems of governments in the developing world by making government more responsive to citizens, increasing citizen involvement in governance decisions, and increasing government-citizen accountability.

Government responsive to citizens (as more than customers)

A first emphasis in the participation movement is on government responsiveness to citizens. This emphasis is broadly evident in the literature, embodied in the link between communities and street level bureaucrats (Lipsky,1980) and the new public administration concentration on citizen service and activism on behalf of citizens (Marini,1971; Frederickson,1981). An example of a reform focused on such theme is the Batho Pele civil service reform act in South Africa, which focuses in name and substance on putting ‘people first’. The act “seeks to transform the public administration to be people-customer orientated” by “educating and training civil servants on the new public administration ethos, building people’s awareness so that they can assert their constitutionally guaranteed rights, and building their capacity to demand better service” (Naki,1998). The emphasis on citizens is intended to enforce “democratic notions regarding the obligations of government officials and elected representatives as public servants” (Jenkins and Goetz,1999: 605).

Citizen involvement in decisions (direct democracy)

The literature argues broadly that involved communities foster ‘better’ bureaucracies. The improvement in governance arises as citizens help inform governments about the services they prefer and the best technologies for producing such, citizens hold government accountable for what they produce, and directly involved citizens develop ‘ownership’ over publicly provided infrastructure. A number of literatures speak to the improvements in service provision, including new public management which argues that, when citizens work with street level bureaucrats, service provision is argued to be more responsive and efficient (Osborne and Gaebler,1992; Frederickson,1996; Pierre,1996). The accountability argument relates to a political science argument in favor of democratic governance that extends beyond the voting booth. This perspective is captured by Pimstone’s description of the constitutional commitment in South Africa, which “rejects

- 27 -

the stark and exclusionary notion that democracy is to be equated with an occasional mark on a ballot paper. While formal electoral power is a prerequisite for democracy and indeed was a focus of determined struggle by disenfranchised South Africans, it is in itself insufficient. Democratic governance demands an active engagement between government and the governed…” (1998:132). When citizens are involved in making (or influencing) public decisions, such decisions are bound to be more responsive and accountable.

Decentralization as devolution (local accountability)

A third pivotal aspect of participative reform is its link with decentralization initiatives, and the effect of such link on the nature of decentralization. Local government is undoubtedly the most appropriate level for effective participation, as government officials are close to citizens (in theory and practice). For effective participation in the local government (or any other level) it is necessary that the government be structured to respond downwards (to citizens) rather than upwards (to central government). This is a major problem in many decentralization reforms that structure local governments as satellites of their central creators, forcing a dependency that dilutes local government accountability to its citizens. The participatory emphasis in reform should challenge such structures and reinforce “democratic notions regarding the obligations of government officials and elected representatives as public servants” (Jenkins and Goetz,1999: 605).

2.2. Chapter conclusion

This chapter has presented a number of different elements and themes common to reform strategies in the developing world. The elements and themes mentioned are not definitive, but do reflect the general range of options faced by reformers in developing countries. These reformers, generally in consultation with development partners, have the unenviable job of sifting through the options and choosing reform strategies appropriate for their country situations. Such a job resembles puzzle building, with the reform elements and themes comprising larger puzzle sections into which individual pieces fall: the job of the reformer is to choose pieces such that they ‘fit’ with each other and with the reform context.

The degree of reform ‘fit’ (pieces with pieces, and pieces to context) will determine the effect of the reform on the public organization. A well ‘fitted’ reform is expected to counter problems evident in the public organization, as listed in chapter one. A poorly ‘fitted’ reform is likely to have either no effect on public sector problems or could, in unfortunate circumstances, actually lead to the entrenchment of problematic incentives, processes and organizational ills. Unfortunately, few observers would agree that reforms are often well fitted and successful in redressing ills. This is the starting observation of the following section, which attempts to provide some understanding of why.

- 28 -

CHAPTER THREE:UNDERSTANDING REFORM EXPERIENCE IN THE DEVELOPING WORLD

Reforms generally take the shape of a combination of the elements discussed. It is impossible to definitively say whether such reforms have been successful or not. Indeed it is difficult to define what ‘success’ means in reform, and very few initiatives have undergone objective evaluation. However, balancing the norm-based evidence from reform literature and the stories of interventions in continents like Africa, South America and South-east Asia does provide a picture of reform experience that begs description, and facilitates judgment.

Such judgment is based on the degree to which reforms are generally believed to influence the governance weaknesses they target, theoretically and practically. Viewed through such a lens, there is enough evidence in the literature to suggest that reforms are generally failing to significantly improve governments. Basic evidence comes from case studies as well as general observation of the many interventions over the last twenty years that have failed in brining about real improvements in the public governing process. Why?

3.1. Identifying why reforms ‘fail’ in the developing world

Reform failure can be attributed to three reform problems: (1) individual reform effects are questionable (specific elements are not working as planned), (2) crucial areas of public sector weakness (notably evaluations problems) are totally ‘untreated’ by reforms, and (3) when combined, reforms can interact to negate the effects of individual reform elements. Table 3 indicates how the three problems are argued to limit reform success in the developing world.

Table 3. Reform influences on weaknesses in the developing administrationAdministration weakness Stabilization Capacity Results

oriented management

Decentralization Participation

Low capacity Negative ? ? ? ?Organizational centralization and top-down governance

Negative Negative ? ? ?

Service monopolies ? x ? ? ?‘Social insulation’, low transparency and poor participation

Negative Negative x ? ?

Poor organizational evaluation and accountability mechanisms

x x x x x

In the table, a question mark (?) shows where the effect of an individual reform element is questionable, a cross (x) suggests an unaffected administration weakness, and the word ‘negative’ indicates that the reform element could be having a negative influence on

- 29 -

reforms when combined with other elements. In most situations reform experience reflects a mixture of the three problems at work: the expected influence of a specific reform element may not materialize because of the complicating or obstructing influence of other reform elements, or a reform’s peculiar effect may be thwarted by the fact that a specific area of administrative weakness is untreated and creates general organizational incentives that negate reform altogether.

3.1.a. Individual reform effects are questionable

The list of reform elements in table 2 shows that there are many varying ideas as to how governments should be improved. Each element has specific application, usually associated with particular weaknesses in developing country governments. Crucial questions asked in reform literature center on the degree to which reform elements have been successful in tackling the problems with which they are most directly associated. The associations were discussed in the previous chapter, allowing one to analyze whether the following effects have indeed been apparent:

1. Macroeconomic stabilization reforms were applied with broad objectives in mind, generally focused on making government more competitive and productive (with the downsizing and privatization themes countering monopolization),

2. Capacity-building initiatives were designed to remedy the problem of low capacity and to make governments more competitive,

3. Results-oriented management reforms are meant, in theory, to facilitate the development of a performance-based capacity, a ‘flat’ organizational structure characterized by devolved responsibility for service provision, and public sector ‘competitiveness’,

4. Decentralization was introduced to encourage local-level governance, which is associated, at least in theory, with competitive, bottom-up, participatory governance,

5. Participatory reforms are directly introduced to counter the problem of public sector insularity. Bringing citizens into the governance process is expected to lead to bottom-up governance, greater public sector competitiveness and capacity (through community partnership).

Macroeconomic stabilization reforms

When one considers the effect of macroeconomic stabilization reforms on public sector problems, for example, the literature is particularly equivocal. Observers writing in opposition to the stabilization movement (such as the observers in the Bretton-Woods Project) are particularly negative about the effect of fiscal discipline, privatization and downsizing on governance processes: most such observers argue that these reforms removed local autonomy from the reform process and weakened the ability of individual nations to chart their own reform paths. These voices are strengthened by stories of privatization initiatives that have culminated in public monopolies re-shaping themselves into private monopolies (which serve citizens even worse than their predecessors did).

- 30 -

Beyond such negative reviews of this reform element, one even finds questions regarding this reform’s effectiveness in literature emanating from international development organizations like the World Bank In their analysis of reform initiatives (concentrated beyond the public sector) Dollar and Svensson (1998) find that stabilization programs regularly faltered. They explain such experience by saying that many countries simply weren’t ready for reform. Easterly (2000) found little evidence that adjustment initiatives influenced poverty or growth in most developing countries. He explains such evidence by saying that the poor were not well positioned to take advantage of any gains from these reforms (throwing water on the argument in favor of top-down reform: lower public spending may lead to lower deficits, but does this help the man in the street?). Directly criticizing the public sector application of macroeconomic stabilization, De Merode and Thomas state that “macroeconomic policy reform cannot by itself foster the major changes in work attitudes, ethics, and organizational culture that are needed if significant performance improvements are to be realized” (De Merode and Thomas, 1994; as presented in Grindle, 1997, 481).

There are many cases behind these findings and statements that lead one to question the influence of these reforms on public sector weaknesses. In particular, even though macroeconomic stabilization interventions have indeed resulted in downsizing and privatization programs, it is unclear whether the incentives and structures favoring service monopolization in the public sector have been reformed.

Capacity building reforms

De Merode and Thomas also question the influence of capacity-building programs on public sector development. Speaking directly to the issue of pay policy reform (and other employment adjustment), these observers state that “no conclusive evidence was found of better pay and leaner staffing alone leading to major productivity gains…This suggests that pay and employment reforms, although important determinants of performance, need to be supplemented with other measures.” (De Merode and Thomas, 1994; as presented in Grindle, 1997, 481). This kind of comment is common in reviews of the impact of personnel reforms (focused on creating higher quality civil services) on governance (an example is Nunberg and Nellis, 1995). It is also a fair comment on other capacity and process enhancing reforms, which appear to have had limited effects on the quality of public organizations even though they are ubiquitous throughout the developing world.

A problem with capacity building programs is their one-dimensional concentration on ‘capacity’ as a supply side concept. Authors like Grindle (1996) and Coston (1998) have recently pointed this out, analyzing the ‘capacity’ concept in terms of both demand and supply. According to such perspective public sector capacity is understood to be more than just a question of staffing quantity and skill (or having the right processes): capacity also involves having the right incentives for staff to develop themselves. The message for capacity-focused reformers is that holding capacity-building workshops or providing developing country governments with the latest management tools or processes does not guarantee that lessons will be learned (and applied) or that new tools will be used. Dia (1996) notes, for example, that while capacity-building reforms in sub-Saharan Africa have meant that the region is reasonably well-endowed with educated personnel, good

- 31 -

policies and access to technologies, it still suffers significant ‘capacity’ problems (it still appears that many countries have poor personnel and systems capacities) (see also Qualman and Bolger,1996). Because capacity-building initiatives seldom involve more than the supply-side, workshop-type activity, their potential effect on capacity is highly questionable (where capacity is understood to be a more holistic concept involving demand-side issues as well). Supply-side initiatives ignore the incentives that limit spontaneous skills or process development or the use of existing (or new) skills and processes.

Results-oriented management reforms

The results-orientation is attracting significant interest across the world. This attention is both positive and negative, wt the latter rasing questions about the element’s reform effect. One dimension of critique comes from academic and practitioner observers who decry the lack of consistency and comprehensiveness in results oriented management reforms. Such reforms have been described as a “loose collection of ideas, derived primarily from the private sector, and traveling between countries” (Polidano1998). Desai and Imrie (1998: 645) say explicitly that this kind of ‘loose combination’ leads to a reform element that is “contradictory and flawed.” The ‘fit’ of many such ideas to the developing world is highly questionable, and in many cases it appears as if results-oriented tools have simply been introduced as add-ons to pre-existing process oriented organizational structures. This is even the case in the industrialized world: performance-based budgeting initiatives in the 50 United States states, for example, have been fully introduced in law but have yet to markedly affect budgetary allocations or organizational incentives on any impressive scale.

This kind of evidence suggests that results-oriented management reforms are not having defined effects of improving capacities to produce services, reducing centralization and top-down governance, and transforming service monopolies into competitive producers (at least through the development of competitive incentives). A key problem with implementation appears to be the lack of strategic delegation: where this element is meant to have improved intra-organizational decentralization and delegation, non-industrialized countries generally apply it in a centralized way. Centrally important agencies like finance or treasury tend to dominate performance-based processes, for example, limiting the ability of line agencies to develop critical results-based competencies and capacities, as well as thwarting any move away from the top-down, hierarchical structures criticized earlier in this monograph. This kind of results-based management reinforces organizational incentives to adhere to process (asset by central agencies), and also promotes insular behavior: because goals are set internally, procedures are ‘audited’ internally, and results are evaluated internally, governments have incentives to internalize their operations.

Decentralization reforms

Decentralization reforms are intended to have broad effects in public organizations, redressing problems of low capacity, centralization and top down governance, and non-participation and insularity. Evidence suggests that these reforms seldom have such

- 32 -

effects. This has resulted in numerous observers questioning the structure of decentralization initiatives (Prud’homme, 1995; Litvak, Ahmad and Bird, 1998; Shah, 1998). Studies by Shah (1998, 2000) and Huther and Shah (1998) find that decentralized governments do appear to have positive effects on public governance,9 but that the structure of decentralization bares significant influence on such effect. Devolved governments (where decision-making responsibilities are decentralized) tend to be more effective than deconcentrated governments (where minimal decision-making authority is devolved) (Shah, 1991, 1994).

It appears that many decentralization initiatives involve deconcentration rather than devolution, characterized by delegation of production responsibilities to local entities only (instead of total decentralization of decision-making authority pertaining to which services are produced, in what quantities, and by whom). A major rationale for such centralized decentralization is the apparent low levels of capacity at the local level: surely central governments cannot decentralize responsibilities to entities without the capacity to meet such? This kind of argument has its flip side: limited decentralization has meant that local level governments exhibit notoriously low levels of capacity (Stewart and Greenwood, 1988). Provincial governments or municipalities with tight mandates from central governments have no reason or incentive to create personnel or process abilities related to decision-making, prioritization, planning and allocation (this is evident in cases of South African municipalities, where central influence over municipal entities retards spontaneous local level capacity-building incentives).10 Given this argument, although in most cases decentralized governments can indeed be seen to have poor capacity, centralized decentralization exacerbates the problem.

Tight accountability lines between central and sub-national governments also have an effect of entrenching top-down governance (a problem discussed earlier). Organizational hierarchies are maintained, simply running over formal intergovernmental boundaries, with sub-nationals responding to the demands of higher-level governments (rather than their own citizens). In most instances the nature of central-sub-national interaction is through top-down controls, which bind provincial and local level entities to specific processes—limiting any chance of citizen-based, results-oriented governance. This plays out in a further question regarding the influence of decentralization reforms on participation and social insularity. The literature shows that centralization may increase the participation of people at the local level, but because of the top-down, centrally dominated structure of most decentralization initiatives it is only a small privileged elite who get to participate (Oates, 1994; Pelling, 1998). Participation is often regulated on the basis of national political affiliation in such instances (with those affiliated with nationally powerful parties also achieving standing at the local level).11 Disempowered

9 Huther and Shah (1998) find that 38% of variance in governance quality is explained by decentralization alone.10 South African studies include Foundation for Contemporary Research (1999), Oranje et al (1999), Otzen et al. (1999), Planact (1999) and The Planning Initiative (1999). 11 An example of this is Pelling’s discussion of decentralization in Guyana (1998). In this case a case study of local government shows the dominance of national political party members in the governance process. Disempowered segments of the society were not empowered in the decentralized government, and it maintained its non-responsive, insular character.

- 33 -

citizens under a top-down centralized system tend to remain disempowered under a deconcentrated system.

Participatory reform

The final area in which reform effects are questioned is participation. Participatory reforms are directly intended to enhance social involvement in governance processes and to open traditionally insular government organizations. Participation is the most recent piece in the reform puzzle, but unfortunately there are already question marks over its impact on public sector problems. There are many unsolved questions when it comes to involving communities in the administration process (Brinkerhoff and Kulibaba, 1996; Turner and Hulme,1997; Schneider,1999;Blair, 2000; Brinkerhoff,2000). Even experience in the developed world yields the generic example of limited success with participation programs, whether one looks at England (Watt et al.,2000; Woodward,2000) or the United States (Tauxe,1995; MacManus,1996), for example. Most examples of participatory reform in the developing world, from Nigeria (Wunsch and Olowu,1996), to India (Dhesi,2000), to Bolivia (Blair,1997 and 2000), and Brazil (Atkinson et al,2000), tell a tale in which citizens largely remain disempowered and excluded from the governance process after interventions take place.

Tauxe’s case study of local level participation in America shows one of the main problems of participatory reform: reforms tend to be taken up as centrally driven programs within insular structures, rather than bottoms-up initiatives aimed at opening such. In many developing countries this kind of outcome arises because administrative decision-making (and participatory reform decision-making) is centralized, administrative processes are complex, and information is unavailable (the process is closed). In such situations, technical administrators enjoy a powerful influence over the nature of participation, facilitating access to narrowly defined groups only. Such participatory reform designs are also prevalent in situations where political processes are un-democratic, centralized, or unrepresentative. In such situations disadvantaged citizens are disenfranchised in the governance process and enjoy little influence over the kinds of participatory processes adopted.

The opposite is found where social capital is strong. Strong social organizations, especially when horizontal in nature, empower citizens (especially disadvantaged citizens) and the administrators working with these citizens (street-level administrators) (Pelling,1998). Such organizational forms are typically representative, broadly accountable, and socially active. They mobilize citizens to act in common interest, pooling resources and minimizing transaction costs associated with social expression and involvement. These social structures often act as conduits for natural participation and ensure that participatory reform programs are structured by participants (and not just government authorities). Unfortunately most participatory reform programs do not consider existing forms of social capital as building blocks for reform, and few reform initiatives look to developing necessary social capital as part of participatory reforms. When these shortcomings are noted, and one understands further that socio-economic conditions often make participation overly costly for participants (especially the disadvantaged and disempowered), one can understand why participation programs

- 34 -

seldom yield high levels of participation and generally fail to open traditionaly closed governance structures to public scrutiny.

3.1.b. Evaluation weaknesses are ‘untreated’ by reforms

Apart from the problem of questionable reform effects on targeted governance weaknesses, reforms can also be faulted for paying insufficient attention to the evaluation problems of developing governments. Interventions to develop evaluations capacities in developing countries are relatively new and have yet to be reflected in main-line reform elements (Feinstein and Picciotto,2000). Evaluations are important for a number of reasons, chief being that they shape and enforce behavioral incentives in social settings. It is important to promote evaluations that facilitate institutional and incentive changes necessary for effective reform. Without such evaluations, incentives do not change from those associated with traditionally flawed public organizations, and behavior cannot be expected to change either.

Examples of the evaluations problem abound. Dia (1996) talks of African countries with the personnel, skills and process abilities to govern well, yet with records of poor governance (perhaps because of poor governance evaluations?). Andrews (2001) speaks of inconsistencies in results-oriented reforms related to the lack of attention given to changing the nature of process-based audit evaluations facing ‘results-oriented’ managers. Andrews comments that, “In such situations, managers are receiving a conflicting message: “Manage for results…but remember that you will be audited on your adherence to process (not on the results)” (Andrews, 2001: 10). In essence, the point is that an ‘untreated’ public sector shortcoming as important as this acts as a negative to any reform positives introduced. Reforms will be prone to failure if they fail to incorporate evaluations that introduce incentives conducive to public sector reform.

Apart from changing the nature of evaluations in public organizations, reformers should also be aware of the need for evaluations of reforms themselves. If one believes that “what gets measured gets done” (Picciotto, 2000:361; Andrews, 2001: 10), one should also understand that failing to measure (or evaluate) reform implementation and adoption almost guarantees that reforms will not be ‘done’. Critical questions reformers should be asking of the pubic entities targeted for change include (Andrews, 2001):

Evaluating reform mechanics:Has the entity adopted a strategic plan to implement reform?Has the entity ‘shaped’ reform ideas to its specific situation?Has the entity implemented any new reform mechanisms, and does it have a plan for evaluating such implementation?

Evaluating reform adoption (and inbuilt accountability mechanisms):Are reform mechanisms being used?

- 35 -

Does the entity communicate with customers (citizens or others) on a regular basis about the reform? How does it communicate? What does it do with the information received from customers regarding reform implementation and value?

Evaluating the influence of the reform on the organization’s culture:Is the reform considered important in the entity? Does the internal management have any alternative reform ideas and if so how are these ideas dealt with (are they actively considered or not)?

There is very little evidence that reformers ask such questions as part of their reform initiatives. This means, essentially, that reforms often go unevaluated. Intuitively, this poor reform evaluation can be seen to relate directly to the problem of low reform adoption and success.

3.1.c. Reform interaction can be negative

A third identifiable problem with public sector reform in the developing world is the negative results that arise out of element combinations. While Qualman and Bolger (1996:2) are correct in observing that “one-dimension (reform) approaches (such as human resource development, and traditional institutional development) have not, over the decades, met expectations for sustainable results,” reform approaches combining the five identified elements (and others) have also failed to meet expectations. This is because of tensions existing between reform elements:

Macroeconomic stabilization themes tend to relate poorly with all other elements. Fiscal austerity, downsizing and privatization are claimed in many circles to reduce public sector capacity (and morale), focus public organizations on process and discipline (rather than results), and strengthen the roles of centralizing entities within governments that make decisions in a top-down, insular fashion. These characteristics clash with reforms focused on capacity building, results-oriented management, decentralization and participation.

Apart from conflicting conceptually with macroeconomic stabilization themes, capacity building initiatives are often implemented in such a way as to negate the influence of other reforms as well. Capacity-building initiatives are invariably introduced (with the assistance of development organizations) through central government agencies, and aim to consolidate the very abilities and processes that centralize and insulate bureaucracies. Other reforms, like results-oriented reform and decentralization are often implemented as second-fiddle to capacity-building initiatives (and thus take on the same centralizing character) or are implemented after capacity building has been completed (which, especially at the local level, is seldom).

A further tension exists between results-oriented reforms, as introduced in the developing world, and participatory reforms (Peters, 1998; Ingraham, 1998; Rockman, 1998; Christensen and Laegredi, 1999). Managerial reforms in the

- 36 -

developing world often take the form of top-down decision-making and goal identification, the introduction of tools that further insulate administrators from society, and a perspective that the public are customers, rather than citizens (Shah, 2000). These often dominating aspects of results-oriented management result in any participatory initiatives being downplayed. Shah identifies this problem on a global level and terms it “Consumer Sovereignty and Democracy Deficit” (Shah, 2000:4).

The problem, in short, is that when elements are combined, their interaction can have negative effects on governance processes and outcomes. Their combinations lack positive, problem solving synergy. Instead of such synergy, reform combinations in the developing world commonly reinforce problems associated with the public sector problems of process orientation and centralization.

In combination, reforms are commonly ‘process’ biased

Figure 1 shows Shah’s public sector results oriented management and evaluations chain (Shah, 2000). The chain presents different phases of public production, from program and project identification through social impact. Earlier discussion held that most governments in the developing world, structured in the mould of the Weberian bureaucracy, are overly focused on process and ‘hard controls’ emphasizing input management (Shah, 2000: 16). Citizens are understood to have an interest in government production in the chain, concentrating on reach (who is served?), outcomes (what are the products received?), and impact (how do the products develop the under-developed state?). The disjoint between what government organizations focus on (inputs) and what citizens look for from governments (outcomes and impacts) is evident in the figure, with one focal point close to the extreme left and the other close to the extreme right.

Figure 1. The process bias of common reform combinationsProgram/ project Inputs Activities Outputs Reach Outcomes Impact

Instead of introducing mechanisms and changing incentives such that governments shift their attention from inputs on the left of the chain to outcomes on the right, common practice is for reforms to concentrate (at least initially) on consolidating hard input controls. This is achieved through the dominance of macroeconomic stabilization reforms (reflected in most fiscal and economic policies at national level in developing countries) and capacity-building initiatives (typically run from the center). Reforms tend to concentrate on introducing controls and ensuring probity and central capacity before they move into areas of results management, decentralized and devolved service provision, and participation. In those instances where reforms have progressed to include results-oriented management, the focus is still limited (to outputs) and governments are still not given any motivation to consider their more important social effects. This is evident in the comment by Desai and Imrie (1998:645) that new managerialism is “characterized by de-democratising tendencies and a fixation with procedural and technical processes.”

- 37 -

There are two explanations why reform combinations focus on process rather than social outcomes and impact:

1. Reforms are sometimes viewed as cumulative, with more complicated elements like results-oriented management, participation and decentralization building on other elements—most importantly capacity building. Capacity building initiatives usually dominate the organizational reforms in developing countries (with macroeconomic stabilization issues prominent in the macro-policy arena). Poor capacity is considered the most serious impediment to good governance and is countered through initiatives usually focused on administrators: teaching administrators how to plan and manage resources and providing administrative entities with necessary processes. The concentration of these programs is explicitly at the left hand side of the results chain, setting the focus for reforms that follow.

2. A recent argument, which could be called the bureaucratic stage argument, holds that administrative entities need to pass through specific stages in order to mature. The argument, offered by authors like Schick(1998) is that bureaucracies have to learn about the importance of hard controls (focused on inputs) before they can successfully implement soft controls( focused on results). The argument is used to legitimate interventions in the developing world that continue to concentrate on the introduction of basic budgeting and civil service controls. The idea is that, once governance processes are strengthened, reforms can re-focus on other sections of the results and evaluations chain, steadily moving from left to right. The argument makes two critical assumptions: (1) that reforms will ever be successful in establishing such controls, and (2) that administrators conditioned to focus on process will be able to shift their view to results and performance.

In combination, reforms are commonly centralizing and top down

Reforms also commonly combine in a centralizing fashion. Figure 2 shows participants in the results and evaluation chain, and reforms at different stages in such. The figure reflects the central emphasis on and influence of administrators and executive office holders in the reform process. The most important reform participants in these categories tend to play centralized roles in governance structures, and often govern in a top-down fashion with legislatures and citizens left out of crucial decision-making and implementation processes. Reforms thus reinforce the centralization, top-down problem earlier identified as an ‘ill’ of governance systems in the developing world.

The view that reform combinations centralize governance processes arises from the observation that dominant reforms (related to macroeconomic stabilization and capacity building) are almost wholly devised and run by central government and central agencies within central government. Even the reform elements designed to redress the centralized character of governance processes are driven centrally: hence results-oriented programs, decentralization initiatives and participation programs are generally shaped and manipulated by central agencies. Andrews (2001) argues, for example, that participation

- 38 -

programs often fail because they are conceived by high-level government agents with no knowledge of participation needs and no interests in ensuring that participation programs actually work. Citizens and their direct representatives in legislatures are often either co-opted (to support) centrally devised programs, invited to participate in reform development on a selective basis (which is usually non-threatening to centralized interests) or kept out of the reform design stages completely. The result in all three cases is that top-down reform influences of dominate bottoms-up influences of citizens and their representatives.

Figure 2. The top-down, centralizing nature of common reform combinationsProgram/ project Inputs Activities Outputs Reach Outcomes Impact

ADMINISTRATION and EXECUTIVE (assisted by ‘development’ partners):* Craft macroeconomic reforms, requiring policy decisions about where to cut spending, how to cut spending, etc.* Introduce capacity building in internal processes and organization, focusing on strengthening internal abilities (and increasing their own importance in the governance process).* Incorporate results-oriented management tools into extant programs, alongside capacity-building initiatives (once again focusing tools on their internal operations and goals). * Fashion decentralization reform to suit internal interests, with process-oriented ties facilitating control over deconcentrated ‘agents’.* Promote limited participatory programs involving politically ‘passable’ mechanisms based on principles of ‘normal professionalism’ and ‘controlled participation’.

STRONG TOP-DOWN REFORM INFLUENCES

WEAK BOTTOMS-UP REFORM INFLUENCES

LEGISLATURE and CITIZENS:* Have very little influence over macroeconomic stabilization issues.* Are seldom consulted about capacity building.* Have no say on results management.* Are either encouraged to support limited decentralization (legislature) or have little say in decentralized structures, which are answerable to higher level government authorities.* Are uninvolved in determining what participation programs look like. Participation programs shaped by other reform interests.

There are two potential explanations of the centralizing effect in reform combinations:

1. All reform elements and themes fall into a reform hierarchy, which yields some elements more important than others. The nature of the reform hierarchy is dependent on a number of factors, including the normative values ascribed to different reform elements, the ‘reform incentives’ created by external reform participants (like international organizations) and economic ‘partners’ (such as the international business sector often expected to facilitate development via foreign capital infusions into under-developed economies), and the chronological order of reform introduction.

- 39 -

In most developing countries these factors yield macroeconomic stabilization the most important reform element, a fact borne out in the fact that other elements generally take their cue from such (deficit targets shape results-oriented reforms in most countries, for example). This reform element is usually charted, implemented and driven by central agencies (usually as high up as the prime minister’s office of the national treasury. All other reforms, like results-oriented reform, decentralization and participation, are nested in this reform element. This results in the reforms that counter centralization theoretically being ‘hidden’ within reforms that reinforce centralization.

2. Another explanation of reform centralization, given by Shah (2000) relates to the direct influence external reform ‘partners’ have on the kinds of reform adopted and the location (within governance structures) of reform initiatives. Shah (2000:33) argues that external reform partners favor centralizing reform structures and initiatives because “a centralized hierarchical system lowers transactions costs for external assistance and enlarges the comfort zone for external participants in terms of monitoring the utilization of their funds for intended purposes.” These preferences are often manifest in the lending-based incentives international organizations create for reforming governments, encouraging such to adopt top-down, control-oriented reforms.

3.2. Chapter conclusion

Not all public sector reforms fail, but experience of failure certainly outweighs experience of success. This chapter has put forward a number of reasons why this is so. The chapter argues that individual reform elements often fail to affect governance problems in ways expected, either because the reforms are poorly conceived or implemented. Reforms also fail because, while many public sector ills are challenged by common elements, few reforms act to counter the problem of absent evaluations in governments. The evaluation problem manifests in weak reform incentives, which could be argued to negate any reform initiatives. Finally, reform combinations tend to re-enforce process and centralization-biases that pervade the public sector in non-industrialized countries. These biases are considered ‘problems’ requiring reform: in reinforcing such, reforms can potentially worsen governance weaknesses (fostering greater insularity and poor representation and accountability).

A new approach to reform is needed, that matches reform elements to the specific problems facing countries in the developing world, that tackles issues of evaluation, and that ensures reform combinations act to counter (rather than reinforce) public sector problems. The next section introduces such an approach: Results Oriented Management and Evaluation (ROME).

- 40 -

SECTION TWO: PRESENTING ROME

The introduction presented ROME as a different model to those preceding it, not so much because of the kinds of elements in the model, but rather because of the way the model is organized and because of the elements emphasized. ROME focuses on creating the right institutional environment for results oriented reform. This entails focusing on developing participatory, localized structures through which citizens are empowered to demand better results from government.

With this kind of institutional environment in place, it is observed that governments have the incentives necessary to adopt results oriented institutions (rules and tools) themselves. Thus, these rules and tools should not be the focus of reform (as they often are) but rather an available resource provided to enquiring managers.

ROME comprises three sections: 1. Participatory decentralization2. Results oriented management, and3. Results oriented evaluation.

In this section these three elements are discussed, as well as their implementation and factors that influence the institutional environment into which they are adopted.

- 41 -

CHAPTER FOUR:ROME: RESPONDING TO REFORM NEED

When one builds puzzles, one needs to ensure that pieces are not improperly chosen, left out or incorrectly combined. When these puzzle-building errors occur, the result is an incomplete or distorted picture. And so it often is with public sector reform. Choosing incorrect reform elements, leaving elements out, and combining elements in contradictory ways has resulted in general reform failure throughout the developing world. The problems of governance in non-industrialized countries remain thorns in the side of visionary leaders who would see their nations develop and mature but lack the governance structures to do so. Even with the many reform options at their disposal, such leaders find themselves in need of new avenues for intervention. ROME constitutes such an avenue, strategically combining prominent new public management, new public administration and new institutionalist reform perspectives into a single approach designed to facilitate real improvement in governance processes and outcomes.

ROME, which stands for Results Oriented Management and Evaluation, is a reform model that combines three elements from common reforms, results-oriented management, decentralization and participation, with a new reform element, results-oriented evaluation. Through the selection of these elements as well as the synergy between them, the ROME model directly challenges the top-down governance models entrenched in the developing world. ROME’s bottoms-up, results-oriented dimensions focus public entities on outcomes and impacts rather than inputs and process, and center attention on citizens as the final principals and customers of public entities. In so doing, ROME tackles the problems and weaknesses that plague governments in developing countries, providing both the focus and incentives necessary for real governance improvement as well as the tools required to respond to such.

4.1. ROME: Making sense of the reform puzzle

ROME arises out of lessons learned from past reform failures. The main differences between ROME and other common reform approaches are:

1. The central focus on results, 2. The inclusion of an evaluations component, and 3. The overall arrangement of reform around a participatory-decentralization

element.

The model emphasizes bottoms-up reform involving citizen participation in local government. This focus arises because citizen involvement is considered the key to public sector results-oriented reform in developing countries. In ROME, citizen participation forms the basis of all government decisions (with public entities responding to citizen demands), the framework for government accountability (with citizens evaluating what government does and ‘rewarding’ them for such), the central motivating factor for civil servants and politicians alike (with citizen ‘evaluations’ driving their

- 42 -

behavior), and the foundation of government capacity (with governments drawing from their constituency to achieve the capacity levels needed to meet demand).

ROME’s focus on participation and decentralization develops from both observation and theory. The observation is simply that governance problems in the developing world generally relate to excessive centralization, and ‘centralizing’, top-down reforms are not seen to visibly improve the organization, service performance or accountability of public entities.12 The theory is varied and includes the following: Public economic theories generally posit that governments should focus on citizens.

Probably the most prominent of these is the median voter theorem, heavily used by economists in research on service provision, which assumes that governments respond to citizen preferences when they provide services.

Broader economic and political science theories also emphasize the importance of citizen participation in governance processes. Consider Hirschman’s (1970) theories about ‘voice’ and ‘exit’ mechanisms, as they relate to other work by authors like Tiebout (1956). These theoretical perspectives generally argue that, when citizen participation in government processes is enabled (through voice or exit mechanisms, the latter of which facilitates a special type of negative participation), governments are more efficient, responsive and accountable.

Work in the evaluations literature increasingly emphasizes the importance of citizen involvement in governance. In his discussion of evaluations and corruption, Kaufmann (2000:292) argues that “the evidence is now clear that participation is the key to development,” a sentiment echoed by Jackson (2000). Sartorius (2000: 133-4) identifies the potential benefits of involving citizens in evaluations processes, which generally reflect the information gains achieved by participatory evaluations, the improved capacity, more representative (and meaningful) goal identification potential, and greater social buy-in to governance decisions resulting from evaluations (which are considered legitimate).

The Public Administration and political science literature presents participation as “one of the dominant…themes of the 1990s” (Peters, 1996: 47). Various approaches to participation exist in this literature, all calling for greater citizen involvement in decision-making and governance. Putnam (1990) finds that where citizens bound together and participate naturally (where social capital is high) governments work better, for example.

The prominent voice of social conscious, as expressed in the context of the struggle for development and growth in under-developed regions of the world, also increasingly emphasizes participation. Expressing the importance of citizen involvement in their government and society, and of social capital, one of these voices, Bishop Desmond Tutu says that, “To be…is to participate” (Krog, 1998:110).

Effective participation that actively results in citizen empowerment is difficult to achieve in large centralized governments. Thus the participatory concentration in ROME is conceptualized at the local (or provincial) level: governments at this level are small

12 Peters (1996:48) argues that participatory approaches in governance reform emerge because of the belief that “the hierarchical, rule-based organizations usually encountered in the public sector (are) severe impediments to effective management and governance.”

- 43 -

enough to facilitate citizen involvement. In decentralized governments citizens should have fairly close access to political representatives and administrators alike, not just street level bureaucrats. Living amongst constituents is assumed to be a most effective way of encouraging political and administrative officials to take such constituents seriously: citizens have a direct and natural line of influence over officials who live alongside them.

Figure 3 shows the ROME model concisely. The direction of governance and reform influence is obvious: contrary to figure 2’s representation of common top-down reforms, ROME is bottoms-up. Citizens, in the bottom right corner of the figure, are the central role player in the governance and governance reform process (with the participatory decentralization element captured by the clear block). ROME sees these citizens interacting directly with legislatures, comprising chosen representatives (note the explicit importance of some form of democratic process to the model). At the local level such communication goes beyond election processes (which are too irregular to guarantee effective participation), rather taking the form of implicit social and explicit political ‘contracts’ between citizens and their representatives. These contracts are built on the social pressure citizens can exert on public servants (who live in their midst) as well as creative political and economic pressures that can be institutionalized through reforms. Such contracts center on the provision, by government, of specific results—outputs, outcomes, reach and impacts—in society.

- 44 -

It is assumed that a shift to participatory decentralization will lead to a natural results focus in the governance process (with the participatory decentralization element located at the right hand side of the results oriented management and evaluations chain in figure 3). The rationale is simply that citizens are more interested in government outcomes, reach and impacts than they are in controls over inputs (citizens want to know what government does for them, not how they do it).

The figure shows that the participatory decentralization element overlaps with the second and third reform elements, results-oriented management and results-oriented evaluations. These graphical overlaps emphasize the fact that all reform elements revolve around the participatory decentralization component. It is citizen involvement in decentralized structures that creates the focus and provides the pressures that make the other elements both important and viable. The participatory decentralization concentration holds ROME together, with citizen demand providing the basis for a results-orientation and comprising the source for results identification and evaluation.

Results-oriented management arises in ROME because of citizen demand in the participatory, decentralized government. Because citizens demand results from their political representatives, those politicians charged with running the administration (the executive) are pressurized to deliver. In developing countries such a pressure is problematic, however, as administrative processes are focused on inputs rather than results (as discussed in preceding chapters). Results-oriented management (ROM) reform enters into ROME to solve this problem. With the executive carrying the results-based demands of citizens (identified through participation mechanisms facilitating citizen voice), it uses ROM interventions to focus and capacitate the administration to achieve such. These interventions take a number of forms, including:1. Devices used to effectively decipher citizen demands, 2. Mechanisms creating ‘output’ contracts between the executive and administration,

and 3. Tools used by administrators to transform their management processes from an input

to an output and outcomes orientation.

The first two forms are considered the most important in ROME. Devices to decipher citizen demands are the core instruments administrators use to develop their results focus, while ‘output’ contracts create the incentives for administrators to respond to demand. The management tools that help administrators in achieving results, like planning tools and accounting tools, are of secondary importance and should only be offered as options for administrators to appropriate as they see fit. The rationale behind this is that if administrators can identify the results they have to achieve, and if they have incentives to achieve such, they will develop the necessary tools and processes themselves (or take advantage of capacity-building opportunities presented to them). The flip-side of this rationale is that if results-oriented tools are introduced, and capacity developed, without the results identification and incentive enhancing mechanisms, administrators will not use the newly acquired tools to serve citizens (because they have no reason to do so).

- 45 -

ROME’s final element is results-oriented evaluation, which plays an important role in strengthening results-oriented incentives facing politicians and administrators in the governance process. The figure shows that this element is intertwined with participatory decentralization (as the two overlap graphically, with the arrow labeled (1) connecting citizen evaluations to outcome contracts). The element also relates directly to results oriented management (with two arrows labeled (2) and (3) connecting internal and citizen evaluations to output contracts). This third element simply involves the evaluation of results. Such evaluation is important for the ongoing analysis of results-oriented contracts, between citizens and political leaders (where the concentration is on outcomes) and between citizens and political leaders and administrators (where the concentration is output). These evaluations constitute important accountability and transparency devices, which help to inform and enforce important behavior-binding and shaping influences in the ROME system: 1. Citizen evaluations of outcomes are necessary in ensuring that citizens have the

ability to assess public sector performance in terms of negotiated ‘outcome contracts’. Political leaders are only expected to take such contracts seriously if citizens, either directly or represented by groups in civil society, actively and regularly force them to do so (by evaluating performance and holding political leaders accountable for such).

2. Political evaluations of outputs are similarly important to bolster the effect ‘outputs contracts’ have on administrators. Administrators are only expected to take such contracts seriously if political principals force them to do so (by evaluating performance and holding administrators accountable for such).

3. Citizen evaluations of outputs are necessary to strengthen the results-orientation of administrators. While the political-administrative output contract sets formal incentives for administrative behavior, citizen interaction with administrators is an often overlooked but powerful informal influence on administrative performance. If citizens actively evaluate outputs as well as outcomes, they can consistently engage with administrators (who live in their midst) about results, creating incentives for administrators to work hard at maximizing their performance.

The three types of results-oriented evaluations thus reinforce the citizen-focused results orientation in ROME. They complete the reform puzzle by consolidating new incentives in the public organization that not only support a new view of governance but also help to facilitate reform itself: the more regular, focused and participatory results-oriented evaluations, the greater the incentive to make change work and produce results.

4.2. ROME: completing the puzzle, tackling the problems

The ROME model is argued to tackle problems facing governments in the developing world. These problems were introduced in chapter two. Chapter four provided a number of reasons why reforms often prove ineffective in solving such problems. These reasons related to the poor ‘fit’ of reform elements to situations (either conceptually or in implementation), the absence of any evaluation concentration in common reforms and the adverse effects accompanying reform combinations.

- 46 -

The ROME idea looks to avoid such reform shortcomings by focusing on reform elements that have intuitive appeal in the developing country context, by incorporating an evaluations element in reform, and by organizing reform elements so that the holistic reform effect counters public sector problems. The ROME effect is shown in table 4, as the individual elements relate to individual governance weaknesses.

Table 4. ROME influences on weaknesses in the developing administrationBureaucratic

weaknessParticipatory

DecentralizationResults oriented management Results oriented

evaluationLow capacity

Participation = a source of capacity.

8. Administrators have the incentive to develop necessary capacity for

ROM, and take advantage of opportunities to develop new tools,

as well as developing capacities within their communities (participation in action)…

Organizational centralization and top-down governance

1. Citizen participation counters these weaknesses directly as governments are bottoms-up and focused on the demands of their citizens, who act as final principals and customers…

7. Administrative entities are less centralized, being accountable for results but independent to produce such…

‘Social insulation’ and non-transparent

management

10. Results oriented evaluation increases transparency

Monopolization of service provision

6. The results focus, and adoption of ROM devices, leads to more ‘competitive’ and efficient governments, who change processes as required to improve results…(their incentives have changed!)

Permanence and non-innovation

Process orientation

2. Citizens are interested in results, not process, countering the process orientation…

5. In developing new ‘hearing’ mechanisms, contractual devices and management tools, executive and administrative role players automatically develop a results-orientation…

- 47 -

Internally biased, non-responsive incentive structure

3. Focused on citizens, who are interested in results, local-level politicians and administrators have an external, responsive incentive structure…

4. Politicians and administrators have an incentive to develop mechanisms and tools that maximize results…

Results-oriented evaluations strengthen ROM incentives

Poor organizational evaluation and accountability mechanisms

Citizens, through their participation, provide regular evaluations of governance performance…

9. Citizen/internal results oriented evaluations provide constant mechanisms for informing results ‘contracts’

The figure shows the basic argument as to how ROME counters administrative and governance weaknesses in the developing world. In the first instance participatory decentralization directly counters weaknesses related to top-down, centralized and insular governance structures. This effect is achieved by focusing government on citizens (usually at the bottom of the governance hierarchy) and decentralizing public sector structures (‘1’ in the table). This reform element is also expected to automatically stimulate changes in public sector organizational orientation and incentives (‘2’ and ‘3’ in the table). Citizens are assumed to be interested in results rather than process, and as they express their preferences so it is expected that public organizations will divert their attention to outputs and outcomes, having explicit incentives to do so (the better their results, the greater their citizen support).

Results-oriented management enters the equation to bolster the new incentives and results orientation (‘4’ and ‘5’ in the table). Political representatives, influencing administrators through output-based contracts, create incentives for these administrators to focus on results rather than process. Beyond these incentive mechanisms, administrators also develop new tools for evaluating citizen demands and meeting such. These tools facilitate the change to a results orientation and lead to government agencies being more creative, efficient and competitive in their provision processes (‘6’ in the table). New ROM incentives and tools also lead to decreased centralization as government entities are held accountable for results but given significant discretion over how such are produced (‘7’ in the table). This discretion leaves capacity building in the hands of actual service producers (‘8’ in the table). Administrators have an incentive to improve their capacity for service provision, can access ROM tools, and also lean on their participating communities for necessary skills and processes. Once again, the elements work together to counter weaknesses of centralization and poor capacity.

The final weakness, usually untreated in reforms, relates to the poor evaluations and monitoring devices in public sector entities. This weakness is addressed directly through the results-oriented evaluations reform element (‘9’ in the table). As results are evaluated, so governments are held accountable for performance. These accountability mechanisms bolster incentives created by results-oriented management tools and are themselves facilitated by the high degree of citizen participation in ROME. Citizens too

- 48 -

conduct evaluations, which also increase transparency in the governance process (‘10’ in the table).

4.3. Chapter conclusion

The point of table 4 is two-fold:1. ROME’s individual elements blanket public sector weaknesses (including evaluations

weaknesses), and 2. When they interact around participatory decentralization, the three elements interact

in a way that reinforces their individual effects.

In essence, ROME meets the bill in answering the reform problem stated at the end of chapter three:

A new approach to reform is needed, that matches reform elements to the specific problems facing countries in the developing world, that tackles issues of evaluation, and that ensures reform combinations act to counter (rather than reinforce) public sector problems.

The reform model has been presented in a simplified fashion in this chapter, with the aim of showing the general logic of the approach. The simplified presentation certainly belies the complexity of each element and of the interaction between elements. The next three chapters address the details of all three reform elements, participatory decentralization, results oriented management and results oriented evaluation, in an effort to provide more information about what each involves. These discussions aim to locate common tools and reform ideas into the ROME model, as they relate to the greater elements.

- 49 -

CHAPTER FIVE: PARTICIPATORY-DECENTRALIZATION IN THE ROME MODEL

ROME stands for results oriented management and evaluation, correctly communicating the reform’s central focus on results. This said, the results focus is not the starting point for ROME-based reform. Any results-orientation first requires a results-identification: all governments have to answer ‘which results matter?’ before they manage for or evaluate such. The question of results identification is not trivial, and the identity of those answering it has a great influence on the character (and effect) of results-oriented reform. While common results-oriented reforms have this question answered by technical experts in central governments, the ROME model begins with citizens providing the answers. The first element of ROME, participatory decentralization, provides the foundation for such citizen involvement.

Focusing at the ‘bottom’ of the governance process, the participatory decentralization element forms ROME’s foundation, and constitutes a radical change to traditional governance models and even reform designs so often espousing top-down decision-making and reform. The two core components of the element are participation and decentralization:

Participation in the ROME model differs from that experienced in many reforms worldwide, however: where common reforms introduce participatory programs with limited benefit to citizens, participation in ROME requires programs that empower citizens (particularly those traditionally disempowered). ROME’s vision is for a “locally or grass-roots focused democratic strategy” that will “place utmost priority on shifting power from the center to the people.” It emphasizes “the need to construct institutions close to and accessible by the people, through which they” can “take collective choices, and structure and sustain collective action developing over time from these choices” (Wunsch and Olowu, 1996/97: 67-68).

The decentralization concept in the ROME model also differs from that commonly implemented: instead of deconcentration in which local entities are merely agents of central government, decentralization in ROME requires devolution (or near-devolution) in which local entities are responsive to citizens (not higher level governments). ROME’s vision is for democratic local governance that is built on popular participation and facilitates local governments being more responsive to local citizen desires and more effective in service delivery.13

Combining forms of participation that genuinely empower citizens with locally-responsive, decentralized government structures facilitates an active citizen influence on governance and government reform. This combination not only gives citizens a voice, but also channels through which to express such, and entities who have incentives to listen, and are close enough to hear what citizens say. The result is a government focused on citizens as their final principals (to whom entities are accountable) and customers (to

13 This sentiment is reflected in a number of recent publications, including Blair (2000).

- 50 -

whom entities should be responsive in service provision). With such concentration, ROME-based reforms are founded on the idea that the results that matter are the results that matter to citizens—their needs and their demands. It is these results demands that inform the public sector bottom line—citizen service.

This idea is the core of reform in countries like New Zealand, Australia, the United Kingdom, and Malaysia (Chiu, 1997; Kaul,1997). It is also assumed in much public economic theory, where governments are expected to respond to the wishes (or demands) of their constituency (represented, for example, by the median voter). In developed countries citizen demands have been shown to impact on what government does and how it does it (MacManus,1996; Hayes,2000). Many observers even cite citizen demands as the central reason why countries like New Zealand, the United States, and the United Kingdom have shifted to a results-orientation (Lam,1997; Kaul,1997). The ‘bible’ of such reform in the United States, Osborne and Gaebler’s ‘Reinventing Government’ introduced the idea as a need for “community-owned government” (Osborne and Gaebler, 1992: 48).

In developing countries, however, citizens seldom play a large role in the governance process, either through democratic systems (like the electoral process) or through financial means (such as tax influences) or other forms of participation. Closed governments in these settings fail to facilitate a results-orientation. This is obvious when one considers the number of governments trying to implement results-oriented tools like performance-based budgeting without success. The primary reason for failure, it is argued, is the lack of results demand (and evaluation) from a citizenry that is disengaged from its own governance process, and an organizational structure that discourages responsiveness to citizens.

ROME provides a different model to the one that traditionally fails, emphasizing participation and decentralization prior to the adoption of results-oriented tools. The process of participatory decentralization, of engaging citizens in governance and governance reform, involves at least five kinds of reform activity, each of which is discussed in detail: Providing citizens with a voice, creating demand for results, and Decentralizing governments, positioning government to hear and respond to voice.

5.1. Providing citizens with a voice, creating demand for results

Citizen ‘voice’ is a term used to relate to citizen demand (Hirschman, 1970). Such voice usually calls for governments to produce results. Consider an example regarding education provision in figure 4, the results oriented management and evaluation chain. The figure shows that educational provision entails identifying programs and organizing inputs and activities to produce outputs for certain groups. These groups then reap the ‘outcomes’ of the process, which yield larger social impacts.

- 51 -

Figure 4.The results oriented management and evaluation chain (an educational example)Primary school education program

Budgeted funds, number of teachers, classrooms, etc.

School administration, teaching activities, etc.

Enrollment levels in specific schools, pass rates.

Identities of those receiving education.

Quality of graduating students.

Student employment rates, student graduation to higher education or other productivity gains.

Program/ project

Inputs Activities Outputs Reach Outcomes Impacts (goals)

Traditional process orientation Results orientationWhat citizens are interested in: what they ‘voice’ their

interests about

The shaded portion of the figure identifies aspects of the chain usually emphasized by governments, the traditional process orientation. These are not the aspects citizens are interested in, however. Citizens are more interested in how many students pass through schools, who these students are, how well the schooling they receive prepares them for life challenges and how such schooling enables them to contribute to society. If citizens are enabled to express their preferences for services, or ‘voice’ their demand, governments will automatically be focused on the results dimensions of the management and evaluation chain. Citizen demand forms the basis of knowing which results matter, and creates the incentives for public organizations to focus on such.

When citizens are not able to ‘voice’ their demands, governments become overly concerned with supply because they lack outward-oriented goals. When citizens are able to ‘voice’ their demand, government is given a clear goal (producing results that satisfy citizens), potential partners in reaching such a goal (the citizens themselves), and automatic evaluators of their goal-directed performance (the citizens, especially direct customers). As such, citizen voice is the basic element required for governments attempting to graduate from a process orientation to a results orientation in the developing world. Results-oriented reform should always start with developing mechanisms that facilitate and ‘amplify’ such voice in the governance process. As stated by Grindle, “organizational performance in developing countries can be improved through the creation of client demand” (1997: 486).

Evidence would suggest that ‘client demand’ is seldom effectively voiced in the developing world. Literature presents many diverse arguments why this is so. These are distilled into two categories that form the basis for ‘voice enhancing’ intervention in the ROME model: weak access channels for ‘voice’ into governance processes and poor social organization for demand expression. ROME solutions to these problems involve applications of participatory and decentralization reform elements: (1) organizing ‘voice’ channels into the governance process, and (2) building on, fostering, and informing social organization. The logic of these solutions is shown in figure 5, where the combination of channels facilitating demand expression and social organizations ‘amplifying’ such leads to active results demand from society and the development of a results orientation in

- 52 -

government. Adding information about governance processes and production outcomes into the system allows more effective demand.

Figure 5. Social organization, information, and the demand for government resultsDemand for results from

society

Channels of demand expression +

social organizations ‘amplifying’ citizen voice

Public sector develops a results orientation

Information about government processes and

government outcomes

5.1.a. Organizing ‘voice’ channels into the governance process

Theoretical perspectives on the governance and administrative process call for channels citizens can use to interact with public officials. These perspectives include the new public administration, which emphasizes the links between citizens and administrators, and the new public management, which argues for a variety of participation paths that facilitate community owned government (Osborne and Gaebler, 1992). In leading results-oriented governments like New Zealand and Malaysia, citizens enjoy significant access to government through such channels. This access facilitates citizens’ ability to ‘voice’ their demands to governments.

Evidence suggests that political systems institutionalize “opportunity structures that can facilitate or hamper collective action” (Mohan and Stokke, 2000:260). Such systems tend to hamper collective action in the developing world. Traditionally, government structures in developing countries are closed to citizens, reflecting the top-down, centralized, insular structures discussed in earlier chapters. Such organizational structures are more apt to facilitate top-down attempts “to repress or attempt to destroy” (Coston, 1998: 486) community, rather than release the ‘voice’ of community into the governance process. For society to influence government, it must be provided ‘voice’ channels. In some instances informal channels are effective, and demand for results can be well communicated through such. For governments genuinely attempting to embrace a results-orientation, however, more formal and legitimate avenues of influence should be created. These provide channels for civil society voice into government as well as the necessary channels for government response—with the result being communications channels facilitating effective social voice and public provision. Two forms of channel development are the basics of the ROME reform model:

1. Localized democratic structures in which political representatives are chosen by and accountable to local citizens.

- 53 -

2. Specific, regular channels of ‘voice’ expression (apart from the prevailing electoral system).

1. Political representation basics: democracy and decentralization

The democratic election process is the most basic form of citizen voice in the governance process. While the importance of such has been questioned in studies testing the relationship between democracy and economic growth, in the present governance context the value of democratic political structures is beyond doubt: “The record of industrialized countries shows that democratic participation is the only form of government with a consistent record in ensuring good governance” (Shah, 2000: 15).

Localized democracy is seen in many corners of the literature to increase civil society participation in and influence over the governance process (for example, Ostrom and Whitaker, 1999). These are central aims of ROME. An important clarification is the need for genuine representation, however, in which democracy at the local level is equivalent to governance decision making at the local level. In such situations civil society is given a clear and meaningful political voice and influence over the governance process, reflecting the essence of what Blair calls ‘effective participation’: “The central idea of participation is to give citizens a meaningful role in local government decisions that affect them, while accountability means that people will be able to hold local government responsible for how it is affecting them” (Blair, 2000:22). This kind of democracy has also been called “radical democracy” (Mohan and Stoke, 2000) and “participatory democracy” (Osborne and Gaebler, 1992). Creating such representative, radical and participatory democratic structures is as close to a first step for ROME-based reform as any.14 Top-down structures associated with other political forms are likely to be associated with political opportunity structures that hinder collective action, a situation antithetical to ROME.

Creating new political and governing systems requires “a change in attitude and behavior of the people in these institutions” or what Chambers calls ‘reversals’ in the behavior of organizational ‘uppers’ (Schneider, 1999: 530; Chambers, 1997). Specific ‘reversals’ are required in the following ‘design’ areas (all of which emerge as problems in local governance development). An illustrating example of the importance of such ‘reversals’ is Wunsch and Olowu’s case study of Nigerian local government (1996/97), where many of the following problems have not been addressed and local democracies remain limited:

Representation location: Too many developing countries concentrate on developing democratic structures at the national level only. This is also the level at which other governance and administrative reforms are usually undertaken. Where local representation exists, representatives are often appointed through national party lines or through the influence of central government leaders. Examples include Pelling’s case of local government in Guyana (1998) and Breslin and Campbell’s work on

14 Taylor (2000:1030) emphasizes the importance of democracy to effective participation: “While participation may form an important basis for citizenship, it is perhaps most valuable in the context of a strong and representative democracy—where accountable representatives have the authority to evaluate needs, balance demands, establish priorities and monitor the outcomes or the political system.”

- 54 -

reforms in Pérez Zeledón, Costa Rica (1999). ROME argues that democracies should be localized, facilitating effective local voice into political processes (see Ostrom and Whitaker, 1999). Brinkerhoff (2000:610) supports such a perspective in his comment that “the literature and experience converge strongly on the desirability of expanding beyond intervening at the central level.” Bottoms-up reform implies the local election of national representatives as well as active political representation at the local level, not local level appointments by high-level government (whether such are explicit or implicit, they still thwart local-level accountability). Organizational ‘reversals’ thus require that national governments alter their democracy building initiatives to concentrate on the local level.

Representative authority: In developing countries where local representatives do exist (either through election or appointment), these representatives usually enjoy very little authority. Local-level representatives are delegated implementers of higher-level decisions, rather than independent, locally responsive decision-makers (consider Pelling’s case of Guyana (1998)). Bottoms-up, ROME –type reform requires a locally authorized polity, however. If political processes are to facilitate the channeling of citizen voice, political representatives holding office closest to citizens must have the authority to respond to such. If such authority is lacking, it is expected that citizens will not use the channel (as they will soon become aware of its low value to them). This is indeed the case in Guyana where citizens have purposefully disassociated themselves from formal channels of representation (Pelling, 1998).

Representativeness: An important critique of decentralized (and centralized governments) in the developing world is the lack of representation for disempowered minorities. Political systems often bias representation against such minorities, either through the structure of districts (creative gerrymandering) or through the type of representation introduced (direct representation versus proportional representation for example) or through the abuses allowed within the system (what Blair calls “guns, goons, and gold” in the Philippines, for example (2000: 27)). ROME requires that all communities have an opportunity to voice their demands, through processes like elections, such that civic pressures in the political process reflect all demand profiles and social accountability structures involve all communities.

Representative accountability: The United States is widely considered a successful democracy, but even in this country citizens question the accountability and allegiance of political officials. Osborne and Gaebler (1992: 73) state that by 1989 “three out of four American surveyed agreed that “most members of Congress care more about special interests than they care about people like you”.” ROME emphasizes local political representation under the assumption that such system leads to greater responsiveness and accountability of political representatives to local constituents. Political systems need to be purposively designed with such in mind, ensuring that elected representatives have the incentive to be accountable and responsive to their constituents.

- 55 -

The attention given to these issues by organizational ‘uppers’ such as national level political leaders, dominant political parties and even international development organizations reflects the degree to which those promoting reform buy into the ROME emphasis on bottoms-up change. Examples of such attention, as identified in relation to the issues mentioned, are:

Representation location: Various developing countries follow the example of industrialized nations like Switzerland in legislating an important role for locally elected governments. The South African Constitution, for example, reflects the belief that municipal structures should play a central role in the substance of governance, not just its ceremony (Pimstone, 2000). Legislation plays a necessary role in facilitating a move to local-level democracy, with the Bolivian case being instructive. Local governance was not effective in the 1980s and 1990s, resulting in the passing of a Popular Participation Law, which emphasized the importance of local entities, and consolidated the de facto move to democratically elected, representative local government (Schneider,1999). Building local democracies requires going beyond legislation, however, and ensuring localities have necessary political capacities to facilitate effective and competitive democratic processes. South Africa is an example of democratic capacity building at the local level, working in conjunction with legislation facilitating the political shift from the center. In provinces like Kwa-Zulu Natal religious groups and other NGOs (like the ecumenical Diakonia organization in Durban) are actively working with Scandinavian donors to educate and inform local constituents of the functioning of their democracy.

Representative authority: Intergovernmental structures are usually characterized by competition between different government levels, over functions provided at different levels, resources accessed by different governments, and influence over key administrative areas (like procurement and personnel management). The essence of locating authority locally lies in the legal and institutional arrangements resolving such competition. Shah (1998) argues conclusively that effective decentralized entities only develop when the following are clarified through such institutions: Expenditure and functional assignment—Who does what? How much and what

kinds of expenditure are individual governments given autonomy over? Revenue assignments—Who pays for what? Who raises what revenue and if it is

to be shared, how is this done? If revenues are shared, what ensures that local entities have authority over the money they receive?

Issues of cooperation—What institutions are required to ensure that decentralization is conducive to national goals such as macroeconomic stability but also accommodate a high level of local responsiveness?

Issues of dispute resolution—What institutions and mechanisms are available to facilitate the resolution of intergovernmental authority complications?

Elander (1997: 152) reflects on the need for clarity in local level authority by saying that, “To be able to exert some degree of autonomy local government needs a minimum of resources in terms of constitutional basis, political legitimacy, professional competence, and money.” These issues are best resolved directly in law,

- 56 -

as in most industrialized federal countries where the rights and functions of local entities are protected by supreme forms of legislation (like the United States Constitution). Recent legal reforms in countries like Costa Rica reflect effective adjustments in this regard, giving local entities greater authority over shared revenues collected nationally (Breslin and Campbell, 1999).

Representativeness: In response to low representation of minorities and low income groups, many countries are looking to legislate representation ‘quotas’ at the local level. While this creates a modified form of democracy, it ensures that these groups are represented and that their voice is heard. An example is the Bolivian Popular Participation Law, which requires representation of minorities in local governments (Schneider, 2000). Another example comes from Costa Rica, where legislation requires that various groups in local communities are actively involved in allocation decisions taken by local councils (Breslin and Campbell,1999). Once again this legislation is a necessary aspect of ensuring that local democracy is representative, but it is hardly sufficient. Democratic capacity building is required to consolidate the effect of legislation, with minority constituent groups informed about how they should participate and assured of their protection from hostility. This is especially important in the many developing countries where political processes are dominated by groups that use intimidation to silence other voices.

Representative accountability: There are a number of ways in which political processes can be structured to ensure that representatives are accountable to citizens. These devices are usually grounded in legislation, which informs political representatives of their ethical and substantive responsibilities while in office and of the limits to their political role. These form the basis of a core ROME reform element, the political-citizen outcomes contract. Such contract mirrors that in place in many New Public Management initiatives between politicians and administrators as well as the ethical contract embodied in the Athenian Oath:

The Athenian OathWe will never bring disgrace on this our City by an act of dishonesty or cowardice.We will fight for the ideals and Sacred Things of the City both alone and with many.We will revere and obey the City's laws, and will do our best to incite a like reverence and respect in those above us who are prone to annul them or set them at naught.We will strive increasingly to quicken the public's sense of civic duty.Thus in all these ways we will transmit this City, not only not less,but greater and more beautiful than it was transmitted to us.

The Athenian Oath combines an ethical commitment as well as a substantive commitment. ROME concentrates on the latter element, arguing that political representatives should be bound by a commitment to producing particular results (as outcomes) that they will produce in transmitting government “not only not less, but greater and more beautiful” than when they received it. This contract idea is obviously not new and is already implicitly embodied in election campaign platforms, where politicians make commitments to constituents regarding the kind of outcomes

- 57 -

they will concentrate on producing. ROME advocates institutionalizing such platforms, as the Athenian Oath was institutionalized in Athens, by making legislators formalize their outcomes promises to citizens. Such commitments should only be open to change at the end of a political term or during a term when constituents agree through direct democracy ballot box initiatives (Osborne and Gaebler,1992). Direct democracy institutions like referenda and initiatives have been shown to influence fiscal outcomes and governance results at the local level in the United States (Besley and Case,1995, Kiewiet and Szakaly,1996)and Switzerland (Feld and Kirchgassner, 1997, 1999). These democratic tools are used to keep politicians accountable for their mandates. Given that such mandates are based on outcomes, or results, such tools ensure effective citizen voice and influence into and over the political process and make it difficult for political representatives to ignore such voice.

Building localized democratic structures is complex and demanding. It is seldom seen as the central element in general public sector and economic reform. The ROME model accords it this central position. Central governments, donor agencies and citizens (especially through NGOs and civic organizations) are all important roleplayers in democracy building at the local level. Central governments play an important part in making laws and in facilitating a shift to decentralized democracy and representation (central governments need to relinquish authority, power and influence). Central governments are also the main actor involved in developing structures required for effective democracy (like independent judiciaries and organizations needed to mediate between levels of government).

International donor agencies are vital in setting the tone for reform. Different agencies have different histories regarding who they work with (which level of government) and which elements of reform they emphasize. ROME’s emphasis requires that these organizations focus on sub-national governments and civic organizations, long the practice of Nordic-based organizations (Brinkerhoff,2000). Working with these entities as well as central governments to build social voice through representative structures provides “ balanced approach” to ROME, that emphasizes the importance of getting the appropriate partners in place for effective reform, and of creating the appropriate incentives for the kind of reform prescribed. Local NGOs and civic organizations are similarly important, as reform watchdogs and democracy builders. As the Inter-American Development bank concludes, this kind of reform requires that “governments…their citizens, and donors must work together in an integrated, coordinated effort to change the incentive structures and preventive measures in society…to create economically stronger, politically accountable democracies for the future” (IADB, 2000).

Legislate regular channels for social ‘voice’ into the governance process

While democracy provides an important channel of social voice into the governance process, there are problems with relying only on elections processes for this purpose. The problems are reflected in Blair’s critique of elections: “elections are crude instruments of popular control, since they occur at widely spaced intervals…and address only the broadest issues” (Blair,2000: 27). Blair’s sentiments capture the problem of

- 58 -

facilitating genuine and regular civic communication into the governance process: he writes that, “People must be able to express their likes and dislikes between elections, as well as their views on specific proposals.” This is a second dimension basic to ROME’s focus on creating voice channels, the development of regular arrangements for demand expression between voting cycles. These arrangements are vital in ensuring that political-citizen contracts are well monitored. Arrangements for such are generally lacking in developing countries, but do come in a number of forms, for example:

Public meetings: Many governments are finding that public meetings are an excellent way of engaging citizen voice on a regular basis. In Ukraine, for example, a number of mayors have launched public budget hearings, which are also common at the local level in the Philippines. In both countries regular council meetings are also open to the public (Blair, 2000). A detailed description of how these meetings often work comes from the Inter-American Bank’s work in Paraguay:

“Before the hearings, community-based organizations are encouraged to meet and discuss their priorities. They present their concerns at the hearing where local officials describe available resources and current commitments. The communities’ priorities are integrated into the yearly budget to the degree possible, and all participants specify how they will help bring the plans to fruition. The subsequent year, the government holds a public hearing to review the implementation of the budget and to explain any discrepancies” (IADB, 2000).

In a number of countries and regions, including Karnataka in India, Honduras, the United Kindgom and South Africa, national level legislation or policy can require local-level governments to hold meetings such as these (Africa, 1999; Blair,2000; Woodward, 2000). In other countries national or regional governments (like Mali) go beyond legislating meetings to actually convene them (as moderators and facilitators). Meetings could also be planned and facilitated at the local level by municipal entities (as in the Dakotas and in Florida (MacManus, 1996; Tauxe,1995). Civic organizations also play an important role in developing these meetings in some countries, like India (where MKSS convenes public meetings in Rajasthan (Jenkins and Goetz,1999)).

Opinion surveys: Opinion surveys and ‘report cards’ are a second voice mechanism facilitating regular communication. These are generally used at the local level, having gained popularity in the 1990s. The Bangalore Public Affairs Centre’s Report Cards are the most commonly discussed. Questions are levied through the centre, of citizens in Bangalore, regarding satisfaction with service provision decisions and activities. The aims include: “To generate citizen feedback on the degree of satisfaction with the quality of services provided…; To catalyse citizen groups to a pro-active stance by demanding more accountability, accessibility and responsiveness from public service providers…; To encourage public agencies to adopt client friendly practices and policies, design performance standards and facilitate increased transparency in operations” (Gopakumar, 1997:281). Some countries also levy similarly focused surveys at the national level (used to connect citizens with national

- 59 -

politicians). Examples are the USAID Governance and Local Democracy Project in Philippines, which includes a survey of citizen satisfaction with government services and Uganda, where diagnostic surveys are seen to “provide vital information for decisionmakers when institutional weaknesses inhibit a more regular flow of information” (Reinikka, 1999: 1). Gopakumar (1997: 282) links such report cards with the results-oriented approach governance by saying, “There could be no better way to gauge performance than the ‘voice’ provided by the end user.”

Formal grievance procedures: A traditional approach to engaging citizens involves creating formal procedures for grievances. These processes are often a part of established processes in public bureaucracies, and are potentially as limiting as other bureaucratic processes. This said, new grievance procedures are proving important in results-oriented reforms around the world. O’Roark (2001) finds that performance-based budgeting at the state level in the United States is more effective in influencing expenditure when citizens have formal lines of grievance and communication (these could be as simple as dedicated telephone response lines). Other examples come from Latin America, where hotlines are used to engage citizens regarding public sector performance and corruption (IADB, 2000) and Hong Kong, where citizens have established procedures for responding to poor performance and to corruption (Efficiency Unit, 2000). The Malaysian Public Complaints Bureau (PCB) is a formal entity designed to provide a channel for citizen ‘voice’, and has been seen to improve “responsiveness to the problems faced by the public in dealings with public sector agencies” (Schneider, 1999:529).

Public accountability officers: Connected to formal grievance procedures, public accountability officers can play an important role in facilitating citizen voice. In Bolivia, Territorial Base Organizations provide such officers, with a mandate to “monitor the provision of public services to committees and to make suggestions which reflect local needs.” Vigilance Committees in the OTBs have a particularly accountability-enhancing role “to check the equitable allocation of resources in urban and rural areas” (Schneider, 1999: 526). In South America generally, these kinds of officers have proved important institutional mechanisms for regular participation. “Key institutions include: (1) Controller General to investigate the misuse of public funds; (2) Ombudsman to represent the concerns of the population to the government” (IADB, 2000:2). The controller general has a multi-faceted evaluations and monitoring role (related again later in this paper). Latin America Offices of the Human Rights Ombudsman (Defensores del Pueblo) are focused singly on the influence of governance on human rights, but ombudsmen in other countries (like Costa Rica and El Salvador) monitor and scrutinize the administration of public services and functions. The influence of these officers is visible in Uruguay, where the ombudsmen function has “strengthened existing formal avenues for citizens’ input and monitoring” (IADB, 2000: 2).

Formal committee access: MacManus (1996) argues that the formal committee access is the most traditional method of soliciting regular citizen responses to capital budget (and other financial) decisions. Governments create committees and invite

- 60 -

citizens to sit on committees, giving civic representatives a voice into allocations decisions. Increased participation through committees is argued to lead to greater buy-in and support, and tailoring of government activities (Miller,1984). Citizens are engaged through advertisements for committee openings as well as extensions into community-based organizations (DeSantis, Pollock and Moore, 1993). An example from the developing world is the Village Development and Security Committee (JKKK) in Malaysia (Schneider, 1999: 529). This committee is described as the “authoritative body for the representation of the village communities in voicing their opinions for project identification and implementation, as well as identification of the poor and hard core poor.”

Schneider (1999: 530) describes these more regular channels of citizen voice as “windows” or “dedicated bodies” through which citizens can access governments. Schneider emphasizes the importance of such channels for disempowered and poor groups in developing countries, who can access government for “targeted service provision, tailored to their needs, or to have administrative activity examined and corrected where they feel they have suffered prejudice.” These arrangements facilitate the creation of incentives for ongoing public response to citizen voice. This is evidenced through the work of MKSS in Rajasthan where public meetings called jun sunwais, at which citizens have a chance of questioning government officials about allocations and activities, have had a dynamic effect of motivating government to produce results that satisfy constituents (Jenkins and Goetz,1999). The incentives to respond largely arise because access mechanisms create regular checks and balances in governance (increasing the probability that low responding officials will be identified and publicly ostracized). These checks only emerge if participatory mechanisms actively empower citizens, however. In examples where this is observed to be the case, two lessons are learned:

Be specific in policy and legislation: Authors like Tauxe (1995) and MacManus (1996) explain that administrators often shape participatory programs to serve their purposes. Andrews (2001) shows, with examples, that this results in low levels of empowerment. In South Africa, for example, legislation calls for the establishment of general ‘participatory mechanisms’, which most observations show tend to be weak and unaccountable. To counter such weakness, policy and legislation needs to specify requirements of participation mechanisms. This means mentioning, for example, how many meetings should be held, where they should be held, when they should be held directly in policy or legislation. Furthermore, legislation should explicitly require responses to citizen communications as such responses instill confidence and trust.

Evaluate participation: If legislation is specific, then governments must have a way of evaluating such specifics. While evaluation will be discussed in a later chapter, it bares relevance at this stage given the importance of evaluation in reform itself. Governments with effective regular participation have effective means of evaluating participatory mechanisms. In Hillsborough County in Florida, United States, for example, the government collects statistics on meeting attendance, meeting length, and meeting engagements (who spoke about what) to allow evaluation of the quantity

- 61 -

and quality of participation. This gives a clear perspective on whose vision is embodied in the minutes and facilitates mailing of a meeting summary to each attendee, and posting of such summary in public places (MacManus, 1996). This kind of active feedback keeps officials accountable for the participatory programs they promote.

The responsibility for creating regular and effective access channels lies with national and local governments, development organizations, and civic structures. National governments often provide necessary capacity (with surveys, for example) and have an important legislative influence on participation. An example of legislated intervention is the 1994 Popular Participation Law in Bolivia, which provides a legal basis for the creation and constitution of village-based ‘vigilance committees’ who provide formal accountability officers (Schneider,1999; Blair,2000). The Development community’s role is multi-faceted. Donor agencies can enhance the quality of regular participation pathways by financing their development or by requiring that fund allocations decisions be made through regular participatory structures. Valenzuela (1999) provides an example from El Salvador where money from the Inter-American Foundation set up a development fund that formed the focus for regular participation. Donor agencies can also contribute evaluations technologies and abilities, given the importance of such in the arrangements discussed (Bendahmane, 1990). Local civic involvement is also important, with examples showing that civic groups ensured the momentum of participation was maintained. In countries from Honduras to Karnataka to South Africa it appears that the quality of civic involvement is directly related to the effectiveness of regular voice mechanisms.

5.1.b. Fostering social organization (for effective ‘voice’)

‘Voice’ channels are only effective if citizens actually express their demand through them. A strong argument exists that citizens have greater ability to express such demand when they do so in groups. Social organization, in which groups are developed, takes a number of forms. One such form is the interest group, which represent the demands of similar-minded citizens regarding specific issues, pressuring governments at all levels to give such issues attention. A second form of organization emerges in societies where citizens gather regularly to interact, for purposes of relaxation or religion or even ongoing business.

Communities in which citizens gather in this way are often seen to have high degrees of ‘social capital’: “trust, norms, and networks, that can improve the efficiency of society by facilitating coordinated actions” (Putnam, 1993: 167). Specific actions that are facilitated by high degrees of social capital (the result of social organization) include the expression of social demand into the governance process. Putnam’s study of social organization in Italy found that governments were more responsive and effective in areas where citizens participated in social structures. The benefits of civic behavior were seen to impact governance processes on both demand and supply sides:

“On the demand side, citizens in civic communities expect better governments and (in part through their own efforts), they get it. They demand more effective public

- 62 -

service, and they are prepared to act collectively to achieve their shared goals. Their counterparts in less civic regions more commonly assume the role of alienated and cynical supplicants.One the supply side, the performance of representative government is facilitated by the social infrastructure of civic communities and by the democratic values of both officials and citizens. Most fundamental to the civic community is the social ability to collaborate for shared interests” (Putnam, 1993: 182).

Social organization, especially the type that promotes trust and collaboration, facilitates demand expression by combining many voices into one. This kind of voice combination manifests in the development of clan mechanisms that influence the operation of government (the organization representing collective interest) (Ostrom, 1996). In developing economies questions exist as to the quality of social organization and social capital, and the ability of citizens to express themselves through their clan mechanisms. The questions focus on problems of both social alienization (whereby entire groups lack the organizational structures necessary for demand expression) and defective social organization (whereby social groups are organized along vertical lines of dependence, clientilism and patronage that also limit ‘voice’) (Blair,1999). These issues are seen to influence levels of social organization and reform in various countries: Blair’s study (1999) compares countries like the Ukraine (with little social organization or social capital) with Honduras (where civic organizations have actively “pressed for better service delivery in health and sanitation” (Blair, 1999: 29)) to find that the basis for reform is greater in the latter types. Overall, however, such research finds low levels of social organization and capital in the developing world, and where such is evident, it is usually very localized and difficult to identify (as shown by Blair’s discussion of the localized nature of social organization in the Philippines).

Founded on the understanding that citizen voice should form the basis of any public sector results orientation, and that such voice is amplified in situations of social organization, ROME promotes the creation of social groups that voice their demands to government, putting “pressure on government to improve public services (Kaul,1997,16). Analyzing the literature on social capital development, participation, and democratic local governance yields two principles for such social organization promotion:1. Build reforms on existing social organization and capital, and promote further social

organization building through reform. 2. Give social groups access to public information, to provide a rallying point for

interaction and a basis for informing social demand.

Build on existing, and promote new, social organization and capital

Many reforms do not start with citizens and actually isolate them from the reform and governance process. ROME counters this with the fundamental emphasis on developing voice mechanisms (as discussed earlier in this chapter) and on identifying local level social organizations through which demand is voiced, working with such, and building others. The principle for reform has a number of dimensions, which all contribute to the

- 63 -

goal of integrating “antecedent and new structures” as the “basis on which ‘civil society’ can be built and strengthened” (Vaughn, 1995: 501):1. Where reform is sourced by central or external entities (as is usually the case)

considerable attention should be paid to identifying social organizational activity (both quantity and quality) at the local level.

2. Where social organizations are evident, these should be actively involved in reform (both in determining appropriate reform designs and in implementing and monitoring reforms). These local organizations become key reform partners.

3. Where social organization and social capital are lacking (and ‘voice’ mechanisms either don’t exist or are dominated by narrow, unrepresentative interests), reformers should seek to understand why, and: If key groups are socially alienated, identify potential reform partners or

interventions to facilitate social organization, or If key groups are organized according to vertical lines of dependence, set rules of

engagement requiring broad representation in governance matters.

The first aspect of this reform principle involves identifying social organization types and quality. Geographic Information Systems are often used to identify social organizations in place in different localities. The types of organization emphasized in the literature reflect civic involvement traditions and that “instill in their members habits of cooperation, solidarity and public-spiritedness” (Putnam,1993: 89). Where civic organizations have such effect, “feelings and ideas are renewed, the heart enlarged, and the understanding developed” (Tocqueville, 1969: 513-514) and citizens learn “public cooperation, discipline, the benefits of mutual compromise and “an appreciation for the joys of successful collaboration” (Putnam, 1993: 90). Examples of such social organization structures include:

Political parties: Political parties are a form of organizational arrangement through which like-minded people can express their demand. Putnam found that this was the most basic form of social organization in Italy (Putnam, 1993: 109). Blair argues that political parties are important in presenting alternative visions for doing public business (Blair,2000:28). Of particular interest are the influences opposition parties have on governance outcomes, as these provide channels for non-establishment voice into the governing process.

Cooperatives and mutual aid societies: The literature shows that cooperatives and mutual aid societies have the effect of consolidating group interests and influences over resource allocation and governance structures. Philippine Irrigation cooperatives, or Zanjeras (meaning “cooperative irrigation society”), for example, have proved very effective in structuring water access in rural areas. Members of the cooperatives have shares, and cooperate in irrigating their lands. The cooperative is internally regulated and rules are internally enforced, with members learning the importance (and benefits) of civic cooperation, trust and reciprocity. The cooperatives are locally developed and organized, which is cenytrally important to their success: “the zanjeras make efficient use of water through well-maintained structures of local design” (Ostrom, Schroeder and Wynne, 1993: 88).

- 64 -

Local association memberships : The basis for local involvement in social action groups is the ability of such groups to tap into and build on existing forms of social capital, especially among the poor. Organizations like peasant associations are often the springboard for both activist and government inspired forms of social organization in the developing world (Schneider,1999;Blair,2000). Associations take different forms, from credit associations (like ROSCAs in Africa) (Klitgaard, 1996) to professional associations. In Ukraine, a recent USAID study found “essentially no civil society at the local level,” but also recognized the beginnings of such society in the development of residence associations in apartment buildings (Blair,2000: 28). In these associations citizens find a vehicle for demand expression, that amplifies their individual voices.

Activist organizations: Recent studies show that active social organizations operating on behalf of the poor or disenfranchised can make a difference when it comes to governance quality, and mobilize results demand (Blair,1999; Schneider,1999; Brinkerhoff,2000). These organizations are usually developed within communities, on the strength of charismatic intervention by activists. Three Indian examples are DISHA (Developing Initiatives for Social and Human Action) in Western India, the Public Affairs Centre (PAC) in Bangalore, and MKSS in the Rajsamand district of Rajsathan province. Successful activism in these organizations is the result of local involvement in the activism. Similarly, apartheid era change in South Africa, and the subsequent change in governance structures, was driven by activists in civil society, loosely defined as the “non-governmental democratic force representing the interests of ordinary people independently of the state” (Bond,1991:72). Activism took different shapes in South Africa, ranging from organized consumer boycotts in towns like Boksburg to the activity of civic organizations in areas like Alexandra (Mashabelo and Narsoo,1989; Bollens,2000). Such community activity led commentators to suggest a strategy of ‘non-reformist reform’ in the new South Africa that would be “grounded in the strongest civil society in Africa”(Bond,1991:72).

Religious organizations: Another form of social organization relates to religious

involvement. Such involvement brings citizens with like demands (moral, social and physical) together, facilitated by common structures (individual churches and mosques, denominations and religious covering groups, or ecumenical structures). An example is the role of Oaxacan Catholic clerics in Mexico who act as “critical mediators, helping communities to establish positions of strength and stability … in the articulation of a blueprint for a new kind of democratic society that challenges the hegemonic policy and practice of the Mexican state” (Norget,1997:96). A particular example of this work is manifest in The Union of Indigenous Communities of the Isthmus Region (UCIRI), which was formed in early 1980s under guidance of a local priest because living conditions were waning and government officials (on whom peasants were dependent) were proving unresponsive. In 1997 it included 50 communities and 3000 coffee producers, all voicing common demands for agricultural assistance and service provision. “Members have achieved what they call

- 65 -

“self-capitalization” or the appropriation of the production process, with the goal of self-sufficiency” (Norget, 1997: 105).

Other indigenous communal structures: Another form of social organization that reformers should identify is the indigenous communal structure. Blair found that Bolivian indigenous communal institutions were central influences on social capital and on new organizational structures in the country, including neighborhood organizations (Blair, 2000: 28). Indigenous structures are important contributors to communal living in many other countries as well, with a “rich tradition of associational life and strong interpersonal networks” being evident in Mali, for example (Blair, 2000:29). These structures are often either ignored by reformers or drafted into programs in a way that fosters repression, not association. Forms of social organization should not be seen as political structures for ‘capture’, but as mechanisms vital to amplification of civic voices in indigenous societies. In this light, although local traditional leaders in Nigeria are found to contribute to significant political and governance complexity, they are also seen to provide an active voice of demand on behalf of community. (Vaughn, 1995).

Beyond identifying the number of social organizations in place (through some kind of GIS system or through field work), reformers should also assess the quality of social organization. Putnam (1993) tells stories that most in the development community can relate to, when he investigates the negative influences of political organizations and other forms of social organization in southern Italy. He suggests that reformers ask a number of questions about the kind of social organization they see: Who leads organizations? What is the basis of membership? How regular is interaction? What are trust norms of reciprocity networks? These questions relate to a number of problems with forms of social organization in the developing world, including the dominance of vertical organizational structures and the influence of organizational agendas on organizational quality:

Avoid vertically structured organizations: Many social organization types are vertical in nature, reflecting the very bureaucracies critiqued in earlier chapters. Pelling’s (1998) example of political representation in Guyana is an example. In this setting national government used various methods to influence local political leaders, exhibiting a vertical power structure that limited local influence or voice. Research in South Africa similarly suggests that the pressure of internal politics in leading and opposition parties poses a potential limitation to the ‘representativeness’ of the parties. Putnam concludes that “It is not the degree of political participation that distinguishes civic from uncivic regions, but its character” (Putnam,1993: 109). Social and political organization that is conducive to effective ‘voice’ and results-oriented reform, are strongly representative, horizontal, and responsive to members (at the most local level possible).

Check the organization’s motivation: All organizational structures have their own agendas. While agendas are necessary for the organizations to provide a voice into the governance process (with such voice developed around a set agenda), agendas can

- 66 -

also harm the legitimacy of such organizations. This can ultimately negate any kind of social representation of an organization, and needs to be carefully monitored and understood. Social organizations that constitute good reform partners are those with vibrant memberships, transparent agendas and organizational processes, and a sense of independence from the reform process itself (such that, even as reform partners, they can be objective voices into the process).

Following the identification of social structures, a second important aspect of building and developing social organization involves fostering relationships with existing, voice enhancing, civic entities. This means developing social organizations as reform partners, which is a radical departure from the conventional partnership of central government with external donor in reform. Brinkerhoff (2000) finds that donor agencies are increasingly attempting to make such connections, and to build on localized forms of civic strength, but other critiques reveal that the centralizing approach still dominates. Social organizations can only provide the voice to identify results that mater if central governments allow them to. As Vaughn states, “the activities of ‘civil society’ at the local level (and other levels) is intimately connected to the institutions of the state” (1995: 501). Similarly, civic structures are ‘enabled’ entry to the reform process (or prohibited access) by the incentives created through donor involvement. Donor involvement in many rural development projects increases or reinforces “the centralization of recipient countries’ governments” (Ostrom, Schroeder and Wynne, 1993: 156). In order to increase social organizational access into the governing process, a number of changes are required, in the mindset of reformers and in the institutions of government:

Place civic access at the top of the reform agenda: Social participation is too often an aside in reform. It is considered too ‘messy’ and complicating to introduce representative civic organizations into reform decision-making processes. This perspective limits the potential of the civic, as both a source of demand for results (facilitating results identification) and results evaluation. In this light, Vaughn (1995: 501) speaks of “the importance of sub-national environments where communities feel, breathe and express themselves and thus, where development programmes are relevant.” Until reformers acknowledge that civic organizations provide sources of civic expression that are central to reform, it is unlikely that the potential of these organizations will be tapped.

Reduce institutional impediments to partnership: Donor agencies and central governments set the tone for reform, and structure the reform process. Institutions and incentives created by these role players can either reinforce existing, closed structures, or can challenge such by requiring partnerships with local organizations. These effects are evident when comparing social emergency funds developed in the last decade. Whilst civic partnership was a prominent aspect of the original World Bank Bolivian Emergency Social Fund vision, and relationships were closely developed and monitored, other social fund experience has not seen many partnerships or close relationships. Recent evaluations find that poor institutions (either limiting civic access or facilitating abusive access by vertically organized,

- 67 -

opportunistic groups) have undermined social fund performance in many countries. For both governments and donor agencies the critical issue is whether a viable system exists “where local communities can harness the productive resources of civic institutions and structures within the context of a decentralized political structure” (Vaughn, 1995: 516).

The third aspect of building on and promoting social organization arises in situations where social organization and social capital are lacking (and ‘voice’ mechanisms either don’t exist or are dominated by narrow, unrepresentative interests). This is often the ‘default’ situation many reformers claim and use to justify non-participation in reform. In ROME, a non-reform is better than a non-participatory reform, as participation is the essential building block of change. Thus, where reformers face the challenge of low social organization, they should see their first job as one of developing such. This involves two strategies, focused on the reason for the lack of civic structure:

De-alienate the alienated: If key groups are socially alienated, such that they are systematically limited from developing their own forms of social organization, reformers need to facilitate their de-alienation. This could involve identifying potential reform partners from outside a locality to facilitate social organization. Bringing ideas, and influences, from outside a locality can spark social arrangements within such, as has been the case in initiatives relating to social auditing in India (Jenkins and Goetz,1999). Providing groups with information is one way that such alienation could be countered: the information provides a rallying point for interaction, which can be further promoted by external leadership (as in the case of DISHA in India).

Institutionalize horizontal social organization: If key groups are organized according to vertical lines of dependence, such that organizational leaders dominate members, reformers should set rules of engagement requiring broad representation in governance matters (prohibiting participation monopolies). Examples of organizations purposefully created by governments or donors to organize horizontal structures in civil society include the Village Development and Security Committees in Malaysia and the Territorial Base Organizations and Vigilance Committees in Bolivia. Although different, these organizations are impacting on their governments and providing a means for citizens to voice their demand for results (Schneider,1999)

Informing citizens about government, stimulating ‘voice’

Information is a key factor necessary in mobilizing civic groups around the cause of good governance and informing civic demand. Without information access, citizens cannot develop effective and realistic demand profiles, and government cannot be held responsible for the results it produces (simply because, if government controls access to information on results, it can prohibit anyone from ensuring it produces results). Furthermore, information access shapes relationships between social organizations (be they activist, political or religious) and those holding information (generally central governments). Social organizations that have information access “are likely to be best

- 68 -

placed to access resources and enhance security for themselves and others” (Pelling,1998: 471). This point is echoed in Freire’s (1985) work, which suggests that information defines ways in which actors relate, legitimating authority in specific groups and excluding others. This is obvious when one compares the conventional developing country experience with successful forms of civic involvement: Centralizing information, centralizing governance: Commonly, governments in the

developing world internalize information and elevate the knowledge of technical, managerial and political elites over the local ‘time-place’ knowledge of grassroots actors. This underscores an official and unofficial limit on citizen information access—information is not provided to social groups because their input is not valued or required—and legitimates the exclusion of citizens (especially poor citizens) from decision-making. It reinforces an inward, managerial model of governance, and prohibits the adoption of results orientated accountability. Civic organizations are blunted in their ability to amplify voice as well, because they have no subject for amplification and no ‘information legitimacy’ in the eyes of their supporters.

Spreading information, spreading governance : All of the ‘successful’ civic organizations mentioned in this monograph derive a certain degree of influence and relevance from the information they access. MKSS in Rajasthan is one entity that is directly concerned with enhancing citizen rights to information in their region, while DISHA and the Public Affairs Center focus their activism on accessing and disseminating information about what government promises to do, versus what government actually does. In these cases, as well as the Malaysian and Bolivian examples of social engagement, the ability of society to demand results is enabled by the combination of information access and voice (through social organizations). In these cases, “the struggle for access to information challenges the obscurantist and remote culture of the bureaucracy, and reinforces democratic notions regarding the obligations of government officials and elected representatives as public servants” (Jenkins and Goetz, 1999:605). Further, information access enhances the role and legitimacy of social organizations that can provide meaningful voice into complex governance dilemmas.

In developed economies, citizens usually have easy access to government information. This access is often mandated by law, and exercised through organizations representing citizen interests—whether they be special interest groups, or explicitly political bodies. This access is seen as the basis for governance interaction, and facilitates demand for results (Osborne and Gaebler,1992). In developing countries these forms of access are often lacking, and demand for results, though latent, is not empowered by information about results production. To counter this, reforms focused on results should concentrate on increasing access to information about government (especially information provided in the budget). This information provides a rallying point around which social organizations can organize themselves, and on the basis of which citizens can demand and evaluate results. In Malaysia and Uganda, governments facilitate public information access by posting information for citizen viewing, a practice similar to that in New Zealand (Chiu,1997; Reinikke,1999). In all three cases, enhanced access to information has made the government more accountable and has facilitated a results awareness. This results awareness has matured to a point where the kind of information published is

- 69 -

focused on results—the outputs and outcomes of production—rather than inputs, which were the focus of initial publishing drives. In Malaysia information is reported ‘store front,’ and in Uganda mass media broadcasts details about government allocations, and clinics and schools report similar details on notice-boards and in classrooms.

These examples provide a number of important pointers for reformers: Information access must be formally institutionalized: If governments are committed

to results, they should legislate access to information about results (and the production of results). Such legislation should take the form of a ‘Freedom of Information Act’ as in the United States. A similar law enabled DISHA (Developing Initiatives for Social and Human Action) to access and analyze provincial budgets in Western India but in other instances legislation does not facilitate information access (DISHA, 2000). This was the case with the MKSS organization’s ‘right to information movement’ in Rajasthan, which involved informal information gathering through personal networks because of the lack of a formal channel (Jenkins and Goetz,1999). The importance of formally institutionalizing information access cannot be under-estimated, as experience shows that public actors purposefully limit and frustrate information access. Without a legally enforced right, citizens are likely to face extreme difficulty obtaining information in such situations.

Information type, content and structure matters: The point of increasing information access is to improve civic knowledge of what government is doing and what it promises to do, so that citizens can shape their results demand. Through this, results-oriented reforms hope that citizens would pressure governments to produce efficiently, creating an incentive for government to adopt a results-orientation. The best way of accomplishing this is by publishing the information so that citizens can access it without cost, and can easily link the information to their every-day lives. The Malaysian publication of production promises and results is best practice in this regard, as is a similar program in Hong Kong and the mass broadcasts of Uganda (Chiu,1997; Lam,1997; Reinikke,1999). Key aspects of these reforms are the open publication of information, the simplicity of the information, the emphasis on results (rather than inputs), and the clarity of communication.

One of the problems with government information is that it is often difficult to evaluate and understand. This is especially the case with the budget, which is generally developed according to an accounting system and structure that is peculiar to the public sector and fraught with inconsistencies (Premchand,1996). For a results-orientation to develop from information access, reform should focus on de-mystifying this information for easy readability. In New Zealand, the results movement is structured around a change in accounting systems (Ball and Dale,1998; Ball et al., 1999). The new systems follow the accrual method of budgeting and require financial statements similar to those in the private sector, facilitating easier understanding of the documents framing government activity. Countries like the United States and the United Kingdom are also adopting accrual budgetary methods to make government information more citizen oriented and standardized.

- 70 -

DISHA provides a case study for reformers, particularly those interested in the developing world, where cash budgeting continues to reign and budgets have to be demystified before citizens can truly understand them. DISHA’s Participatory Budget Analysis program has de-mystified the process of governance through pro-active budgetary analysis involving pro-poor NGOs. The budget analysis has helped to keep government accountable for general money flows and policy decisions. The press, and local activist groups, have a device to assist them in understanding what government does, closing the power gap between informed bureaucrats and uninformed citizens (DISHA,1999). The three central lessons for reform are:

Bad information is better than no information: Public information is often produced in ways that are difficult to evaluate and understand, even in developed countries (Niskanen,1972). The experience of organizations like MKSS and DISHA show that complex, input-oriented budgets with little support documentation can provide important information about government. Governments should not stall or limit societal access to information because they think information will be useless. Information should be provided ‘as-is,’ and with as much supporting documentation as possible (Jenkins and Goetz,1999).

Making budgets useful 1, de-mystifying the information: Governments that produce mystified budgets should not think that these constitute ‘information provision’. Information is only truly provided when it is useful, which requires that citizens have the ability to read and understand its content. Governments and groups trying to make sense of public data, particularly financial data, should ask basic questions about what the information structure is trying to convey—and to hide. All information has value when it is read according to the purpose, and in the spirit, in which it was created. Governments and activist groups attempting to facilitate information-based results incentives should arrange existing information in ways that people can understand. DISHA breaks provincial budget details down into a comparison between promises in government manifestos and allocations in the budget, for example, and finds a large audience for such simplified data. In countries like New Zealand and the United Kingdom, performance reports accompany complex financial documents to give citizens and lawmakers an accessible (though simplified) view of what they want to see. This increases the value of information.

Making budgets useful 2, adopting conventional reporting approaches: Information value is directly correlated with information use, and information use is directly correlated with information usefulness. Apart from demystifying budgets, another way of improving usefulness is by adopting conventional standards of communication. Countries like New Zealand produce information according to private sector best practice, and the results accounting and budgeting movement in the United Kingdom has the same aim. In developing countries like South Africa such practices are also being adhered to at local, provincial and national government levels, but the process is slow and there are still major groups who cannot understand the documentation. In countries like this there is a strong argument for adopting information provision standards that are vastly different to international best practice,

- 71 -

but that are most used by citizens. This will allow easy social digestion of information by citizens.

5.2. Decentralizing governments to access and respond to citizen voice

Just as citizen voice is a basic requirement for effective results-oriented government, and information is important to inform that voice, there are some issues to do with government design that constitute a basic element of ROME. These issues revolve around how government structures affect government ability to develop a results-orientation. Simply put, if they are going to adopt a results-orientation, governments need to be designed so that they can listen to citizen voice.

The traditional public sector model develops from Weberian theory, as applied during the colonial period in most developing countries (or as a copy of ‘successful’ systems in mid-century industrialized countries). Through its rules and process-orientation, this model emphasizes input control and activity efficiency and reliability—not results. This emphasis amounts to a concentration on the supply side of governance and on the mechanisms and processes of management—“knowledge of systems, structures, procedures, rules, and legal precedents” instead of the outputs, reach, outcomes, and impacts of public production emphasized in results-oriented governments like Malaysia, New Zealand, Australia, and Singapore (Grindle, 1997: 487). There is no guarantee in the model that “service supply will be in the public interest” (Coston,1998: 486).

To correct for this, reform needs to tackle fundamental design factors to facilitate service that is in the public interest. One of these factors revolves around the question of intergovernmental decentralization. Beall (1997: 951)rightly identifies that the reform community is taken up with decentralization ‘fever’. The decentralization fever she refers to links decentralization with social capital development, regularly considered a recipe for robust governance necessary to the development of results-oriented, responsive government. The thinking behind such fever is that, where social organization (as discussed) is developed in a decentralized setting, the following benefits will follow:

Improved efficiency is realized as governments closer to citizens meet demand profiles better than government operating at a distance from citizens (Tiebout, 1956, Oates, 1972; Musgrave, 1985; Tanzi, 1996).

Greater equity and democratic governance is facilitated by the closeness of government to citizens, and the fact that central governments are no longer all powerful (Weingast,1995; Inman and Rubinfeld, 1997).

Enhanced economic competitiveness and effectiveness, as well as results-orientation, is achieved as citizens exercise their voice and exit options to demand governments that serve them well (Tiebout, 1956).

Mohan and Stokke (2000: 250) say that decentralization is central to modern public sector reforms with the important underlying principle that such decentralization is centered on the public. What decentralization offers is responsiveness, shortness of communication time, and directness of channels between servers and clients—“increased

- 72 -

competition among levels of government, increased local control over resources, increased decision-making authority, increased targeting of services to local needs and desires, and increased transparency” (Brinkerhoff, 200: 604). These all encourage and enable innovation and provide cost efficiency incentives in government, elements of competition that enhance a results-orientation (Shah, 2000).

Experience with decentralization reforms has not been overwhelmingly positive, however, and a number of observers question the use of decentralization as a reform tool. Others have identified the problem with decentralization attempts in terms of factors of design that demand attention in reform, not an active questioning of the general idea of decentralization itself (Shah,1998). These design factors center on (1) providing services, and locating service provision accountability, as locally as possible, and (2) involving citizens in government decision-making and implementation.

5.2.a. Producing services as close to citizens as possible

Decentralization is important in the results-oriented movement because it provides a public sector design that lends itself to efficiency and ‘competition.’ Public economics teaches that different types of government products—or services—are suited to production at different levels of government (Oates,1972; Musgrave,1985). In this light, the subsidiarity principle and the decentralization theorem, regularly used to motivate decentralization, both suggest that services should be produced as close to citizens as possible—all services should be produced locally, unless there is a good reason for production centrally (generally the case with defense, welfare, and macroeconomic management).

In developing economies arguments about limited local capacity and accountability dominate decentralization debates, and result in many local-appropriate products being produced centrally (Bahl,1999). In recent examples of reform, this is however changing. Governments are decentralizing important services like health care, education, and sanitation in countries ranging from Ghana, to Bolivia, to Malaysia, Uganda, and Pakistan (Kisubi,1996; Beal,1997; Larbi,1998; Reinikka,1999; Brinkerhoff,2000). Results-oriented government is developing in all of these examples, with the decentralized entities leading the way. In all cases of success and failure, the following three design factors are seen to influence the ‘quality’ of local governance:1. The degree to which functional authority and accountability is decentralized to local

entities.2. The degree to which results-oriented accountability chains are created from local

governments to local people.3. The degree to which results-oriented accountability chains are created from local

governments to local people.

Decentralizing functional authority and accountability

The move to decentralize is founded on the argument that citizens can hold decentralized governments to account more easily than they can centralized governments. This

- 73 -

argument assumes that local governments have the authority necessary for accountability. In many instances decentralization arrangements are so loosely defined that this is untrue, and governments are not accountable to citizens. This is the case in all areas of practice, where services are provided by all levels of government (with each level weighing in at different stages in the production process). Schroeder and Andrews (2001) discuss this problem with regards to rural water, rural roads, primary education and health care provision in Africa, concluding that governments typically provide services in complex processes that defy accurate accountability because of the character of services themselves (services with spillovers automatically require national and sub-national provision, for instance) and of the service providers (with issues such as capacity and politics proving major obstacles to clear functional decentralization).

The murkiness of decentralization requires significant attention if decentralized governments are to be truly accountable. One form of such attention comes through identifying which services are best suited for production at which level of government. There are two influences on such identification: (1) Standard decentralization theory and the subsidiarity principle, and (2) The question of citizen interest. The former approach is common in the literature whilst the second is more closely related to the results-oriented approach suggested in this paper. The approaches can be summarized as follows:

Decentralization theory: The “benefit principle” should guide decisions of functional and expenditure allocation. This principle holds that the responsibility for a particular function should be located at the level of government where the benefits of the function are accrued (Shah, 1994; Arora and Norregaard, 1997). Any function that benefits a narrow and readily defined group should be the responsibility of that group. Any function that benefits groups should be produced by a central actor representing the cooperative or combined needs of these groups—the central government. Economics-based theories of federalism categorize governmental functions according to this approach, with four general groupings—economic stabilization and development, redistribution and allocation. The conditions that lead to different divisions of the task of government include concepts such as capacity, preference and efficiency, as well as the location of benefit and the mobility between jurisdictions (Shah, 1994; Musgrave, 1976; Break, 1980; King, 1984; Oates, 1991; Rich, 1991; Rivlin, 1992 and Peterson, 1995). Whatever the allocation of functions in any government, economic theory also holds that such assignment should be clear, consistent, institutionalized and stable—facilitating true autonomy on the part of sub-national governments (Shah, 1994; Peterson, 1995; Arora and Norregaard, 1997). If assignments are unclear or inconsistent or if there is no institutional (legal or otherwise) entrenchment of responsibilities it is likely that the gains of decentralization will be frustrated; sub-nationals will not enjoy real authority to allocate services and spend in a way that they deem best represents the preferences and production characteristics of their jurisdiction.

Citizen interest: This second approach holds that, given the emphasis on citizen involvement in governance, it is they who should choose the kinds of services they want to be produced locally. If they are interested in the service they are more likely

- 74 -

to monitor its provision and ensure the success of the local entity. Citizen involvement and interest is vital to ensure decentralized governments work well and are accountable for results. Examples of services that are working well at the local level are health care (Ghana, Uganda), sanitation and water provision (Pakistan, Recife in Brazil), and education (Uganda)—all of which were of great local importance, and which attracted great attention in decentralization. These services were not only of great interest to citizens, but they were also easily evaluated at the local level. This increased the potential for citizens to hold governments accountable for production.

Combining these two approaches to function allocations between government yields two lessons for reform:

Legislate service provision roles to consolidate accountability: Whatever the functional allocations are between governments, they must be institutionalized. In countries like New Zealand the roles of different levels of government are legislated (United Nations,2000). This protects lower levels from higher-level predation, limits potential service overlap, and formalizes accountabilities at the various levels of government. It makes it very clear who produces what, and to whom the public sector machine is accountable. In many developing countries, a results orientation will be difficult to institute at various government levels because there is poor clarity about who produces what—making it difficult to hold producers accountable for their output.

Decentralize services for which tangible results are easily identifiable: The idea of decentralization is closely linked to that of results oriented governance. Citizens play an important role of guiding and evaluating government in both. Ensuring effective service accountability at the local level requires ensuring that citizens have an interest in holding service providers accountable. This is the experience in decentralized systems from Costa Rica to India. In the Costa Rican situation, although citizens were officially isolated from the governance process, they played a major role in shaping and evaluating hurricane assistance services (Breslin and Campbell, 1998). This was simply because of the importance of such services to citizen interests. Necessary conditions for such involvement are that citizens are themselves interested in the services provided, and that the services can be evaluated by citizens. Citizen evaluation is based on the citizen influence in the governance system (how active and influential are social organizations) as well as the degree to which results can be identified. Where citizens are well organized and results are easily identified, service provision should be localized.

Creating accountability chains from local governments to local people.

The benefits of decentralization derive from the space created for the voice of the local population into local planning (Atkinson,2000). In many decentralization cases this space is not created. Guyana is an example in which decentralized government was introduced as part of reform, but failed to emphasize community accountability. As with

- 75 -

many versions of developing country decentralization, the Guyana case was characterized by a lack of transparency about decision-making, an absence of community-level forums for information dissemination and decision-making, and an absence of community participation in ‘public meetings’ (Pelling,1998,481).

It is common that governance mechanisms emphasize accountability to external donors and responsiveness to internal leaders instead of accountability and responsiveness to citizens. These vertical versus horizontal lines of accountability are conducive to ineffective and unpopular local governments. Such lines of accountability are particularly limiting to the development of a results orientation as they emphasize power based, rule-bound relationships rather than service-oriented relationships.

In other decentralization examples, from health care provision in Brazil to sanitation provision in Pakistan, decentralization has fostered greater citizen accountability and led to more effective governments (Atkinson,2000; Brinkerhoff,2000; United Nations,2000). There is a definite correlation in these examples between the strength of citizen-local government connection and local government effectiveness. The connection is developed through ‘social contracts’ between political leaders and their constituents and results-based contracts between administrative officials and their citizen ‘customers.’ Such ‘contracts’ require clear lines of political accountability to citizens at the local level, the formalization of mandates between political representatives and citizens (as discussed earlier in this chapter) and active citizen participation with representatives.

Create clear lines of political accountability to citizens: The electoral process constitutes one line of political accountability, but needs to be supplemented with lines reflecting information about what is being produced, how, and when. Citizens pay taxes and vote for representation to get results, interested in what government does for them. In other words, results matter and should form the basis of accountability links between citizens and their local governments. Lines of accountability should be based on the channels of communication discussed previously in this chapter, allowing citizens to regularly evaluate the results-oriented performance of politicians (against their institutionalized election mandates).

Mandate requirements for local government accountability to citizens: In New Zealand, local governments are required “to prepare annual plans in consultation with their communities and to report on their performance in accordance with the plan at the end of each year” (United Nations, New Zealand study,2000,9). South African municipal governments are similarly required by law to practice regular, “co-operative, co-ordinated, strategic and participatory planning” (Otzen,1999:1). These plans are then meant to drive budget processes and to enhance transparency and accountability to citizens. The requirement creates and safeguards a results-oriented accountability link between decentralized governments and the citizens such governments serve in New Zealand. In South Africa the results focus is not made explicit enough, and plans fail to constitute citizen-relevant mandates. This minimizes the accountability-enhancing influence of the reforms.

- 76 -

Ensure active citizen participation, in politics and public production: The best way of ensuring that decentralized governments are accountable to their local constituencies is to involve those constituencies in the life of the decentralized governments, through the development of contracts. The basic level of involvement is through the electoral process (where political, social contracts are developed). These are not considered enough to ensure accountability, however, being “crude instruments of popular control” that are settled too infrequently on the basis of issues that are too vague. The electoral process should be supplemented with programs of citizen involvement in decentralized governments to ensure active and participatory government is realized (Pelling,1998). These have already been discussed in this chapter, but are noted again because of their particular relevance at the local level.

Results accountability between governments themselves

There is a reticence in many decentralization initiatives to give local governments autonomy to manage their own financial and production affairs. Higher level governments often exert outside control over decentralized governments through budgetary requirements and controls over personnel management (Bahl,1999). In both cases these controls reinforce a process-oriented government design and frustrate any potential results-orientation. They are shown in the right hand side of figure four, as the ‘usual, vertical control decentralization design.’ In this kind of design, citizens are shut out of influencing their local governments, with the influence being top-down through rules of financial and administrative process.

Figure six Decentralizing for results—as compared with usual, vertical control design Decentralizing for results The usual, vertical control decentralization

designHigher level

government entity

Funds provided, rulesmade in terms ofresults-oriented contracts focused onlocal community

Higher level Government entity

Funds provided, rulesmade in terms oftop-down processfavoring verticalaccountabilities

Lower level government entity

Low level entitiesare accountable to citizens, who hold them accountable,and monitorresults contracts,reporting on results

Lower level Government entity

Low level entitiesare not accountable to citizens, having no link to local citizens. They are accountable upwards, politically and administratively.

CitizensCitizens

Note: Gray shading indicates role emphasis in the respective designs.

- 77 -

The obvious answer to this problem is to allow local governments to raise their own funds and spend such as they need, and to hire and fire as they please. In a study of 29 public organizations in six developing countries, Grindle (1997) found that, where decentralization was matched by such local autonomy, governments were “good performers” (achieving high levels of capacity, completed tasks, responsiveness and effectiveness). Developing countries have not decentralized autonomy to this degree, however, and limitations on local taxing ability and so forth limit the real chance of them doing so in the future.

As an alternative, some governments have developed ways of relating higher to lower level governments through results-oriented lines of accountability that do not stifle performance. The left hand side of figure four shows this kind of arrangement, in which the institutional links between higher and lower level governments are designed to reinforce whatever accountability lines there are between the lower level governments and their citizens. The model of decentralization for health care in Brazil is an example. Results-oriented contracts are developed between central government (providing funds) and local clinics (providing services), ensuring accountability between levels of government without a process bias (Atkinson,2000,623). The contract system institutionalizes greater autonomy for local governments than do other decentralization regimes, holding them accountable for their results—and not binding their processes.

Another example is the idea behind the creation of ‘vigilance committees’ in Bolivia. Comprised of local people, and meant to be representative of local organizations and a localized voting process, these committees are mandated to monitor local governments and have a direct line to the central government for any complaints about such local government (Blair,1999; Schneider,1999). Their monitoring is results-based and places the emphasis on local citizen satisfaction instead of centralized control. Central entities fund local entities, and require the local entities to commit to results contracts in return. These contracts are set with local communities, who are charged with monitoring the contracts and ‘reporting’ to the central authorities on performance.

There are two critical reform lessons emerging from these approaches.

Eliminate intergovernmental links reinforcing process rather than results: Institutions operating through budgets and grants, as well as personnel rules (governing conditions of service throughout government) often bind lower level entities to process-related rules. They need to be actively reconsidered if a decentralized government is to be truly results oriented. It is impossible to hold a local, or mid-level, government accountable for results if they do not enjoy control over the funds allocated to them (often coming ‘with strings attached’), or their major input expenses (usually determined by central conditions of service), or their core programs and initiatives (often decided by central governments).

Develop intergovernmental contracts that focus on results, not process: It is unlikely that many developing countries will be able to develop local and mid-level governments that do not have to relate on an ongoing basis with higher-level governments. This said, it is

- 78 -

possible to develop relational and institutional links between higher and lower level governments that do not limit the performance potential or local accountability of the lower level entities (even when the former acts as banker to the latter). Results oriented contracts, and the introduction of civic organizations as the referees of such contracts, are interesting institutions for governments to use in this regard. Such arrangements should always be developed constitutionally to safe-guard their existence.

5.2.b. Involving citizens in government, facilitating partnerships for results

In conjunction with organizing citizens outside of government, providing citizens with information to frame demands for results, and decentralizing governments to position them to respond to results, examples of successful results-oriented reforms also concentrate on involving citizens in government. Indeed, results-oriented reforms, and many related interventions, depend upon citizen involvement in government. Decentralization, for example, assumes participation by society in the life of government. The very “proposal that decentralized management…will be more responsive to local needs…depends on decentralization being accompanied by increased involvement by the catchment population in some way in order to define those needs” (Atkinson,2000,628).

In Europe, social involvement in government is evident in various versions of discursive democracy. This form of governance, developed from the strengths of social engagement, involves citizens “in the shaping of issues, in the formulation of responses, and perhaps also in the implementation of programs once they have been adopted,” creating strong social incentives for officials to pay close attention to citizen demands as well as a channel through which to communicate citizen demands and consolidate societal/government partnerships (Peters,1996,55).

In developing countries, and in some developed settings, these kinds of participatory structures are not evident. In this regard, “moving towards inclusive planning mechanisms requires open political space which may be absent in those nations that have experienced strong centralized control, have only recently commenced the democratic transition, and where civil society is narrowly defined, and community organizations (and local NGOs) may be absent. In these instances…there may be a role for ‘community facilitators’ to act as catalysts to develop local social capital” (Pelling1998,473).

Examples of such community facilitation are provided throughout the literature, and all involve a steady move towards involving communities in the planning and production of services and the move towards granting communities political and economic autonomy. An example comes from Recife in Brazil, where government was traditionally centralized and unresponsive to the needs of the poor. In particular, the central government imposed excessive standards on engineering that prohibited the supply of water to poor areas (Ostrom, 1996).

When the government decentralized and gave autonomy to local governments, the local governments mobilized residents for the construction of sewer lines. This was done via a series of structured community meetings to introduce ideas and get

- 79 -

community support and commitment, and contracts in the form of signed financial and working obligations. The participative, results-oriented governance structure involving citizens in sectoral projects, led to improved services and a robust local government. It was characterized by decentralized governments relying on: 1. Social involvement in the production process, and participative planning and

production.

Involving citizens in government production and planning processes

Traditional government structures, based on Weberian theory, emphasize professionalism and limited entry over broad involvement and openness. In developing countries, this kind of organization is often criticized for quashing society and not attempting to serve. Furthermore, it is generally limited in its ability to serve because it does not have a direct line of engagement with its citizens and does not have the internal capacity to respond to citizens even if it did. In systems of discursive democracy across Europe, governments operate differently—with officials appointed on a part-time basis and local citizens involved daily in advising on policy and weighing in on policy implementation (this is especially the case in Switzerland).

In these forms of government social involvement proves vital in drawing government closer to the society it serves, accessing social capacity to expand government service capacity, and allowing citizens the space to speak into government and hold government accountable. By increasing participation in sectoral projects like sewer production in Recife, for example, citizens develop a close understanding of the workings of government, and create and consolidate social links that empower them in their relationship with government (links developed with other citizens involved in the projects, or with non governmental organizations already partnering government agents).

This knowledge, and these links, increase the interest of citizens in government and the potential voice of citizens into what governments are producing and how governments are producing such services and products (Brinkerhoff,2000). In order to create participatory structures that facilitate a results orientation in government, as well as improved results themselves, reform interventions should be characterized as bottoms-up and inclusive, taking the following lessons into consideration.

Involve citizens in bottoms-up initiatives: The most effective participatory programs include the local population in decision-making from the earliest opportunity in projects, give the local populace a defined role in decision-making in projects, and provide a clear channel of communication to political leaders.

Ensure that involvement programs are inclusive: Social involvement in government can easily be hijacked by particular interest groups (Pelling,1998). If involvement is to really impact on the results of government, it should appeal to the constituency as a whole, and not be dominated by specific interest groups (social, political or managerial). In particular, citizen involvement should not amount to the mere expansion of the public sector work force—dominated by administrative officials.

- 80 -

5.3. Conclusion: The citizen-government contract

The introduction to this monograph argued that the Results-Oriented Management and Evaluation model (ROME) differs from other reform options in two ways: (1) It combines reform elements around local participation, with citizens creating the pressure for results-oriented government, and (2) It emphasizes participation to the point that a non-reform is better than a non-participatory reform.

This chapter has laid the foundation for the rest of ROME, by providing some detail of the participatory decentralization element. The argument is simply that citizens form the basic focus of a results-oriented government, and thus governments need to take issues of citizen participation more seriously in reforms. This entails, as laid out, creating mechanisms for civic demand into the governance process, working with civic organizations towards reform, informing citizens and their organizations about government production processes and performance, and locating government close to its citizens to facilitate hearing and response. When combined, these aspect of participatory decentralization facilitate the development of a citizen-government contract. The contract is most directly between citizens and political representatives, in which the latter are held accountable for their performance, as measured against their platform mandates. Figure seven shows how this contract is formed.

- 81 -

The figure shows the public sector production process from earlier chapters to identify the role players typically associated with each stage of the process. The administration is usually concerned with motivating the adoption of programs and managing the production of outputs from such programs (by translating inputs, through activities, into outputs). The executive concentrates on setting policy in terms of specific outcomes of government, and usually oversees the administration in its production of outputs. The legislature is accountable to citizens about the impacts of government production on their lives.

The graph shows the roles as they usually are in the public sector—top-down with the administration holding a power over the rest of the system, and the concentration resting on the inputs and outputs of such process. Citizens are largely excluded. The reform elements embodied in participatory decentralization transform this traditional approach by ensuring that government is bottoms-up, that citizens are the main voice and source of demand for results, that citizens are informed to demand results, that government is structured to respond to such demand (being decentralized), and that citizens are involved in planning and producing results.

When combined, these elements yield the core parts of the citizen-government contract: citizens voice demand, and hold politicians accountable for results (using the information they access to evaluate the contract) and use voice mechanisms to respond to contract performance (having the regular ability to ostracize or commend representatives as well as the periodic ability to re-elect or replace representatives). When individual elements are missing (citizens are poorly organized, or lack information, or have to deal with distant governments) governments are likely to experience the problems. Examples include the following:

Although the MKSS in Rajasthan has mobilized and organized citizens to demand results, and has accessed the information necessary for such demand, government continues to organize itself in a top-down fashion. The provincial government is large and unresponsive and resists any active decentralization or citizen involvement in its production process. As a result, although there have been gains through the ‘right to information’ movement, public organizations are still input-oriented and fairly unresponsive.

Although the Bolivian experiment with decentralization at the local level is designed to increase the participation of locals in their government by creating structures allowing social representation into local governments, participation has not been as inclusive and bottoms-up as desired. Social organization has been assumed to a large extent, and the ‘democratic’ demand for results that is meant to flow from the experiment has not been as evident as was hoped. The mechanisms of decentralization were well thought out, but social organization and information coverage has not effectively facilitated inclusive involvement in decentralized governments. The government is thus not as results and citizen oriented as planned (Blair,1999).

- 82 -

ROME involves implementing programs designed to mobilize and inform the public, while at the same time decentralizing government to ensure that public organizations are close to citizens, and involving citizens in public planning and production. Individually, these elements constitute excellent ideas of reform, but lack impact because of blind spots (as evident in the examples above). When the elements are combined, they cancel blind spots out and provide the first section of the ROME puzzle. This section constitutes the necessary foundation for ROME. When it is further combined with the elements of the other two sections (results-oriented management and results-oriented evaluations), ROME is completed and constitutes a complete and comprehensive approach to reform that is far superior to the piece-meal practices of the present.

An example of participatory decentralization in action helps to illustrate how the interaction of the elements feeds a results-orientation in government, but is still dependent on the development of other ROME elements (results-oriented management and results-oriented evaluations sections). The example is from the county of Pérez Zeledón in Costa Rica, where a county representative has championed reform by ensuring that local citizens have authority over discretionary funds allocated to them from higher levels of government (Breslin and Campbell,1999). These discretionary funds were usually used by representatives to develop lines of patronage and clientelism in their communities, but are now channeled directly to district councils for local allocation. The availability of the funds at the local level has spawned, and been accompanied by, a number of other developments at the local level: Old community development organizations have been revived (social organization). These organizations receive direct information about available funds from district

councils (information). These organizations work together with the local governments, who are accountable

to the people through the development organizations (participatory decentralization). The community development organizations decide on where the money should go,

prioritizing in terms of the outcomes they deem most important in their environment (participation in planning).

The community development organizations contribute the labor to complete the projects (participation in production).

The new system of governance involves a direct move away from the old top-down system of deciding what communities need at an administrative level. It now involves telling the communities what money is available and asking them what they want to do with it (Breslin and Campbell,1999). The reform has transformed the local governments into results-oriented entities in which communities can see work being targeted and completed. The involvement of communities in government has provided a foundation for results-oriented management and evaluation as well. Although these other dimensions of ROME are still being developed, there is evidence that administrators are receiving more precise demands for actual results (and are also getting the autonomy needed to produce these results). Public entities in the county are relatively more efficient and output-oriented than in other counties, borne out in relative response times to disasters. Pérez Zeledón’s quick and effective response, and results-oriented

- 83 -

management, is explained in terms of the “level of organization and the commitment of local communities” (Optiz,1998).

At the same time, results-oriented management and evaluation are not being as driven as the participation and decentralization elements have been. This is causing problems for reform, with too much centralization of power and bureaucratic control over funds and process rules limiting progress. These factors are now receiving attention (Breslin and Campbell,1999,30). As they receive attention, so the county moves in the direction of ROME—combining the basics of bottoms-up government with the elements and practices of results-oriented management and results-oriented evaluation.

- 84 -

CHAPTER SIXRESULTS ORIENTED MANAGEMENT IN THE ROME MODEL

The second section of ROME is results oriented management. This section has many pieces, often reflected in management practices transferred from the private sector where results are considered central to the business process. These practices, including total quality management, activity based costing, results-based budgeting, and the balanced scorecard, are often implemented stand-alone as individual pieces, and not in connection with one another or in conjunction with the citizen focus emphasized in participatory decentralization. This has led to great criticism of results oriented reform in developing countries, particularly focused on the non-democratic bias of such reforms (tools like total quality management have been implemented in insular, top-down ways, thwarting participation). The disjointed application of individual results-oriented management tools is also a dominant cause of constant reform failure: implementing a tool for one area of management does not assure appropriate tools throughout the organization or a change in culture or incentives that facilitates a results orientation in public administrations.

ROME has a very different approach to results-oriented management:

Results oriented management is not the dominant piece of reform, nor is it the prime focus of reformers. A results orientation in management requires a participatory, localized structure to work in the developing world, as it is the influence of citizens that define what results are and that create the incentives for managers to pursue results. Administrators will be expected to pursue the adoption and use of results-oriented tools IF citizens demand results from them. As such, results oriented tools are provided in ROME as options for government managers to appropriate when they have the incentive to do so, not as the thrust for reform.

ROME advocates results oriented management only when managers respond to the social demands for results and the incentives emanating from civic organizations. When managers respond in this way, indicating the will to change, they are understood to require assistance in developing results oriented tools to meet demand. This assistance should be provided for all areas of administration, not just through the provision of fashionable tools, but also through the adjustment of organizational structures, rules and relational connections. A change to results-oriented management requires a significant change in how governments do what they do as well as how governments conceptualize their role.

When business practices are connected and built on the fundamental client orientation and citizen focus reflected in the participatory decentralization section, and in conjunction with results oriented evaluation (to be discussed in Chapter 6), results oriented management resembles a set of clearly related roles and contract based relationships that center on the outputs and outcomes of government production, as in figure eight.

- 85 -

As with figure seven, figure eight shows a public sector production chain, which begins on the left with program identification and extends on the extreme right to public sector social impact. From top to bottom, the figure shows the dominant actors in the public production process. If traditional public sector management were reflected on the figure, one would begin by emphasizing the role of the administration in a top-down, bureaucratic organizational structure. The role of the administration in these structures is concentrated around the procedures of managing inputs in the activities of production (on the left hand side of figure eight). The executive plays a largely observational role, sometimes directly involved in managing at the inputs level as well. The legislature and citizens are often uninvolved in the processes of service generation (at the bottom of a top-down process). The input-concentration and hierarchical structure unite to create a process-orientation that dominates questions about actual production results.

Results-oriented management is completely different, as shown in the figure. Results oriented management is: Bottoms-up, with citizens comprising the starting point of the production process, Focused on citizens who provide the bottom-line for public managers in the form of

results-oriented goals, Focused on the right hand side of the production process, with an emphasis on

outcomes and outputs, Based on clearly identified results-oriented roles (focused on outcomes in the case of

politicians and outputs in the case of administrators), Based on contractual relationships between interconnected parties: citizens depending

on governments to impact their societies positively hold political representatives accountable for such; political representatives, working through a defined executive, hold administrators accountable for production of outputs required to produce the outcomes they promise to citizens.

- 86 -

The results orientation is therefore reflected in management concentration for outputs at the administration level and outcomes at the political level. The two concepts, outputs and outcomes, are of vital importance in the results and performance movement. Outputs are the direct products of the administration, the results of input allocations and activities. Examples include the number of patients treated by a hospital in a specific period of time, and with a specific level of quality, and the number of pupils educated to a certain level in a specific period. Outcomes are less tangible but more important to citizens. They are the social results achieved by the effect of outputs on society (the mix of outputs and reach in figure eight). Examples include the health of a community, or the level of education in a country or locality—difficult to measure but vital to society.

The key of results-oriented management is coordinating the achievement of outputs and outcomes, and developing output-based contracts between the administration and the executive, and of outcome-based contracts between political representatives and society (as discussed in the participatory decentralization section). Both contracts are designed to motivate and create incentives necessary for results oriented management. In countries like New Zealand, output contracts are emphasized, while other experiences emphasize outcomes contracts (the state of Florida in the United States, for example). Campos and Esfahani (2000,225) note that all policy is in essence a contract, implicitly emphasizing the outcomes contract between political representatives and their constituency, and the importance of “commitment mechanisms” in the process of democratic governance. Haggard (1999, 38) also emphasizes contracts in reform, saying that “all democratic politics can be seen as a complex chain of delegations: from voters to legislators, from legislators to executives and party leaders and from executives and legislators to government agencies.” He says that reform should comprise “a political process of re-writing the contract between politicians and bureaucrats.”

Contracts emerge from civic pressures for reform and create incentives for the adoption of results-oriented tools. Results oriented management focuses on developing the kind of relationships that facilitate such contracts, as well as the tools managers will require when such contracts motivate them to produce results. These, therefore, are the two main elements of results oriented management: Results-oriented relationships in the administration, and Results oriented tools and systems in the administration.

6.1. Results oriented relationships in the administration

Results-oriented management requires a change in the role identities and relationships of public sector role players—citizens, the legislature, the executive, and the bureaucracy (Whiteoak, 1996; ANAO, 1997; Clatworthy, et al, 2000). In traditional public organizations the lines of role differentiation occur at the frontier between political and administrative officials. Political representatives are argued to make policy, and administrators are argued to implement policy. Empirical and theoretical literature shows that role players cannot be boxed into such identities, and that interactions between

- 87 -

players are usually complex and multi-faceted (see Mountfield, 1997, for an example of role differentiation problems from the United Kingdom, and Therkildson, 2000, for an example of role conflict in Tanzania). Political representatives are often involved in implementing policy, and administrators are often involved in making policy.

Results-orientated management moves beyond the simple differentiation of roles according to who makes and who implements policy. It stresses the results that different role players are responsible and accountable for and the results-oriented relationships between those players, acknowledging that all roles involve active decision-making but that at the same time, all role players must be held accountable for the decisions that they make. Political representatives, and especially the executive, are held accountable for outcomes of the public production process. Administrative heads are held accountable for outputs of this process.

These roles are common across countries that have adopted results or performance-based government. These countries include New Zealand, Australia, the United Kingdom, Malaysia, Hong Kong, and Singapore. In all of the examples the roles are consolidated by the devolution of authority to administrators, relevant to the accountability borne by administrators. Furthermore, there are new lines of communication and accountability between role players. These lines include explicit, budget based contracts between political executives and administrative heads, in which the former promises resources and potential rewards in return for the latter’s promise of specified outputs. Similarly, political executives are held accountable within their political structures and by the electorate generally, for the smooth production of outputs and the ultimate generation of outcomes.

The results focus is also enabled by the development of active coordination mechanisms between administrators and politicians. Such coordination helps to ensure consistency between outputs produced and outcomes demanded, and to smooth the transition from process focus to results focus. New results oriented role development requires three kinds of reform intervention: (1) Focusing executives and administrators on citizens, and (2) Interpreting internal administrative roles in terms of the results orientation. 6.1.a. Focusing executives and administrators on results

Traditional bureaucratic governments are closed to citizens and do not engender involvement or participation. They emphasize administrative internal control and process, organizing public entities around ubiquitous rules instead of external relevance. Given that citizens are the source of results demand, in order to transform governments into results-oriented entities, it is important to open them up to citizen influence. In short, instead of embracing the traditional process-oriented design, governments in the modern age of results and performance need to be organized around citizens (Osborne and Gaebler,1992; Kettl,1996; Peters,1996).

In order to do this, government reform should look to develop procedures based on the perspective that citizens are active, freely choosing customers rather then passive

- 88 -

recipients of monopolistically provided services (Kaul,1997,16). In countries like New Zealand, the United Kingdom and the United States, this new emphasis is wrapped up in a new design for government process. Instead of a concentration on how services are produced, there is an emphasis on what is produced and how well it is produced. The new emphasis is a result of a public sector awareness of its ‘customer-hostile’ image and begins at the point of accessing citizens as principals and customers (Dale,1993,31; Lam,1997). This is reflected in the drive to develop new methods of accessing citizen demand for government results, and citizen perspectives about government performance. These interventions are also popular in developing countries. Governments in Malaysia, Hong Kong, Uganda, and Singapore have designed public entities so that they interact directly with citizens and are concerned about citizen-specific service quality (not just quantity), and possess the faculties to evaluate and classify citizen voice as the central form of production mandate.

Management tools designed for these purposes have been in place in the private sector for many years, and developed country governments have embraced them recently. In a study of effective government entities in developing countries, Grindle finds that “in 12 of the successful cases, management style emphasized such practices. Moreover, 10 of the poor performers had managers whose style was characterized by top-down decision-making, favoritism, lack of consultation, and poor capacity to organize or direct work. In these cases, then, management style appeared to be a factor in distinguishing good and poor performers”(1997). ‘Good performing governments’, focused on producing results, are designed to interact directly with citizens, and emphasize providing services to such citizens—seeing them as customers of their products. These governments are characterized by:1. Incentives for citizen access, and2. Tools for citizen access.

Creating the incentives to access citizens

Very few governments interact in defined ways with citizens. Governments are more concerned with obeying procedural rules. Successful examples of results-oriented government, like New Zealand, Australia, and the United Kingdom, all stand out as different in this regard. They all based their reforms on the development of a client-based culture in which organizations are structured around their citizens for effective interaction—with citizens seen as customers who need to be satisfied (Dale,1993). The United Kingdom’s ‘citizen charters,’ for example, are manifestos of citizen commitments designed to make agencies more responsive to citizen needs and demands.

In countries like Malaysia and Hong Kong, this culture has been replicated. Through the requirement that departments and agencies design their production processes around citizens, these countries have transformed the basis of their government culture and process. Public entities have to show that they engage citizens in discourse at every stage of the production process, getting input on production decisions and feedback on their output results. Such discourse leads to constant learning in the system, and also provides the basis for process improvement and service tailoring. This is because public

- 89 -

organizations have an incentive to improve their performance when they are in active dialogue with the consumers of that performance, and dialogue constitutes a line of communication into what kinds of performance are most demanded.

A variety of mechanisms characterize a customer-focused design, including courtesy campaigns, customer-care training, and comprehensive complaints procedures (Chiu,1997;Lam,1997). The following were used in Hong Kong, “to raise awareness and understanding” of “Serving the Community” (Hong Kong Efficiency Unit,1999): “Serving the Community” Booklet—describing the aims, principles, and values of the

public sector; “Service the Community” week—provided a platform for departments to promote

their services to the community and to raise awareness among staff of the importance of their individual contributions.

These kinds of devices have helped to change both the culture and structure of governments in developing and developed countries. Their influence is easy to under-rate from a distance, but the change in philosophy that they bring to the public organization (especially in situations where governments tend to ignore citizens) has enormous influence on how governments see themselves and their production activities. The following lessons arise from governments like Malaysia and Hong Kong:

Formalizing citizen/administrator interaction: Most governments, especially in the developing world, do not actively connect with their citizens. Such connection is central to the development of a results-oriented government, especially one that is bottoms-up. Without administrative devices to access what citizens are saying and demanding in terms of actual products, it is impossible to focus production on citizens. The production process becomes top-down in such instances.

Formalize the use of specific tools to access citizens: There are many simple, low-cost devices that increase access for governments to citizens. These tools should be introduced centrally to motivate their use and to facilitate a change in the production culture in favor of an outward focus. Reformers should not assume that tools will be automatically developed or used. Such use should be mandated, as in Malaysia, Hong Kong, and the United Kingdom.

Providing tools to access citizens: The Appeal of Total Quality Management

Beyond the requirement that public organizations listen to citizen voice, a results-orientation demands that governments also respond to citizen voice. This has generally meant a change in concentration from producing quantity services to producing quality services, as governments realize they are providing services which discerning customers have an opinion about. New Zealand, Australia, United Kingdom, Singapore and the United States all focus on service quality for consumers, while Hong Kong and Malaysia have formalized such concern by embracing forms of Total Quality Management (TQM).

- 90 -

These applications are designed to translate the ‘voice of the customer’ into specific managerial targets and mandates (Dale,1993;Chiu,1997, Kaul,1997; Lam,1997, Hong Kong Government,1999). These applications have proved vital in the results-oriented reforms of these governments, as they provide management mechanisms that facilitate a change in business process. They encourage an outward focus on the part of the administration, and are used to identify citizen product preferences, and to adjust production processes in favor of meeting such preferences.

Figure nine. The Malaysian Total Quality Management Program (Chiu,1997,176).

Program/ project

Inputs Activities Outputs Reach Outcomes Impacts (goals)

Orientation Traditional process orientation

Results orientation

TQM in results-oriented

management

TQM step one. Identification of customers and their needs

TQM step two. Identification of quality features of services valued by customers

TQM step four. Development of relevant work

processes

TQM step three. Formulation of quality standards of services

TQM step five. Service targets and formulation of the client’s charter

As shown in figure nine, the first step in the Malaysian TQM program involves identifying and understanding the needs of citizens (referred to as customers). The various instruments discussed previously, including customer surveys and customer campaigns, are used for this purpose. Once the needs are known, quality features such as well-defined service targets are identified, formulated, and set. The service targets then impact on the work processes in the organization, providing inputs necessary for governments to develop work processes that are conducive to quality service provision.

The service targets are then embodied in a ‘Client’s Charter,’ publicly displayed at the point of service delivery. Service targets are dynamic, adapting with improvements in capacity, and focused on providing “total customer satisfaction”(Chiu,1997,178). They spell out what services are to be rendered, when the delivery is to take place, and sometimes the cost of producing the service. Quality standards such as timeliness, reliability, accessibility, orderliness, comfort and courtesy are also reported on, as citizens have expressed interest in such indicators.

- 91 -

The entire process concentrates the government entity on the ‘results-orientation’ section of the public production process, with four out of five steps asking questions about the impacts, outcomes, reach, or outputs of production. Only one step in the TQM process sees governments examining their processes of production, and this step involves an evaluation of production processes in terms of the demands for outputs and outcomes identified and evaluated in steps one through three.

TQM has helped the Malaysian government transform its orientation from inputs to results. It is a tool that is well suited to assisting governments to listen to the voice of their citizens. Through the Malaysian version of TQM, and others like it in the United Kingdom (driven by the Customer Care Act of 1990) and Hong Kong, lessons emerge about the importance of developing a quality orientation in results oriented reform as well as the practice of developing a quality orientation:

A quality orientation can facilitate a citizen/results orientation: Governments are not normally concerned with results. They are not normally concerned with asking citizens about results that matter, either. Government production processes tend to exhibit an input, or process, bias and a concentration on how much is produced, rather than on the quality or appropriateness of production. Getting governments to think about quality helps to change this mind-set.

Develop new tools to identify and understand citizen voice: Total Quality Management (TQM) is a useful management tool for making sense of citizen voice, but any tool that links listening with evaluating and operationalizing what citizens are saying will do the job. What is important is that the tool used accesses what citizens say about services, and then transfers that voice into the discussion about what is produced and how it is produced. This is a bottoms-up approach to production.

Mandate the use of tools like TQM to facilitate client oriented organizational reform: Governments without a history of effective client orientation probably won’t adopt customer oriented reforms on their own. Governments can be motivated to adopt these reforms through legislation or administrative mandate. The Hong Kong experience of providing government-wide initiatives to motivate this kind of approach shows the importance of developing a customer orientation culture, with active endorsement throughout the organization.

6.1.b. A results interpretation of executive and administrative roles

The hierarchical, Weberian, public organization is based on a clear separation of roles between political and administrative officials. In this organization, politicians make policy and administrators implement policy in a separated though hierarchically linked process of production. Results-oriented reforms in countries like New Zealand, Australia and Malaysia also emphasize a clear differentiation of roles in the government production process, but the basis of differentiation is different to the traditional political-administration dichotomy.

- 92 -

Effective results management demands role clarity and tight lines of accountability within the context of redefined relationships between political policy-makers and administrative policy implementers (Kaul,1997,15). In this new relationship, political representatives (especially the executive) are responsible for public sector outcomes. In their election manifestos, and sometimes in even more direct form, political representatives make social contracts to produce such outcomes, via policy commitments. Examples could be promises to reduce the level of crime or to increase the general health of a constituency. Their main role is to choose which outputs will be produced, and by whom, to achieve the outcomes demanded by, and promised to, citizens (Ball and Dale,1998). Once this has been done, administrative heads are responsible for the outputs they are specified to produce, but enjoy some freedom in producing those outputs, with the understanding that ‘street level bureaucrats’ not only interpret policy but make active policy as well (Lipsky, 1981).

This mimics a private sector market in which political representatives are considered service purchasers, while departments are service producers. This market metaphor is used to describe the results-oriented roles in New Zealand. The different accountabilities synonomous with the separation of political buyers and administrative sellers is considered central to creating the incentives necessary for both to concentrate on the generation of results. The ‘market-like’ roles in places like New Zealand are characterized by:1. Results oriented accountabilities,2. Results-oriented communication between role players,3. Active coordination of role players, and4. Devolution of authority, to meet results accountability requirements.

Clarify results oriented ‘accountabilities’ in the organization

Traditionally, administrators and political principals engage in role play focused on abiding by and enforcing rules and processes (see figure ten, below). If these roles are carried into results-oriented government, the alternative results-oriented identities will be contradicted and complicated. In such instances the results-based roles have failed to work because there are conflicting signals sent to role players as to what is expected of them (“follow the rules” often clashes with “produce the goods”) and what will happen to them if they do not behave in certain ways. This means matching roles with accountability and changing important managerial tools so that they are in keeping with, not in opposition to, new results-oriented roles.

- 93 -

Figure ten. New and old public sector roles, accountabilities, and relationships Results oriented roles and relationships Traditional, process oriented roles and

relationshipsPolitical representatives determine broad policy directions on the strength of social demands for

outcomes. Political representatives are Accountable to citizens for producing these outcomes, and are the principals with whom administrative agents

contract and to whom administrative agents are accountable for outputs produced

Contracts link political leaders and the administration

Political representatives make policy but do not participate directly in the public production

process. Rather, they develop rules of process to control what administrators do within the

public sector production process.

Rules bind the hierarchy’s behavior

Central agencies help to coordinate between political representatives who are accountable for outcomes and

administrative heads who are accountable for outputs—products that lead to outcomes. Central, coordinating agencies, help to coordinate between the two by overseeing the development of output-based

contracts with administrative headsContracts permeate the system

Central agencies concentrate on applying procedural rules and ensuring that the

administration adheres to these rules. The central agencies control financial allocations on

the basis of rules guiding such allocations

Rules bind the hierarchy’s behavior

The administration is given resources to produce Outputs that will ultimately fulfill the political program,

delivering outcomes promised to the electorate. The administration is accountable for the outputs they

produce, and enjoy the managerial freedom Necessary to produce these outputs

The administration carries out the programs of political policy makers. It concentrates on

adhering to rules of process and procedure and not on providing results. This is in keeping with the perspective that the administration

implements policy, and does not actively shape or influence what policy looks like

New roles, as defined in terms of results, are required for political executive and administrative role players in the results oriented management structure:

Identify executive accountabilities for outcomes: In the results-oriented system it is important that all the main players have some direct connection with results. The first section of the ROME puzzle is designed to get citizens to directly express their demand interest in results. Similarly, accountability for outcomes needs to be identified with the role of political representatives. In order to do this, the electoral process should be structured to include some form of ‘social contract’—in which leaders are elected to limited terms on the strength of published promises for specific outcomes. In this way, the political executive is given a clear mandate to produce outcomes and is made to commit to a mandate for such. These outcomes are clearly communicated and are measurable and open to evaluation—and lines of political accountability exist to ensure their evaluation and reward and redress on the strength of such.

Identify administrative accountabilities for outputs: The second lesson is similar to the first lesson. Administrators need to be clearly identified as the role players

- 94 -

accountable for outputs. This involves clarifying what outputs are expected, as well as the measuring and evaluation tools that will be used to monitor such—as well as ensuring that lines of administrative accountability exist to facilitate evaluation, reward and redress on the strength of such.

Promote communication necessary for results-oriented accountability

In traditional government structures, just as roles revolve around rules, communication and accountability is also rule-related. This is evident in figure ten, on the right hand side. The political representatives develop policy and attempt to control the administration through rules binding their behavior. Accountability is measured through adherence to rules and control—with important questions focusing on whether money is spent on that which it was meant to be spent on, administrators were appointed according to the rules of appointment, and the like. This system has proved more successful in some settings than in others (Christensen and Laegreid (1997) show it is successful in Sweden, for example, while Schick (1998) argues that rules have not been effective in developing countries). In all settings, this approach concentrates the public organization on process instead of results.

As expected, therefore, the results-oriented organization has different relationship and accountability lines. Kaul emphasizes “new management structures” that enhance accountability by tightening task definition, and through “measurement of performance, devolution of resource control, strengthening monitoring, and clarifying incentives” (1997,15). In this new system, executives are accountable to taxpayers and service recipients for outcomes produced through political and social lines: policy commitments, electoral promises, and in developing countries especially, through structures of social evaluation and response (Ball and Dale,1998).

Because of this final accountability, executives have complete authority over the broad allocation of resources—how much money goes into each department’s balance sheet. In allocating money to departments (through coordinating agents) the executives demand contractual promises, in terms of which administrative heads commit to producing specific outputs requested by their political representatives. Within the balance sheets of individual departments, the head of the department has authority over the resources but is held accountable for the outputs those resources were allocated to produce. This accountability is formalized through budget-specific contracts for resources. This is the New Zealand model, sometimes called contracturalism (Schisk,1998).

In Uganda, there are no explicit contracts in place yet, but the budget acts as an implicit contractual device, binding administrators to expected outputs in return for publicized allocations. The agencies are given some autonomy to pursue objectives without excessive control, but are kept in check by public record of disbursements; “monthly transfers of public funds to districts are now reported in the main newspapers and broadcast on radio…transfers to primary education (are) displayed on public notice boards in each school and district center.” The results are clear—in a survey in 1998 the Ministry of Education “found major improvements in the flow of funds”

- 95 -

(Reinikka,1999). There are four outstanding lessons for the development of effective accountability lines.

Emphasize a results accountability framework, and new relationships: Lines of relationship and accountability link role players to facilitate a smooth system of government production. Wherever possible, the basis of the accountability lines should be outputs and outcomes produced. These are the results of administrative and political processes. Concentration on such results motivates administrators and political officials to concentrate on these results.

Specify the contract/accountability basis in advance: Administrative accountability processes are effective if the contract basis is well specified in advance, and published. In New Zealand, important lines of accountability are developed in terms of purchase agreements, or contracts, that spell out the outputs that are to be delivered. In Uganda, the ‘contracts’ are specified in terms of financial probity with regards the published inputs allotted to agents. In Malaysia, inputs and promised outputs are specified and published.

Supplement internal lines of accountability with lines involving society: In developing countries especially, lines connecting society to political and administrative officials are vital to effective accountability. Countries like Uganda and Malaysia show that society has a vital role in evaluating results and holding government accountable for results. When ‘contracts’ are published, the fear of social ostracism, especially at the local government level, is an effective line of relationship and accountability.

Specific ideas for developing lines of political accountability: Political contracts are difficult to specify, primarily because outcomes—for which political representatives are accountable—are difficult to measure. The idea of a contract-based relationship should be based on some performance requirement, however, and should be both limited in time (there should be regular election cycles), and enforceable by society. This requires a social contract developed in a political system where citizens have influence on their political representatives, a situation demanding social organization and mobilization as discussed in the first section of ROME—the basics of ROME.

Specific ideas for developing lines of administrative accountability: Administrative contracts are easier to develop than political contracts. Outputs are easier to evaluate, and administrative appointments can be made according to performance more easily than can political choice. These are the main reasons why results-oriented government in New Zealand and other successful governments concentrate on results in terms of outputs. In these countries, administrative accountability is consolidated by the move from life long employment to medium-term contracts for administrative heads, renewable on the basis of performance in terms of output targets. These targets are specified in the performance-based budget (where performance is specified in terms of outputs, not outcomes). The budget is the main contracting device.

- 96 -

Coordinate the administration for consistency in results

A central coordinating institution has proved important in all governments actively developing results oriented management (Schick,1998). The coordinating role is often played by the central ministry of finance, underscoring the importance of financial processes in the results government. Coordination is needed to ensure that outputs produced relate well to outcomes demanded. The coordinating entity also provides a central location for long term planning and visioning (Kaul,1997).

An example of a coordinating mechanism comes from New Zealand where the executive specifies and publishes the outcomes the government wishes to achieve. Coordinating committees are established, made up of senior government officials from departments with broad interest in government activity. Their role is to ensure that policy options are developed in a coordinated way across government. In governments like Malaysia and Hong Kong coordinating institutions involve specialists with citizens, administrators and politicians. In Hong Kong, the Efficiency Unit has an important role of coordinating, publishing, and evaluating the development of results reforms. The following lessons are learned from these experiences.

Coordinate between outputs and outcomes to ensure consistent production: If a system concentrates some members on outputs and others on outcomes, that system needs a communications valve—the coordinating mechanism. In countries that have adopted the results oriented approach a dedicated institution fulfills this role. In most countries this institution is a direct spin-off from the central finance agency.

Develop skill and capacity in the coordinating institutions: A coordinating institution needs effective capacity and skill to coordinate production. Coordinating between outputs and outcomes requires planning abilities and capacity related to project evaluations. Governments developing coordinating institutions generally focus such around high capacity specialists providing advice to political representatives.

Clarify who the coordinating institution is accountable to : The coordinating institution receives its mandate from politicians, accountable for outcomes. The coordinating institution is not designed to advise on what outcomes government should concentrate on producing. That decision is a political decision, reserved especially for the executive (Ball and Dale,1998). The coordinating institution received its mandate from the executive to advice them on what outputs should be produced to realize outcomes.

Clarify the role of the coordinating institution: The role of the coordinating institution must be clearly specified, and such institution must not be allowed to dominate the production process. The coordinating entity acts as a coach, not a captain. It works upfront before the production decisions are finalized to ensure that such decisions are the right ones. Once these decisions have been made, the coordinating entity works in the background, if at all. It does not carry accountabilities for either outcomes or outputs, and does not directly produce.

- 97 -

Devolving authority to administrators, as required by accountability profiles

Linked to the new lines of accountability in the results-oriented government structure, is the need for authority devolution to administrative heads. The argument for devolution is simple: “If enhanced accountability has to act as a pressure for performance improvement, it must be matched by enhanced managerial authority. Devolving responsibility to senior administrators allows them the financial and procedural latitude necessary to deliver the outputs for which they are held responsible” (Kaul, 1997, 18).

In practice, devolution is focused on breaking the traditional government controls over how administrative agents provide services—focused on mandated processes and bound by fiscal and personnel-based rules of procedure. Reforms focused on devolution aim to give authority to service providing departments and agencies so that they can provide services in the best way they know possible. In granting such autonomy, the agents are, however bound by contracts for results—with the emphasis changing from influence over how they produce to influence over what they produce.

The devolved government example is provided by New Zealand, where a chief executive for each agency or department producing services is contracted to produce certain outputs related to strategic outcomes. It is understood that the chief executive’s performance is maximized only when she is allowed to be responsible for the incentives, processes, and costs involved in performance. In short, she must have the organizational ability to perform if she is held accountable for performance. In traditional governments the concentration on control limited both the incentive for performance and constrained the ability to perform—as managers could not hire or manage with a results or performance focus. The New Zealand model breaks this process stronghold and focuses the organization on results. It is replicated in various forms in a number of countries, from Malaysia to Uganda, where contracts are generally embodied in budgets and often published so that citizens can vet and respond to performance. Lessons for results-oriented management concentrate on the need to devolve autonomy over financial management, personnel, and costs in production.

Give managers autonomy over personnel: In a study of public sector entities in developing countries, Grindle (1997) finds that organizations with some autonomy in personnel matters have greater capacity to set performance standards and develop a results orientation. In New Zealand Chief Executives have authority for departmental staff, with the power to determine the conditions of employment and salaries of their employees, a situation reproduced in Singapore. In Malaysia the New Remuneration Scheme allows flexibility in compensation (Chiu,1997). In all of these cases, giving autonomy to managers has facilitated management for results.

Give managers autonomy over finances (within budget): Managers should be given charge of their financial resources, and the management of those resources. In New Zealand, this flexibility is evident in the freedom Chief Executives have in deciding

- 98 -

what money is spent on, and in the fully delegated authority to design and acquire financial systems. In both cases, the pre-existing controls over the details of financial management have been replaced by commitments for outputs and the provision of standardized financial reports (which will be discussed later in this paper).

Give managers authority over resource purchases : Departments need autonomy so that they can concentrate on minimizing costs of production while meeting output commitments. In New Zealand, Chief Executives have authority to purchase inputs from where they decide is most appropriate and efficient. They are not required to use central services for basic services. They are, however, required to evaluate costs and report on costs according to standardized procedures (which will be discussed later in this paper).

6.2. Providing the tools required for a results-oriented response to citizen demand

Managing for results is not only about re-focusing relationships and roles. It is important to ensure that processes within which role players operate are also appropriately focused. In traditional public organization structures the focus is on process, not results. The process focus is underscored by an emphasis on adherence to rules of internalized procedure, costing mechanisms that concentrate on short-term flows instead of long term and competitive commitments and products, and budget systems that classify behavior in terms of line items and inputs. Results-oriented managers in countries adopting such reforms operate in systems different to these traditional systems. The new systems are based on the coordinated application of management tools more commonly used in the private sector. These tools permit comparison of business processes and results, effective costing of production outputs and not just inputs, and the effective evaluation of public organizations in a long term, strategic sense.

The adoption of these kinds of tools is vital in moving towards a results-orientation. In her research into effective public sector entities in developing countries, Grindle (1997) finds that these kinds of ‘private sector’ management practices were evident in the ‘good performers.’ In particular, she found that results-based planning and incentive schemes had a very positive effect on public organizations. Of the 26 case studies examined, “12 out of 14 good performers set and applied performance expectations while 10 of 12 poor performers set no such standards for their employees. In these latter cases, employees were left to conform to the performance expectations for the public sector more generally—standards that were usually very low” (1997,481).

6.2.a. Tools for results oriented management

While results oriented reforms are often centered on the application of tools, ROME introduces these elements as optional reform instruments that managers should request. Essentially, they provide the best ways of responding to social and institutional demands and incentives for reform. They do not in themselves constitute reform interventions. As such, results oriented tools should not be applied simply because of ‘fashion appeal’—

- 99 -

rather, they all have identifiable value in the move to results-oriented government, and in most cases there are examples of public sector application that suggest their ‘fit’ to the sector.

The management tools specified here include the use of budgets that show allocations in terms of results instead of inputs, benchmarking to define results in a comparative sense, and activity based and resource based costing to evaluate costs for outputs, not inputs. Also included is the introduction of capital charging to account for full organizational value, and a move to private-sector type accrual accounting and the production of standardized and readable financial reports to ensure that all commitments are accounted for, and that results are reported on as well as inputs. Finally, in order to successfully focus managers on results, governments need to provide managers with the necessary tools and capacity for managing for results. This includes helping managers develop and adapt planning tools like the Balanced Scorecard to their organizations, and implement incentive schemes to encourage a performance culture within individual organizations.

The tools that governments should make accessible to inquiring managers are:1. Performance-based budgeting,2. Benchmarking,3. Activity-based costing,4. Full costing,5. Accrual accounting and full costing, and6. The balanced scorecard.

Performance-based budgeting in ROME

Traditional budgets focus on inputs. They show how much money is being allocated to departments to cover specific types of expenditure—notably line-items like salaries. They do not provide information about results, either expected or realized in previous periods. They also seldom provide any information as to why allocations are as they are—although the general practice is to allocate resources in an incremental way. This involves adding to a base of funds so as to cover an entity’s existing expenditures (assuming they continue) and provide some resources for new expenses (the increment).

This budgeting approach is obviously limited and limiting, especially in terms of facilitating resource allocation or a results-orientation. It was never designed to do either of these two things, with a focus on providing controls to government finances—an instrument for holding public entities accountable for funds received from citizens. But in the modern era, the input-based, incremental budget is no longer appropriate. The definition of ‘accountability’ has been expanded to incorporate demand for value for money in governments, and calls for governments to increase efficiency and responsiveness have led to budget reform initiatives across the globe.

In most instances, these reforms involve a move towards results-based budgeting, as shown in figure eleven.

- 100 -

Figure eleven. Moving from input-based incremental budgets to results-based budgets

Program/ project

Inputs Activities Outputs Reach Outcomes Impacts (goals)

Budgeting approach, and the four steps to a results-

based budget

Start. Input-oriented, incremental budget (the

conventional budget)

Step 1. The ‘Functional’ budget (Ghana)—shows

broad fund destinations, for example “infant health care”

Step 3. The performance-based budget

(New Zealand)

—shows more specific fund destinations and

aims, for example “lowering the infant

mortality rate by 1% a year”

Step 2. The Outcomes-basedbudget (exemplified in

Friedman, 1995)—shows specific fund

destinations and aims, for example “improving the

health of infants”

The first step away from these traditional systems involves identifying the functional destination of funds. This involves identifying outcomes that matter to citizens, a process that should be inclusive and participatory, based on hearing and responding to the voice of the electorate (in keeping with the bottoms-up concentration in ROME). In Ghana, for example, a functional budget (linked to a Public Investment Program (PIP)) shows funds in terms of programs, collected into categories by similar activities and focused on achieving broad social objectives (Assibey-Mensah, 1998). Citizens know what the government is doing or expects to do with public funds—in broad functional areas related to targeted outcomes, for example “infant health care.” The functional budget does not provide information on results but provides a step in this direction by identifying the functions targeted by government.

Outcomes-based budgeting is the next step towards results-based budgeting (Martin,1997). It builds on the identification of broad outcome targets to specify what those targets are and to provide measures of such targets. Once identified, outcomes are disaggregated into a set of indicators, which, when taken together, provide a view of government outcome performance, for example “improving the health of infants.” Data on these indicators can be used to establish a baseline against which to track progress, with the outcome budget providing information about the final outcomes government aims to produce, with specific measures and times for delivery.

Specifying outcomes is important for consolidating the social contract between citizens and political representatives. It is not effective in creating a results orientation for the administration, however, as the goals are long-term and often prove difficult to measure. Performance-based budgeting, as developed in New Zealand, concentrates on moving the outcome goals one step backwards, to output-related objectives. An example is

- 101 -

“lowering the infant mortality rate by 1% a year.” This goal is short-term, and is more easily evaluated and measured than the outcomes goal. It is also directly related to the role of the administration, and provides their results-orientation (Foltin,1999 stresses the importance of budgeting for outputs in terms of the incentives created for administrative performance).

In the performance-based budget, government presents the purposes and objectives for which funds are requested (in terms of measurable outputs), as well as the costs of achieving these purposes and objectives (Xavier,1998). Indicators and measures of such purposes and objectives are published with the budget so that the budget provides a complete contract between political representatives (who ‘hire’ the administration) and the administration (who receive budget money to produce specific products). If the outputs are not produced as promised, the administrative head responsible does not receive funding in future periods, and faces replacement by another manager or even provider.

There are five lessons for developing this kind of results-based budget.

Develop program-based, functional budgets as a first-step to results-based budgets: Most developing countries start the route to results-based budgeting from an input-based, incremental budget. It takes steps from this format to a performance-based budget. Between the two formats, the development of functional budgets provides a good transition—constituting a link between the budget and outcomes that matter to citizens.

Develop outcome-budgets to facilitate ‘social/political contracts’ : The budget is the most important document in government and is well suited to the development of contracts. Developing on the identification of functions to which finances will be allocated, the budget should also incorporate specifics about the outcomes being targeted—so as to provide the basis of a ‘social contract’ for political representatives.

Develop output-based budgets to facilitate political/administrative contracts: Performance-based budgets specify output contracts between political representatives and administrative heads. The budgets should be well detailed, so that administrative heads are specifically committed.

Stipulate expected results, indicators of results, and potential responses to results: Results are complex concepts—they should be specified in terms of expected produce, indicators of the product, time allowed for production, and the potential rewards and redress options that can be used to enforce contracts.

Develop a set of criteria on which results/ performance measurement is based: In order to make the budget-based contracts credible, the performance measurements should be clear, developed through a process of interactive negotiation, and relevant to administrators and political representatives.

- 102 -

Benchmarking in ROME

In order to manage and budget for results, managers need the ability to compare their results. The question of “how are we doing?” is always based on an earlier question of “what are we comparing ourselves against?” In Malaysia, benchmarking has been adopted to make comparisons with similar organizations that carry out the same functional activity, and in so doing to see how an agency’s results measure up (Chiu,1997). Benchmarking is a management technique suited to results oriented management—“essential because it provides a standard of reference by which a government entity can measure or judge performance”(Foltin,1999,42) . On a simple level, as in Malaysia, it involves comparing current performance with a meaningful point of reference and using the comparison to guide management improvement.

Three principle forms of benchmarking exist, corporate-style benchmarking, targets-style benchmarking, and the use of performance statistic comparisons as benchmarks (Ammons,2000). In the first of these three, corporate-style benchmarking, processes accorded ‘best practice’ status are identified in competing organizations and used as ‘benchmark practices’ for management to pursue. Results-oriented managers look to high performers and then copy their production processes—as in the Indianapolis managed competitions where city service providers were given contracts to produce service on the proviso that they adopted particularly noteworthy practices of private sector leaders. In the second variant, targets-style benchmarking, current conditions are contrasted with conditions embodied in future plans—especially related to quality of life indicators and sometimes specifically related to service efficiencies. This is largely the way results targets are set in New Zealand, where agencies set goals and are evaluated in terms of such. The third form of benchmarking, performance comparison, is that practiced in Malaysia (as well as being the most common in the public sector). It involves comparing performance expectations and results of an operation with reputable standards or norms achieved by market counterparts.

Whichever method is used, benchmarking helps to identify the shortfall between current performance and desired, compared, or benchmarked, performance (Rivenbark and Carter,2000). It provides a device for both prescribing direction (performance goals) and for assisting in diagnosing the reasons for gaps in performance. There are four lessons for effective benchmarking.

Structure the benchmarking initiative: Comparison is the basis of results management. In developing country governments, however, comparison is not part of every-day management. It will not be part of such management unless such capacity is developed and specifically located. Hong Kong and Malaysia invested in developing benchmarking capacities in their central coordination institutions—in Hong Kong this is the Efficiency Unit, an appropriate name. In New Zealand departments and agencies have benchmarking responsibilities themselves, setting goals and monitoring performance in terms of such goals. These two models differ in terms of the location of benchmarking responsibility, but provide teach that such responsibility must be established and enabled.

- 103 -

Decide how to compare results up front: Benchmarking approaches differ in terms of what results are evaluated against; best practices, internal goals, or external service leaders. Different governments stress different points of comparison, matched to their specific purposes. In decentralized governments comparative service benchmarks are possible and preferable, motivating competition and innovation between decentralized entities. In centralized settings, especially in developing countries, it is difficult to compare internal performance with other providers and manageable, internal goals are the preferred form of benchmark.

Identify points of comparison, and develop ways of tracking them: Once the type of benchmark has been chosen, governments should specifically identify benchmarks for comparison. This involves setting targets for production that are measurable and open to evaluation. Without specific points of comparison, agents could compare their performance on a sliding scale according to how they measure up.

Be as specific and comprehensive as possible: In comparing current conditions with benchmarks (whether ‘best practice’ processes, targets, or compared performance), public organizations need to consider service level, volume, quality, and cost and efficiency. To do this, it is important to go beyond budget figures and gross expenditures and concentrate on developing an accurate understanding of real costs and value. This step is often poorly developed in the public sector where cost accounting issues are generally poorly dealt with.

Activity Based Costing in ROME

Good cost accounting is central to good performance management. Using benchmarking to develop performance criteria and diagnosing process improvements in terms of such, for example, calls for a thorough knowledge of the costs associated with processes, targets, or compared performances. There is a need “to evaluate the costs of producing outputs and outcomes on a continuing basis in order to evaluate performance and allocate resources” (Rodriguez, 1995). Governments generally lack an accepted methodology of cost measurement and evaluation, however, particularly one that allows for cost comparison of output (Tierney, 1994).

In order to facilitate results comparison between public bodies and market leaders or targeted conditions (without which competitive incentives cannot develop) it is vital that governments adopt devices that measure costs in a way that facilitates broad comparison of results and performance (between public organizations and private organizations, and between production states now and production states in the past and future). Examples of such cost accounting approaches are Activity Based Costing (ABC) used widely in the United States, the Micro Accounting System (MAS) in Malaysia, and Resource Based Costing (RBC) currently being introduced in the United Kingdom (Brown, Myring, and Gard, 1999).

- 104 -

Activity-Based Costing (ABC) is the most common device in results-oriented governments in the developed world (Simpson and Williams,1996; Ho and Kidwell,2000). It has been used widely in local governments especially, and has proved particularly popular in entities attempting to compare their performance with private sector standards. It involves relating input costs to activities within organizations, and then relating the activities to the factors that drive costs—generally the output objectives of the organization. An example of ABC is relating the input costs in a university (salaries, and so forth) to the activities of the university (lecturing, research, and administration), and ultimately to the outputs and objectives of the university (PhDs produced, Bachelors degrees issued, journal articles published). ABC gives the university a view on the cost per unit of output—it costs X dollars to produce a PhD, and Y dollars to produce a Bachelors degree, and Z dollars to produce a publishable paper (developed from Goddard and Ooi, 199; Mitchell, 1996; Granoff et al., 2000). In this way, as shown in figure ten, Activity Based Costing facilitates the cost evaluation of organizational results—in terms of activities and the outputs driving such activities.

Figure twelve. Activity-based costing and results-oriented government

Program/ project

Inputs Activities Outputs Reach Outcomes Impacts (goals)

Cost Orientation

Traditional cost

orientation

The ABC process

ABC step 1. Relate inputs to activities, to

cost activities

ABC step 2. Relate activities to outputs (cost drivers), to cost

outputs

The various approaches to costing activities and outputs yield two important lessons for results-minded reformers.

Develop methodologies for cost evaluation that suite the specific results-orientation: There are a number of approaches facilitating the costing of results. Activity Based Costing is popular because it generates per unit costs for outputs, relevant in performance-based budgets. Other costing approaches concentrate on outcomes. The appropriate approach depends upon the kind of results that specific governments concentrate on.

Focus costs on activities and outputs, in order to gauge results in terms of outputs: In most results-oriented governments the concentration is on outputs. Concentrating on costing activities and outputs provides information necessary to evaluate results at the administrative level—the level at which performance-based budgeting is traditionally concentrated.

Full costing in ROME

- 105 -

Tied to the need for costing tools that relate costs to production results, particularly outputs, is the need for costing tools that evaluate the full costs of such results. This need arises because governments generally underestimate their production costs, by concentrating on direct cash and either apportioning other costs (like overhead costs) generally, or not considering other costs (like capital costs) at all. New Zealand provides an important lesson for results-oriented reform in this regard: Defining and monitoring purchase and ownership performance requires information about full resource cost, including the consumption of assets and the opportunity cost of capital, and about assets and liabilities, their utilization, and the return being generated. Without such information it is impossible to compare relative efficiencies of performance (Ball and Dale,1998).

In order to ensure that management can compare their results with benchmarks, it is necessary that they have full costing data. This means accounting for direct costs, indirect costs, and capital costs—full costing—in ways that can be easily related to results that are meant to drive behavior. Full costs should be reported on in such a way that they facilitate comparison and reporting, as well as internal diagnoses of the cost of producing results that matter. This means considering the appropriate cost of overheads in production (a concentration of Activity Based Costing), and the explicit evaluation of capital in the production process (which the New Zealand government concentrates on in their Capital Charging initiative, also in place in the United Kingdom).

Full costing not only provides a more accurate picture of relative production efficiency; it also plays an important role in developing incentives for efficient results production. If administrators are not required to accurately measure their overhead and capital costs, they lack the incentive to manage these resources efficiently and effectively. When these costs are included in their management decisions, managers have an incentive to actively manage how much capital they use, and to strategize about latent capacity.

Experience with full costing provides the following two lessons for results minded reformers.

Explicitly consider overheads in costing results: Overheads are often poorly included in public sector costing. These overheads often constitute an important cost category and demand better evaluation. Whatever cost device is chosen, it should provide an accurate assessment of the overheads used in producing specific results.

Explicitly consider the cost of capital in costing results : Kaul says that “capital charging underlines the significance of good estate management” (1997,120). Capital investment has been a free good for public service managers because it has not entered into costing systems. This has not encouraged the responsible use of assets and needs to be redressed through active capital charging.

Accrual accounting and results reports in ROME

- 106 -

Traditionally, government accounting systems are based on cash (or modified cash) accounting. In countries like New Zealand, the United Kingdom, and Malaysia, this approach was identified early on as incompatible with a results focus. All three countries have adopted accounting approaches consistent with private sector GAAP, based on the accrual system, and directly including information about departmental aims, outputs, and results. The difference between accrual and cash accounting is, simply, that the former registers financial flows when they are accrued whereas the latter registers such flows when cash is exchanged. An accrual system registers commitments and expenses, a cash system registers actual cash payments and expenditures.

The shift to accrual accounting is desirable in that it provides superior information from managers to owners—especially regarding general performance, and especially when augmented by statements of performance. Accrual accounting gives a more accurate picture of organizational value and commitment, especially concerning the future. Cash accounting has a number of major drawbacks. It does not accurately represent the amount of resource usage—for example, the value of capital is not well considered over time. There is also a failure to take into account future commitments or liabilities. Cash payments concentration alone can also lead to a deterioration in fixed assets, and poor management of fixed assets (off balance sheet often means off managerial mind). It is not at all strategic, as the concentration is on resources moving in and out of the system, rather than the outputs produced by those inputs. There is also a distortion of incentives as managers underestimate the costs of programs (capital is not current so is left out) and managers are encouraged to spend the full cost of their appropriations.

Private entities do not follow the cash approach, and public entities cannot be compared with the private sector if public entities have a non-comparable accounting method in place. On the basis of accrual accounting in New Zealand, each department provides a full set of financial statements to the executive and treasury on a monthly basis. They also produce a Statement of Service Performance, outlining the outputs produced versus the outputs agreed and giving information about purchase performance much as a private firm would. Because statements are uniform, the government can sum them up to provide holistic financial statements. These are published monthly and annual, audited financial statements are presented to parliament. Government accounts thus show the net worth of government.

With this information, it is envisaged that government will be able to make resource allocations more effectively, and with a definite results-orientation (Jones, Likierman, Gillibrand and Hilton). Cash accounting does not provide this kind of information. There are s lessons for financial reporting in a results-oriented government, all linked to the basic reporting documents required.

Report on financial performance—including all commitments: The Statement of Financial Performance (the Operating Statement or the Profit and Los Account) shows the financial results of an organization’s activities. Were revenues sufficient to cover all expenses? This Statement should reflect the commitments of the

- 107 -

department, developed according to the accrual method so as to facilitate balanced planning and strategy development.

Report on financial position—again, including all commitments: The Statement of Financial Position (the balance sheet) shows all financial items the organization owns and owes at a certain point in time, providing insight into the organization’s ability to pay its entire debt. Again, this should be developed according to the accrual method.

Report on cash flows : A Statement off Cash Flows is also provided to reconcile the accrual accounts with changes in cash balances. It provides a picture of cash inflows and outflows which is important in government to show liquidity.

Report on performance/results : A Statement of Service Performance is important to provide information regarding the results of the organization. This statement is seldom produced in traditional governments and should be developed from the output contract implied in the budget, showing production quantities, quality, costs, and timeliness.

The balanced scorecard in ROME

One of the main criticisms of results-oriented reform in developing countries is the lack of capacity in such settings (Schick, 1998). Developing country governments are not known for high levels of capacity, and it is felt that the results-orientation will be beyond such governments. What is interesting, is that New Zealand, the leader of the results-oriented approach, also faced capacity problems (Aitken,1997; Matheson,1998). They had to work hard to develop capacity, as well as incentives for administrators, so that the results-orientation was enabled in their government.

The lessons of New Zealand’s experience are found throughout this document. Coupled with experience in countries like Malaysia, lessons also emerge about developing and inspiring capacity to manage for results through the introduction of new management systems and the use of performance-based incentives to focus organizations on producing results well. In relation to Malaysia’s initiatives, Chiu (1997) emphasizes upgrading the capacity of the administration through the development of human resources, technology, and results-oriented systems and procedures.

Important in all of these methods of organizational improvement is the development of tools and practices that assist management in transforming their organizations from a one-dimensional control orientation to a multi-dimensional results orientation in which customers need to be considered, budget contracts now have to be fulfilled, services demand improvement, and internal actors and processes matter more as contributors to results instead of mere cogs in an ever-turning wheel of production.

One tool that has been used in the public sector to combine the dimensions of the new results focus is the Balanced Scorecard. The scorecard relates well to Total Quality Management and Activity Based Costing tools, and provides a picture of the entire

- 108 -

organization—its inward and outward foci. It has been explicitly described as a device for operationalizing a performance and results orientation (Kaplan and Norton,1992). It emphasizes four perspectives of performance, to help public managers understand the complexities of the results-oriented organization: Financial performance, Customer-related performance, Innovation and learning performance, and an Internal Business performance.

Figure thirteen. The Balanced Scorecard and Results-Oriented Management

Program/ project

Inputs Activities Outputs Reach Outcomes Impacts (goals)

Areas of concentration

in the Balanced Scorecard

The Internal Business Perspective The Financial Perspective

The Innovation and Learning Perspective

The Customer Perspective

Figure thirteen shows how the scorecard unifies perspectives at every stage of the public sector production process, helping managers to see their roles in a comprehensive but concise fashion. The scorecard allows managers to evaluate how they are doing in terms of the following process:

1. How do we look financially? Are we managing to fulfill our budgetary obligations?

2. How doe we look to our customers—both political principals who buy our services, and citizen customers? Are we producing services of the type and quality that satisfies our customers demands and that fulfils our contractual obligations?

3. Can we continue to improve and create value (the innovation and learning perspective)? In particular, does our interaction with customers give us any ideas, or reason, to improve the way we do things?

4. How should we focus the internal organization to meet customer needs? How much do the internal processes cost? How can internal business be focused towards achievement of financial and customer goals?

Alongside this kind of approach, spelling out the dimensions of public management in the results dispensation, managers should have access to incentive packages for their employees. In Malaysia, incentive packages are built into the management role (Kaul,1997,20). The Malaysian New Remuneration System ensures that skills and, in particular, personal achievements are rewarded in organizations, and the Public Innovation Award provides an incentive for innovation. Together, these tools help managers to motivate a results-orientation within their organization.

There are three lessons for capacity development through the introduction of tools like the Balanced Scorecard and of incentive programs focused on results.

- 109 -

Develop capacity through facilitating a new results-orientation: Developing countries need not see themselves irreversibly weak in terms of managerial and operational capacity. Even countries like New Zealand did not have effective capacity for results-oriented management. Such capacity is developed through demand, and demand is developed through the introduction of management techniques that facilitate and engender a results-orientation. In a number of governments at various levels, personnel have exhibited previously unseen creativity and innovation once given authority over their production processes, and access to management tools suited to results management.

Link external results-oriented goals with internal strategic management: One of the

challenges of results reform is to change bureaucratic cultures and management behavior from one dimension concentration to multi-dimension concentration. Management needs tools to make this change. The Balance Scorecard is such a tool.

Link incentive programs to individual and group performances: To help change the culture, and to entrench a results orientation in government, management should be encouraged and freed to introduce incentive schemes into their organizations. In developing countries like Malaysia, monetary incentives have not proved effective motivators of public administrators, but opportunities for skills development, professional travel, and public honor are all used successfully to motivate high levels of performance.

6.3. Conclusion: Facilitating results-oriented and responsive administration

Many governments around the world have implemented one or more of the tools and mechanisms discussed in this section. The literature shows, however, that implementing the reform tool does not mean success in achieving reform results. Indeed, results-based reforms have failed in governments from Malawi to Tanzania to Thailand, to a variety in Latin America, and even to a number of states and local governments in the United States because they have concentrated on individual tools instead of the combination of factors presented in this second section of ROME.

As with participatory decentralization, Results-Oriented Management involves an interaction between, and combination of, reform elements and practices. Implementing one practice without the others leads to stand-alone, and poorly used, tools. Implementing all the tools together (especially when implemented in conjunction with the basics of ROME and Results-Oriented evaluation) yields substantial gains and a change in government orientation.

- 110 -

The Results-Oriented Management section of the ROME model is shown in figure fourteen. As this section relates to other ROME sections, it starts with citizens occupying the main role in the government process. Assuming that the basics of ROME are in place, these citizens are the focus of government management. Political and administrative heads develop explicit (election-based) and implicit (social based) contracts to produce goods demanded by citizens. These contracts are consolidated through outreach programs by government, and the development of tools like Total Quality Management to discern demand and facilitate an organizational ability to respond to demand (the contract and TQM elements are shown by the arrows on the right-hand side of the graph, numbered 6.1.a.).

These social contracts underpin the bottoms-up approach to management implicit in Results-Oriented Management. They are consolidated by new relationships and lines of communication in the government production process, which see political representatives accountable for outcomes and administrators accountable for (and enjoying autonomy in the production of) outputs. These new relational lines are shown by the arrows running from citizens to the executive and from the executive to the administration (and by the number 6.1.b.).

Finally, Results-Oriented Management notes the importance of introducing tools and mechanisms that focus managers on results and give them the necessary systems and capacity to generate results. These include performance-based budgeting, various costing and financial reporting tools, and the Balanced Scorecard, which combine to provide an effective arrangement of managerial tools (shown as 6.2).

- 111 -

Table five. Management tools in ROMEQuestion for results-oriented management Management

toolThe entire process

driven by a citizen focus:

Contract information—what is the final product we must produce and what do we receive to produce such product?

Performance-based budget

All these tools are connected to Total Quality Management and such devices used to create a results and participation culture, and work effectively where roles emphasize results.

How do we know how we are doing in terms of the contract, and in terms of other producers from whom we can learn?

Benchmarking

How much does it cost to produce such product (the complete cost)? How can we produce the product better so we can be sure of meeting and exceeding our contract obligation and receiving rewards?

Activity-Based Costing (and others)

How do we report our results? Full reporting using accrual accounting

How do we manage the new reporting, production, and contract obligations we have, as well as run a citizen friendly administration?

Balanced Scorecard

Results-Oriented Management is effective when these management tools are implemented in conjunction with the other elements described. In most reform instances reforms are more piece-meal than this, however. Tools like performance-based budgets are introduced into governments with no effective benchmarks or devolved autonomy in the administration. This results in the performance-based budget being largely ceremonial, as administrators have neither the room to be accountable for their outputs nor the means for evaluating their performance in terms of output requirements. Similar problems that emerge when pieces of the Results-Oriented Management section of the ROME puzzle are introduced without others include:

Spain recently attempted to devolve its public sector and introduce results-based management devices. It failed, however, to devolve responsibility for personnel and other vital inputs to administrative heads. The result has been a contradiction of policy—with managers expected to pursue outputs targets but unable to do so. The two agencies with devolved authority over personnel have been able to transform their management approach (Parado-Diez,1998).

In a number of countries results-oriented management does not begin with citizen interaction and questioning. There is no Total Quality Management scheme or method of accessing citizen demand. Output and outcome targets are largely developed by administrators. The systems have a results dimension to them, but are hardly true results-based governments (they are still top-down).

A number of Latin American countries have introduced results-based budgeting in the last few years. Reports suggest, however, that administrative and executive roles are intermeshed, limiting the ability to locate accountability for outputs with either one of

- 112 -

them. Furthermore, public entities receiving money through performance-based budgets still produce cash-based financial reports. These of not facilitate strategic or results-oriented administration.

Many of the states in the United States say they are running performance-based budgeting systems, but very few have devolved authority to managers to allow them to concentrate on results. Furthermore, many of the states still account for inputs instead of activities and outputs, and have no planning tools like the Balanced Scorecard. Most of the recent literature shows that most states may say they are implementing Performance-based budgeting, but that these problems and others are limiting the manifestation of results-oriented management.

This section argues that the results oriented management section of ROME involves a number of elements and practices that should be considered in concert with each other. These elements and practices should also be considered in relation to the other two elements of ROME, participatory decentralization and results-oriented evaluations.

- 113 -

CHAPTER SEVEN:RESULTS ORIENTED EVALUATIONS IN THE ROME MODEL

The third section of ROME is results oriented evaluations. This section, when coupled with participatory decentralization and results oriented management, completes the Results Oriented Management and Evaluation model (ROME).

Evaluation is not a foreign topic in the public sector. Public organizations have a history of evaluating themselves and of being evaluated. Traditionally, however, public organizations have been evaluated in terms of their financial probity and procedure adherence, and not in terms of their results-oriented performance. Evaluations are usually focused on financial and procedural accountability, with the central question being whether or not the public organization follows the rules and procedures guiding its business. In more recent years, the definition of accountability, and the domain of evaluation, have both expanded.

Countries like Australia, New Zealand, and the United Kingdom all actively promote evaluations extending beyond financial probity and rule-adherence. In these countries, and in a variety of government entities at all levels of operation, evaluations are conducted to ensure that governments produce services that are by nature and quality suited to their environments and especially to their citizens’ demand profiles. These kinds of evaluations concentrate on the results of the government production process (outputs and outcomes) as well as the relationship between such results and the voice of citizens (the effectiveness of production in meeting demand) and the relationship between results and inputs (efficiency).

Developing countries do not have a good record for evaluations of any sort. Even evaluations mechanisms built into traditional budgets do not work well in such settings, and rules are often broken (Schick,1998). Some observers believe that this is a major flaw that limits the potential of results-oriented systems in developing countries—if established rules do not ensure effective evaluations at the simple level of rule adherence, how can one develop more elaborate evaluations concentrating on difficult concepts like results, effectiveness, and efficiency?

This negative perspective is countered by examples of results-oriented evaluations in developing countries that have successfully developed a results orientation in government—most notable of these being Malaysia and Uganda. Experience with results oriented evaluations is a bit different in developing varieties of results oriented management, as compared with developed varieties. Three kinds of results evaluations characterize both types: informal citizen evaluations of results, formal, internal evaluations of results within the production process, and formal, external evaluations of results from outside of the production process (as shown in figure fifteen).

- 114 -

All three kinds of evaluations are important, but differences in organizational capacity and culture probably favor the dominance of one over the others—with the informal option providing the right capacity-based approach as well as the right incentive consequences in developing settings. As a result, although countries like Malaysia and Uganda are like New Zealand in requiring entities to evaluate their own performance, and like Australia in requiring formal external evaluations by institutionalized role players like auditors, informal ‘social audits’ play a larger and more important role in such settings than they do in countries like New Zealand and Australia.

Results oriented reforms depend on evaluations occurring at all three levels and in all three shapes. These evaluations provide both a check on what governments are doing (especially a check against what governments are committed to do in their contracts) as well as an institutionalized device that should motivate governments to concentrate on their performance. The adage of “what gets measured gets managed” holds true here—if government performance is actively evaluated, these governments have an incentive to concentrate on improving such performance. To effectively create a checking and incentive mechanism through evaluations, all three varieties of evaluation mechanism should be developed to some degree. As such, it is important for results-oriented reform to consider: Involving citizens in results-based evaluations Focusing formal internal and external audits and evaluations on results (of regular

governance processes and activities and of reform itself).

- 115 -

7.1. Involving citizens in results-based evaluations

Results oriented reforms are focused on changing the emphasis of government organizations. This entails altering the systems and institutions underlying organizational behavior, as well as the incentives shaping such organizational behavior. Active evaluation, coupled with the results-based contracts featured in results-oriented government, is meant to consolidate incentives for managers and politicians to produce results. In developing countries, the kind of evaluation that is most effective in achieving this is citizen-based evaluations that involve society keeping tabs on government and responding to government performance by either ostracizing them, or praising them. Social ostracism and social acceptance act as powerful negative and positive incentives for public officials to produce results—especially at the local level.

Involving citizens in results based evaluations is particularly valuable where capacity to perform rigorous evaluations is low, and where market-based comparisons of results are not possible. This is the case in most developing countries. Governments do not have the capacity to perform rigorous analyses of production processes and the adherence to rules within such processes—so there is really little point in prescribing new rules for which there is no hope of evaluation. Social evaluations are fairly informal and require few resources, only an involved, empowered, and informed society that is sufficiently organized to speak into the life of its government.

This kind of evaluation is both facilitated by, and in turn facilitates, results-oriented government. In the first instance, citizens are more interested in results, and can more easily evaluate results of programs than they can input efficiencies. Providing results information thus opens the door to citizen interest and voice in government. At the same time, when citizens are involved and interested in what their governments produce, they create a community pressure and evaluation that acts as an important check on government performance as well as an incentive for government production. This is especially so when citizens are organized and informed about government production and government contracts (which facilitate their voice into government) and where citizens are freely able to meet and conduct ‘social audits’ of performance, with lines of redress and reward to follow such audits. The essential intervention requires mobilizing social evaluations.

7.1.a. Mobilizing social evaluations

For effective results-orientated social evaluations, it is important that citizens be both organized and informed (Gopukamar,1997). Organization amplifies citizen voice, while information guides such voice. In developing countries like Malaysia, organizations exist that facilitate organization, and governments are obliged to provide information to citizens—indeed, to publish information openly so that citizens can access such information without any transaction or social cost. Citizens can see, at the ‘store-front,’ exactly what government entities are contracted to produce—with details of the contract

- 116 -

clearly spelled out in terms of the amount of time services should take, the quality of services, and the quantity of services.

Figure sixteen. Involving citizens in results-based evaluationsThe Basics of

ROMEResults Oriented Management Results Oriented Evaluations at a

Social LevelSocial

organizationSocial and administrative contracts

publishedSocial audits:

Informal analysis of results against promises, and more

formal, forum-based analysis of reform details

Social redress/ reward options

Social access to public

information

Management improvements

Community pressure on political and administrative contractors, added to official redress/ reward to motivate

improvements in management

Figure sixteen shows how this kind of citizen organization and information access facilitate evaluation of management in terms of published contracts. Social organization, bolstered by information access, facilitate the evaluation of managers in terms of such contracts. These evaluations take place as ‘social audits’ in which results are checked against promises for results. Such audits lead to the groups exercising redress and reward options, and pressuring government to improve its results.

Examples of effective social evaluation include social organizations mentioned in the participatory decentralization section. DISHA, a non- governmental organization in Western India, combines its strong social capital with carefully analyzed budget information to actively evaluate the degree to which elected officials allocate money according to their election manifestos—the social contract in that setting. In Rajasthan, MKSS, another activist organization, evaluates basic financial probity on the basis of budgets and other financial records.

Both DISHA and MKSS pressurize governments into being more accountable, with social rejection and ostracism befalling officials who are caught performing poorly or contravening rules. In this way, the social evaluations process has generated incentives for officials to behave in an accountable and responsive way. Society proves to be the most knowledgeable source into what government is doing, as well as the most influential player in influencing government to do better.

These examples show that social evaluations can be very effective, proving the ultimate source of motivation for the public sector. They are based on the idea that opening the government up to the scrutiny of its citizens will lead to government responding more readily to those citizens. This has certainly been the case in countries like Malaysia and

- 117 -

Uganda. The examples also show that social organizations are more effective than individuals in constituting social evaluation agents. They have greater voice and influence over government. Where possible, public-based evaluations should be focused around the role of organized groups in society—with open channels for such groups to comment on the results of government. The examples further indicate that, in order to facilitate social evaluation, governments should provide citizens with information about government results—expected and realized. The more information citizens have, and the easier it is for them to access and evaluate the information, the more they will use it to evaluate governments. Information provision should thus be substantive, relevant, easy to access, and easy to evaluate (it should be de-mystified).

These aspects of social audits require that reformers:1. Facilitate social audits by providing social groups with information, forums, and

lines of redress through which to evaluate and enforce evaluations on governments.

Enabling social evaluations

In traditional government structures, citizen voice is not encouraged in the public production process. This often manifests in limited opportunities for citizens to meet and discuss government production. In the results-oriented model, such social meeting and discussion opportunities are central to the idea of effective social evaluation. These opportunities are exemplified in developed economies, especially at the local level, where citizens hold forums to discuss service types, and service delivery. Peters (1996) presents the example of discursive democracy in Europe, where citizens gather regularly to consider how government is doing and to make decisions about what government should do in the future, and who should be appointed in government.

This kind of social forum is shown in figure sixteen as central to a ‘social audit,’ which, when leading to anticipated redress and reward, is an important source of pressure on governments to improve and respond—central characteristics of the results-oriented system. An example is the MKSS influence in Rajasthan. Here, community evaluation devices take a number of forms, including jun sunwais—public hearings—at which detailed accounts derived from public expenditure records and other documents are read aloud to assembled villagers. Local people are invited to give testimony that might highlight discrepancies between official records and their own experiences. “Through this direct form of ‘social audit’” discrepancies have been identified and public officials (politicians and administrators) have been bought to account. Furthermore, there is now an active process of social negotiation regarding how government documents are structured so that they maximize the civil society window (facilitating evaluations) and facilitate government learning from civil interaction (Jenkins and Goetz, 1999).

Another example comes from Malaysia, where the Client’s Charter constitutes a written commitment made by the agency to its user—a guarantee of service—in terms of which social groups can frame evaluations and lodge appeals for redress. It spells out the rights of a customer and the agency’s service recovery mechanisms and redressing grievances, should it fail to fulfill its pledges. The Service Recovery System is aimed at ensuring the

- 118 -

confidence of customers, based on the philosophy that the ‘customer has a right to remedial services.’ When the recovery system fails to work, customers can seek redress for their grievances from a Public Complaints Bureau (PCB) (Chiu,1997).

A final example is the work of the Bolivian ‘vigilance committees,’ formally developed from social organizations in rural areas to monitor and evaluate the performance of local governments. These committees can meet with local people, and because they are themselves locals, they tend to interact with ‘government customers’ regularly. Their interactions are the basis of their evaluations, which carry weight in the Bolivian system largely because of their direct line of communication to the national executive.

In all the examples, society constitutes an important source of evaluation and influence in government, with the following four lessons standing out.

Social evaluation requires information: Provide society with information about results-oriented contracts specifically. Results-oriented evaluations are founded on the results contracts that agents enter into with their principals. If society knows the details of the results contracts entered into by administrators and politicians, it can act to enforce such and to check on results progress in terms of such contracts. The leader in providing such information is Malaysia, which mandates the publication of results targets at source of service. This means that a school will provide a public notice-board showing the services it is ‘contracted’ to produce, and a licensing department will publish its promised results (including targeted times for serving customers) where customers are in fact being served. In the developed world, especially in the United States, the internet is being used as a notice-board where government entities can publish their service promises.

Social evaluation requires social for a for evaluation: In examples of effective social evaluations, forums are evident for engaging citizens to ‘audit’ government performance. These forums facilitate the broad identification of short-falls in performance that are not possible through the eyes of individuals alone. Results evaluation forums should be held in local governments, especially, and should be focused around the information about what government is meant to produce. The presence of public managers and political representatives at such forums should be mandated, so that these officials know that they will be held to account for their performance in front of their peers and neighbors—a knowledge that should motivate improved social service, efficiency, and performance.

Social evaluation should be regular: Irregular and informal forums certainly have value, but also suffer from a number of weaknesses. Attendance at such forums is not as certain as it is when forums are regular and organized, and the status of the forums is not as great. Regular, and formalized forums of citizen evaluation are required to effectively evaluate government results.

The results of social evaluation must carry weight : Evaluation is stressed in ROME because it completes a process of institutionalizing a results-orientation. This means

- 119 -

that it assists in changing the rules of government so that behavior according to such rules is results-oriented. Rules only work if individuals are rewarded or punished according to their adherence to rules. In ROME, if a social evaluation finds an official producing results above or below requirement, the individual must be rewarded or punished. If the social evaluation is not acted upon, and does not have influence, the results-orientation will not be institutionalized.

7.2. Transforming formal evaluations processes

Apart from ‘bringing’ citizens and civic groups into the evaluations process, governments should also adjust their own evaluations focus. Most governments have some form of evaluation in place, usually taking the form of financial audits or corruption analysis processes. These evaluations almost always concentrate on process rather than results. The concentration dates to the emphasis on accountability to higher level principals, either in colonial governments or in hierarchical bureaucratic structures dominating developing countries.

If one believes that ‘what gets evaluated gets done’ it should be apparent that a process focus in evaluations will spawn a process focus in government. In essence, these kinds of evaluations promote incentives for administrators to rule by the book rather than in a responsive, accountable way. To move from such a situation to a results oriented government requires transforming evaluations processes such that they provide incentives for administrators to pursue results and to adopt results oriented management tools and rules. The basic requirements of transforming formal evaluations processes involve: (1) Transforming internal evaluations processes to concentrate on results and (2) Transforming external evaluations processes to concentrate on results

7.2.a. Transforming internal evaluations processes to concentrate on results

In New Zealand, the results movement is firmly structured around the output contracts between political representatives and administrative heads (called Chief Executive Officers). These contracts are consolidated by a requirement that administrative entities evaluate their own production on a regular basis, and according to standardized processes. Each entity produces financial and other statements that reflect their accountability and results-based performance, on a regular basis. This practice is designed to reinforce results incentives in the system, and to streamline the evaluation of administrative contracts by political representatives.

The logic is simple; because administrators are constantly evaluating their results, they are constantly aware of their results and can fine tune their production to maximize the production of results. Furthermore, because administrative heads are constantly producing results information (as well as other information on financial accountability) it is easy and cost effective to evaluate contracts. In New Zealand there are no special externally conducted evaluations of results, because the internal evaluations are so well

- 120 -

planned and so regularly developed. This kind of evaluation also features in governments like Malaysia and Hong Kong. It has an important t role to play in influencing the culture of public administrators, who are reminded of their results focus every time they have to conduct an internal evaluation. These internal evaluations are most effective when they are included as requirements in management ‘contracts’ and when their format is standardized so as to prohibit entity-specific vagueness. The reform focus should be on:1. Including results evaluations requirements in management ‘contracts’2. Standardizing requirements for internal evaluations

Including results evaluations requirements in management ‘contracts’

In New Zealand, administrators are required by contract to monitor and evaluate their department’s performance. The focus is on departments adopting techniques aimed at managing for results and evaluating results—including planning and budgeting techniques. Each portfolio has to develop its own performance monitoring capacity—an ongoing systematic review of the relationship between program inputs, outputs and outcomes in terms of progress against targeted achievements, past performance, and trend projections. Performance indicators are categorized in terms of a production process model, emphasizing both the process of production and the results of production. Three kinds of indicators are reported on—efficiency indicators (measures of the input/output relationship), effectiveness indicators (measures of the degree of achievement of the program element’s objective, or output results), and cost effectiveness indicators (the unit cost of outcome).

In developing these requirements internally, the performance monitoring system has evolved from a control mechanism to a cultural device. It engenders a mind-set in which individual and organizational attention is focused on objectives and results, and on finding better ways of achieving them. Management is now more self-aware and self-critical than previously and enjoys an ‘internal learning loop’ whereby internal evaluations feed information back to management decision-making to facilitate constant improvements and adjustments to production processes (Brignall,1997 emphasizes the learning loops developing from internal performance measurement in Australia).

This internal learning loop is shown in figure seventeen. It is accompanied by an external learning loop running from internal evaluations to those evaluating administrative contracts, usually political representatives. This loop provides these representatives with an automatic source of information for evaluating contractual performance as well as a window on the conditions facing public production—which allows adjustments to outcome targets and policies—an ‘external learning loop.’

These loops and internal evaluation processes also exist in Malaysia, where the main features of programs designed to improve service delivery are:

Service targets presented in an agency’s Client’s Charter together with workflow charts of key processes,

Agreed performance targets within the Client’s Charter, and Easily accessible facilities for evaluation and grievance resolution (Chiu,1997).

- 121 -

Figure seventeen. Evaluations in the results-oriented management processResults Oriented Management Results Oriented Evaluations built

in to results-oriented managementAdministrative contracts written into performance-based budgets: Specific

results must be produced in order to maximize reward and to ensure continued

funding/employment

Management is encouraged to develop and adopt devices to help reach results: Quality

improvement tools, tools for costing activities and outputs

and for registering full costs, a variety of reporting mechanisms,

and tools for managing strategically

Management devices facilitate ongoing internal evaluations—management can

gauge performance throughout the production process

Internal learning loop

Management learns from internal evaluations, and adapts the management and production processes—facilitating the most

efficient and effective production of contracted outputs

External learning loop

If internal evaluations devices are standardized, the internal evaluations can

be used by external evaluators (like auditors) to evaluate performance of management against contracts—as in New Zealand. This information can

form the basis of contractual negotiations in future periods.

Administrative heads produce information for evaluation, which they know will lead to the exercise of certain redress or reward options. The internal evaluations create the means of learning about process improvement and the incentive instrument motivating such improvement. Effective internal evaluations mechanisms can be developed on the basis of the following four lessons.

Link internal evaluations to performance-budgeting and contracts: Experience in many developing countries shows that internal evaluations are either developed ad-hoc or are not developed at all. ROME-type evaluations, focused on results and not process, are definitely rare. In order to ensure that such evaluations are performed internally, it is probably necessary to include a requirement for such evaluation into the contract—whether this be the budget or a separate document.

Link internal evaluation requirements to results, not just procedures: ROME-type evaluations differ from traditional audits and the like that concentrate on financial and

- 122 -

process regularity, not results. ROME is based on the argument that results should be the basis of accountability in government. Internal evaluations should be focused on results—the outputs produced.

Link internal evaluation requirements to other management tools: In the few successful examples of internal evaluation (notably New Zealand), internal evaluations have either been linked to, or carried out through, established managerial tools. Activity Based Costing, Benchmarking, and the Balanced Scorecard all constitute effective tools for measuring elements of performance and for reporting on those elements in ways that facilitate internal scrutiny and evaluation.

‘Sell’ the internal evaluations idea by stressing the learning loops it introduces: Most organizations do not enjoy evaluations—whether they are internal or external. Governments seem particularly averse to evaluations, and the requirement that such be developed internally is often seen as an unnecessary and threatening addition to an already overloaded work schedule. To overcome the hostility towards evaluations, the potential for learning through internal evaluations should be stressed. This learning has been described in terms of two important loops, which are important in a government entity exploring results-oriented management.

Standardizing requirements for internal evaluations

The New Zealand dependence on internal evaluations is reliant on the standardization of internal evaluations in terms of both approach and quality. This is evident in figure fourteen—the feasibility of the external learning loop and dependence on internal evaluations demands that such evaluations are developed according to a consistent and standard approach and meet certain quality requirements. These requirements should stress accuracy, relevance, and regularity of information.

In the United Kingdom, the importance of agencies measuring and reporting on performance has been recognized. Performance, rather than spending, has been emphasized, abut until recently reporting procedures were weak. Departments were neither trained to evaluate their own performance, nor provided a blue-print for such evaluation (Hyndman and Anderson). In New Zealand, such a blue-print was provided, and linked to other financial changes and the introduction of other management tools. Departments were required to provide the following documents when reporting, twice yearly:

1. A Statement of Financial Performance2. A Statement of Financial Position3. A Statement of Cash Flows4. A Statement of Service Performance

The standardization of reporting requirements allows for the easy analysis of internal evaluations. Documents can be evaluated in the same way, no matter which department

- 123 -

or agency they emanate from. This experience provides two lessons to results-oriented reformers.

Standardize the information required in internal evaluations: It is important that internal evaluations are open to external scrutiny, so that the value of the evaluation itself can be checked, and also so that the information from the internal evaluation can be used in external checks on performance. To ensure such openness, the format of internal evaluations and of internal reporting should be standardized.

Train departments to produce internal evaluations according to the standards set: There is probably a need to train departments to develop evaluations according to standards. The onus should not be solely on the department to do this.

7.2.b. Transforming external evaluations processes to concentrate on results

Governments traditionally have an agent or agents dedicated to performing independent, external evaluations (see Carlo, 1997, for an example from Puerto Rica). In developed countries especially, this role is often filled by the auditor and is concentrated on analyzing the financial probity of government departments. Other evaluation institutions include independent committees who check on the adherence to rules and processes in such specific practices as personnel appointments and procurement.

In the United Kingdom, these traditional audit functions concentrate on financial management:

Financial audits cover the examination and reporting on financial statements, and the examination of the accounting systems upon which these statements are based.

Compliance or regularity audits examine legal and administrative compliance, the probity and propriety of administration, financial systems and systems of management control.

Value for money audits constitute a recent addition to these traditional concentrations. These audits assess the management and operational performance of departments in using resources to meet stated results (Davies and Shellard,1997; United Kingdom HM Treasury,1999). As such, they could be called external audits of government performance or results.

These United Kingdom value for money audits are one example of a new type of external evaluation in governments that focus on results. In the United States, “local government auditors are increasingly being recognized for their role in establishing accountability while improving government performance” (Niesner,1999,37). The role of the external audit has been expanded to include all of the methods, systems and functions used to achieve the purpose and mission of the organization. External evaluations by these auditors are both institutionalized checks on production performance, and motivating factors facilitating progress with results-oriented reforms. The external evaluations

- 124 -

encourage change and in many ways provide the direction for capacity and reform development.

The auditors used by United States local governments come in many forms. Some are allied to local government associations, others are related to higher levels of government, and a few are even from the private sector. They are all, however, independent of the governments they evaluate. Similarly, the external evaluating entities in Australia, the United Kingdom, and Hong Kong, are all independent of the agencies and departments they evaluate. Furthermore, all of these external evaluators have a broad brush when they evaluate—concentrating on the details of results, process, and reform. Steps:1. Creating an independent evaluation agent with social credibility and technical

capacity2. Evaluating the multi-dimensional character of government outcomes and

performance, including government results, process, and reform, and3. Giving special attention to budget (and budget reform) evaluations.

The importance of independent evaluation agents

One of the major questions facing developing countries attempting to apply results-oriented reforms is how to ensure that administrative heads carry out contractual obligations. Certain observers suggest that the poor record of rule adherence by these administrators does not augur well for their contract adherence in the results-oriented system. These observers are concerned that a results-orientation could lead to an ultimate breakdown in public sector performance.

This concern makes the role of external results evaluation vital in the developing world. External evaluators with an independence from administrators responsible for service provision in terms of contracts, offer an institutional mechanism to ensure contract adherence. The mechanism works by providing objective information on a regular basis, facilitating broad social evaluations, as well as providing an independent source of evaluation. These kinds of evaluation institutions are relatively new in many developing countries. In developed countries like the United States, the institutions may have a history, but their results orientation is also new.

In a way, then, developing and developed countries are at the same point in setting up independent, results-oriented evaluation institutions. In the United States, a relatively young organization, the National Association of Local Government Auditors (N.A.L.G.A) promotes independent, effective local government auditing (Niesner,1997). The organization exists independently on local governments and provides audit services as well as results-management training and assistance. It is characterized, along with similar organizations, by the following elements:

It is independent and offers an objective approach to results auditing. Its staff is both competent and committed, with skills ranging from performance

measurement, benchmarking, control self-assessment, activity-based costing, and strategic planning.

- 125 -

It offers quality products and services to the governments it audits, assisting them in improving their results-focus while evaluating performance in terms of such focus.

While being independent, the external evaluator works with the entities it evaluates, developing relationships and ensuring that the evaluation experience is as much one of learning as it is one of grading.

These characteristics, and the experience of external evaluators in other results-oriented governments, provide the following five lessons to developing countries attempting to progress with ROME-type reforms.

Develop independent external evaluators: The external evaluation agency must achieve independence to allow the checks and balances of effective democratic government to work. The independence of these agents aside, it is important to ensure clarity as to their interest in the governments they evaluate. In the most effective cases, external evaluators are audit agencies that report to higher level governments (in the case of local governments with results-oriented contracts tied to intergovernmental grants), elected officials, or the public. Central, coordinating agencies often fulfill this role (Davies and Shellard,1997).

Develop evaluators from within social organizations: Experience with evaluations in developing countries suggests that society has a major role in holding government in check, as long as society is organized and is able to access information necessary for evaluations. In experiences such as the Bolivian experiment with ‘vigilance committees,’ external evaluations carried out by institutions built on social representation have proved very effective in holding government accountable and developing evaluations that motivate results-orientation.

Ensure evaluator competence: The expansion of audit services requires diverse skills in evaluating agencies. Skills and experiences required include performance measurement, benchmarking, control self –assessment, activity-based costing, information technology, investigations, and strategic planning. These skills are similar to the skills that need to be developed in results-oriented management, which suggests that the central coordinating agency or other entity providing such services to departments could also develop an evaluations skill.

Structure evaluations agents to provide quality products and services: External evaluators are well positioned to provide the high-quality, reliable and impartial information requested by citizens, elected officials and managers. These services should be provided in accordance with professional standards and objectivity (Niesner,1999).

‘Sell’ the external evaluations to managers—by making them more than just checks: Most managers do not like evaluations. The tension between external evaluators and managers is useful if both know what is expected of them. In the results movement, neither managers nor evaluators are currently well equipped to stand alone and in

- 126 -

conflict with one another. Results-oriented reform requires that, at least in the opening stages of results-based reform, evaluators provide a service that extends beyond checking on managers. Their role is to consult on reform and results-management, as related to the evaluations they produce. Niesner (1999,36) sums this sentiment up by saying that “government auditors are uniquely positioned to promote performance measurement and performance accountability and to help ensure that these systems are used to improve government services.”

Evaluate government results, process, and reform regularly

Probably the most famous quote of the results and performance movement is, “what gets measured gets done.” The quote is particularly relevant in considering the role of external evaluators in the results-oriented government. These evaluators, traditionally auditors, need to focus on a variety of factors without concentrating too heavily on one or the other. In particular, they need to concentrate on traditional financial and process accountability—as in the United Kingdom’s Financial and Compliance audits. They also need to concentrate on the evaluation of results—as in the United Kingdom’s new value for money audits. Finally, they need to provide active evaluations of the progress of reform.

Concentrating on measuring (evaluating) these three factors ensures that all three receive attention from managers, as shown in figure eighteen. Ignoring one or more of the factors in external evaluations gives the message that such factors are unimportant and leads to managers concentrating their attention away from them. This has traditionally been the case with results and reform progress in most bureaucratic governments.

Figure eighteen. Formal, external evaluations and results-oriented reform: What gets measured gets done!External evaluators measure

results…Management has an incentive

to produce results…Management

produces results.

External evaluators measure accountability in process…

Management has an incentive to be accountable in process…

Management is accountable.

External evaluators measure reform progress…

Management has an incentive to progress with reform…

Management progresses with

reform.

Most external agents have some experience with auditing process and financial compliance. Their skills in evaluating results and reform progress are, however, limited. In this light, one of the most important skills that results-oriented external evaluators need to develop is that of results evaluations. They need to ask questions about how well outputs are produced, as well as about how well outputs help to achieve the outcomes the government wishes to pursue (Ball and Dale, 1998). This evaluation point keeps the focus on outcomes even if the results being measured are outputs. It also keeps the allocations question fairly open and prohibits base budgeting (knowing that an external evaluator is going to question the outcome-related efficacy of their programs,

- 127 -

departments should have an incentive to ensure that their production does help to facilitate outcome achievement).

The following lessons on external evaluations emanate from the limited number of multi-dimensional evaluations agents in place around the world. Three sets of questions are also provided as sample questions covering areas not well developed in most evaluations—results, efficiency, and reform progress.

Concentrate external evaluations on the factors that matter: When external evaluators are appointed, those being evaluated look carefully at what is being evaluated to gauge what matters and what does not matter. In structuring the evaluations, therefore, it is important to concentrate attention on all the factors that matter—largely results, process and financial accountability, and reform progress.

Ensure that the emphasis on outcomes is not lost: Because administrators are concerned with outputs, evaluations of results should concentrate on outputs. At the same time, however, evaluations should always question the link between outputs and outcomes, so that the long-term focus of government is not ignored and administrators are forced to see their role in terms of a larger picture.

Ask the right questions, especially of results: Helpful questions for evaluating results (adapted from United Kingdom HM Treasury,1988 and Andrews, 2001) include:1. Have results been achieved in terms of quality, quantity, time?2. To what extent was the achievement the effect of government action?3. How has the result been evaluated internally? Is there a benchmark? How

reliable is this evaluation?4. Has government action induced any unforeseen results (positive or negative)?5. Was the environment for policy implementation the same as or different from

expectation? Did a change in environment affect the achievement of results?6. Are the results-objectives still relevant?7. Is the existing policy still the right one for meeting the results-objectives?8. What steps should be taken to improve/alter the policy?9. Are there any lessons for other areas of policy?10. Are there any lessons for the management of the department?

Ask the right questions of efficiency: Helpful questions for evaluating efficiency (adapted from United Kingdom HM Treasury,1988 and Andrews, 2001) include:1. Are departmental processes transparent?2. Have all inputs been made according to planned amounts, timing, and quality, and

are all inputs accounted for in financial records?3. Have all inputs been included in the cost assessments (including direct inputs,

capital, overheads, and support inputs)?4. How much of the input was wasted (irrelevant or unsuccessful projects,

avoidance, deadweight costs, unused capacity)?5. How does the input requirement compare with input requirements of comparable

producers?

- 128 -

6. What is the cost of a unit of output (distinguish program cost and administration cost)?

7. How effective are cost evaluation and management practices? Has there been an improvement in cost efficiency?

8. Were there any complaints about the administration?

Ask the right questions of efficiency: Helpful questions for evaluating reform progress (adapted from United Kingdom HM Treasury,1988 and Andrews, 2001) include:1. Has the department adopted a strategic plan to concentrate on results?2. Has the department implemented any new management mechanisms? Are these

mechanisms being used?3. Does the department communicate with customers on a regular basis? How does

it communicate? What does it do with the information received from customers?4. Have departmental performance goals been improved upon? Have departmental

capacities to produce results been expanded?5. Is the reform considered important in the department? Does the internal

management have any alternative reform ideas?6. Is the management implementing internal changes to the departmental culture

(like incentives)?7. Is the reform strategy comprehensive, building on a base of social interaction and

focusing management on results? (Or is it made up of stand-alone elements)?

Evaluating budgeting and budget reforms: the heart of governance

Budgeting is central to the governance process and budget reform is central to the public sector reform process. Budgets provide the resources for results generation, and their structure, processes and institutional character constitute a major influence on governance processes in general. In short, budgeting processes are the dominant source of incentives in the governance process. As such, reforms applied to budgets require special evaluation. Typical evaluations of these reforms do not fit the bill however, asking limited questions about whether reforms are in place or whether final outcomes are improving. A set of questions is required that examines whether budget reforms actually influence behavior and incentives in the governance process.

Reform literature is explicit about the fact that sub-optimal fiscal and other governance outcomes are expected for many of the same reasons that inform expected ‘failures’ in free market systems. These reasons relate to ‘budget failures’ that arise because of self-interested behavior on the part of administrators and politicians, which undermine the basic requirements of efficient market interaction. There are four areas of ‘budget failure’: Citizens have poor budgetary access: As consumers of public goods and services

citizen preferences should underlie budgeting decisions. Citizens generally enjoy little budget access, however. Access is indirect through election processes, political parties, and interest groups, which are often no more than “crude instruments of popular control” occurring at intervals spaced too widely to be relevant (and often

- 129 -

selectively accessible) (Blair,2000: 27). The literature argues that politicians lack incentive to represent citizen interests in budget decisions because of fiscal illusion on the part of electors or because electors tend to be perennially uncertain of political policy preferences (Buchanan, 1967; Goetz, 1977; Wittman, 1977, 1983). As a result, theory and empirical evidence points to a budget failure in this area. Citizens have no way of expressing preferences or of holding government accountable for its decisions and performance (Feld and Savioz, 1997; Feld and Kirchgassner, 1999).

Administrators have an informational advantage: An assumption of any market is that producers (administrators) provide consumers (politicians and citizens) with accurate information about the cost and quality of their good or service. This flow of information is necessary in making consumption decisions and creating an incentive for producers to deal with consumers in good faith. Budgeting research places a question mark over whether administrators provide politicians or citizens with appropriate production information. Theory suggests that budgeting processes are characterized by informational asymmetry: administrators have an informational advantage over political and citizen consumers. As a result, administrators have an incentive to provide insufficient information for decision-making or to load information with data that supports their own preferred agendas (Munger,1984). This informational asymmetry is often blamed for poor budgeting decisions, over-spending and poor allocations (Campos and Pradhan, 1996).

Budgeting allocation decisions are seldom efficient: Efficient decisions involve allocating money to projects with relatively low social costs and high social benefits (Matheson, 1998). Decisions are influenced by conflict between parties vying for allocations, formal and informal ‘rules’ governing decision-making, and the political power struggles informing such. These problems lead to a variety of potential ‘budget failures’: common-pool problems, barriers to entry, and budgetary monopolization. Common pool problems involve parties under-considering project costs and over-emphasizing benefits, leading to too many projects and too much budget burden. Barriers to entry arise as ‘rules’ governing decisions yield some proposals admissible while limiting the access of others (Kraan, 1996:157). These rules are often developed by decision-makers protecting private interests at the expense of broad social interests, thus establishing ‘budget monopolies’ (Wildavsky,1988; Kraan,1996). Given evidence of such practices, the budget process is not expected to yield rational and socially responsive decisions or accountable decision-makers.

Budgetary implementation is seldom efficient: Efficient budget implementation requires administrative producers provide goods and services demanded by citizen and political consumers at the best price possible. Studies suggest that administrators have incentives to spend differently and more than their budgetary mandate (Miller and Moe, 1983; Bendor, Taylor, and von Gaalen, 1987; Kraan, 1996; and Campos and Pradhan, 1996). The principal-agent problem in budgetary implementation raises a question about how politicians and citizens can ensure administrators implement budgets efficiently and responsively. The principal-agent problem is a ‘budget

- 130 -

failure’ that impacts central outcomes of the budget process: fiscal discipline (do administrators spend more than they are allotted?), allocative efficiency (do administrators allocate money to the right projects, or do they favor personal projects that have lower social benefit and higher social cost?), and responsiveness and effectiveness (do administrators implement projects using the lowest cost, most efficient approach?).

These budget failures provide a rationale for budget reform, as well as a set of questions relevant to reform evaluation. Figure nineteen provides a summary of this rationale, with reforms needed to redress ‘budget failures’ which thwart the achievement of budgetary outcomes like fiscal discipline, efficiency and responsiveness. Such outcomes are not achieved when budget failures dominate the budget process. Citizen access failures, for example, limit the potential for budgets to reflect citizen priorities while decision-making failures limit the efficiency of budget allocations.

Figure nineteen. Reforms redress budget failures

In terms of figure nineteen, although budget reforms are introduced with the improvement of final outcomes in mind, the direct effect of reform is not on outcomes. If outcomes are sub-optimal it is because of failures in the ‘black-box’ section of the budgeting process. It is in this section that reforms are required to redress failure. It is in this section, therefore, that reforms should be evaluated (in terms of their ability to redress failure). In redressing budget failures, successful reforms alter behavior, which then results in changed outputs, outcomes and impacts.

Successful budget reforms redress budgetary failures and alter behavior in the ‘black box’ by transforming incentives facing budgeters, such that new incentives counter failures. The details of the new incentives and behavioral changes will vary from government to government, depending on the kind of failure being addressed. It could be argued that most countries face all of the failures, however, and reforms tend to have a blanket effect. This said, the point is to ensure that reform elements do in fact target failure areas and are evaluated as such. Successful reforms, which change behavior in specific areas of failure and further influence outcomes, can be discussed in terms of the failures they redress. Such discussion provides the focus for theory-based reform evaluations.

- 131 -

Reforms countering failures associated with poor citizen access: Successful reforms focused on redressing failures associated with citizen access are those that provide citizens with access points into the budgetary process as well as creating incentives for citizens to use such access points in voicing their budgetary demands and in maintaining a watch on budget progress (by ensuring tangible citizen gains from budgetary participation) (Campos and Pradhan,1996: 13). These reforms limit the ability of politicians and administrators to mystify the budget process and place these actors under social pressure to take citizen demands seriously and to be responsive to citizens (with the promise of greater citizen support, or with the threat of citizen censure). Reform evaluations should ask questions about how reforms counter budgetary failure by focusing budgets (and the incentives of budgeters) on citizens:

1. Do reforms provide citizens access to the budget process?2. Do reforms give citizens incentive to participate?3. Do reforms give citizens information about the budget (and de-mystify it)?4. Do reforms make politicians and administrators more aware of (and concerned

about) citizens?

Reforms countering failures in information flows: Effective reforms yield information that is both comprehensive (providing accurate data on a variety of budgetary alternatives in a timely and efficient manner) and easily evaluated by politicians (Kraan,1996: 192-197). Reform success is closely linked to the degree to which administrators have an incentive (not just a tool) to provide such information to politicians. Reforms should thus be evaluated on the basis of their ability to enable and encourage administrative information provision. Reforms that do this will likely counter budget failure arising from information asymmetry, requiring that budgeters change their behavior to allow positive answers to the following:

1. Do reforms lead to administrators providing relevant information to politicians?

2. Do reforms assist politicians in processing information?3. Do reforms penalize administrators who do not provide information?

Reforms introduced to counter failures in decision-making: Reform success lies in the degree to which reforms create tools, rules and incentives for decision-makers (whether politicians or administrators or both) to make rational and responsive budgetary decisions. These reforms give decision-makers the ability to make informed decisions and also reward them for making such (or penalize them for failing to do so). The reforms actively correct failures in the budgetary allocations stage of the budgetary process, which often leads to opposition to such reform. The allocations stage is traditionally dominated by interests and influences that have managed to entrench themselves over a period of time. Examining whether decision processes are changing from traditionally closed and entrenched structures to reflect transparency and responsiveness requires asking questions like:

1. Do reforms make decision processes transparent?2. Do reforms structure decision ‘rules’ to prohibit over-demand and the chance

of ‘decision monopolies’ (by stipulating budget targets and increasing proposal access, for example)?

- 132 -

3. Do reforms create decision processes based on objective measures like cost and benefit information?

Reforms introduced to improve budgetary implementation: Reforms are also effective when they give administrators the tools and incentives to produce goods and services responsively and efficiently. Such reforms institutionalize methods by which administrators, politicians and citizens can evaluate and appraise implementation performance. These mechanisms give budgetary consumers (politicians and citizens) reward and censure options to exercise over budgetary producers (administrative entities) and should motivate producers to produce goods and services efficiently, countering the fourth budget failure. Evaluations should concentrate on whether monitoring and evaluations mechanisms are in place, whether ‘consumers’ feel empowered to use these, and whether administrators are motivated by such ‘consumer empowerment’. Questions facilitating this include:

1.Do reforms establish monitoring mechanisms that focus on aspects of budgeting that politicians and citizens care about (fiscal discipline, performance, etc.)?

2.Do citizens play a role in monitoring? 3.Do monitoring results empower ‘consumers’? 4.Do administrators answer to ‘consumers’ regarding monitoring results?

7.3. Conclusion: Results oriented evaluation and incentives to govern well

Results-oriented evaluation is the section of the ROME puzzle that has received the least attention in the reform community. It has proved difficult to combine the traditional concept of process-based audits and the new results-orientation. In balance, governments attempting to introduce elements of results-oriented management tend to keep their traditional process-oriented external auditors, creating confusion about what government is meant to be aiming for—results, as reflected in the performance-based budget (for example) or rule adherence as stressed in the audits.

Some governments, particularly the examples in this section, are developing results-oriented evaluations to match their results-oriented structures and management focus. Examples in developed countries tend to concentrate on formal evaluations. New Zealand emphasizes internal formal evaluations. Australia and the United Kingdom emphasize external formal evaluations. Examples from developing countries stress social evaluations, with Malaysia being an example. As with the other sections of ROME the three types of evaluation (social, formal internal and formal external) are not presented to run as stand-alone applications. The three constitute a comprehensive approach to evaluation that requires interactive application, as shown in figure twenty.

- 133 -

As shown in the figure, the three kinds of evaluation impact on government in different ways. The social evaluation is important, especially in developing countries, in that it facilitates social pressure for results in government. This pressure is effective in motivating a results-orientation in government, where other forms of motivation (financial and other) are ineffective or cannot be afforded. The second kind of evaluation discussed in this section, internal evaluations, gives management a tool to understand their production process in terms of the new results-orientation. It also provides the basis for external evaluations in some instances, but does not directly facilitate pressure for management to comply with results targets—acting more as a device for learning than a device for enforcement. This enforcement role is played by the social evaluations discussed, as well as the objective external evaluations conducted by entities like auditors offices. These evaluations carry with them the potential for redress or reward which enforce the results-oriented institutional changes mentioned in this paper.

If only one kind of evaluation is in place, the role of the other two will be missed, and the results-orientation will be undermined. Consider the following examples:

MKSS in Rajasthan, and NGO committed to examining budgets and demanding that government comply with the details within such, is an example of a social organization conducting social audits. These audits have resulted in the identification of corruption, and other irregularities, and have in some instances led to changes in

- 134 -

policy and in improved government performance. The social audit is not, however, accompanied by internal evaluations or effective formal audits of government. As a result, although there is some evaluation, it is incomplete—departments do not effectively analyze their own performance to facilitate major improvements, and external auditors do not provide the enforcement mechanism to reinforce what MKSS is doing.

In New Zealand, evaluation is limited to internal departmental analyses of performance. Although performance targets are publicized, they are set with minimal social involvement. Citizens are not provided direct methods of evaluating and auditing results. Furthermore, external audits are developed almost completely on the basis of internal evaluations—which makes the internal analyses both sources of learning for Chief Executives and a potential source of reward and redress for such executives. In ROME, the internal evaluation is primarily a learning device, with the external evaluation constituting the enforcement/response mechanism.

Results-oriented evaluation is central in the results-oriented government. It provides a constant source of learning and a form of enforcement that provides the incentive for managers to produce results. It needs to be implemented as three pieces that inter-relate, not one or two pieces that stand alone. These pieces, further, interact with other elements discussed in the basics of ROME and results-oriented evaluations. In this interaction, ROME institutionalizes the pursuit of results in governments that usually follow a process direction.

- 135 -

CHAPTER EIGHT IMPLEMENTING ROME: ENSURING THE ROAD TO ROME IS NOT ONE OF

WRACKS AND RUINS

Literature shows that the main problem with reforms in the results movement is a lack of consistent application and comprehensiveness. Reforms are often implemented as stand-alone applications, and fail because they are not supported by the general structure or culture of public sector organizations. Indeed, there are many examples of ROME-type applications either failing or having only partial success in transforming governments, because of piece-meal reform. ROME provides a comprehensive approach to results-oriented reform that combines the most common reform concepts into a single model, emphasizing the basic requirements for governments moving towards an output and outcome-orientation, as well as reforms focused on introducing results-oriented management and results-oriented evaluation into government.

The three sections of ROME are effective when they are connected into an entire model. To develop a metaphor from the introduction, the ROME sections need to be combined to make sense of the ROME puzzle. The first section, participatory decentralization, provides the background of the puzzle, not the graphic details of how the ROME system actually works. These are developed in the results-oriented management section, which act as the color in the overall ROME puzzle. The results-oriented evaluation component adds a consistency to the model that holds all the pieces together—building on the basics and enforcing the use of the management elements. The sections need to be combined to ensure that: (1) a results-orientation is institutionalized in governments traditionally characterized by institutionalized process orientation. Furthermore, attention is required to ensure that (2) the institutional environment is conducive to ROME implementation, with a large ‘reform space’ evident.

8.1. Institutionalizing ROME

Institutions are rules and norms that shape and guide behavior. When rules are institutionalized, they are said to influence and guide behavior effectively. Many examples are evident in every day life, including the road laws that inform behavior of motorists and pedestrians. These laws are institutionalized through a complex interaction of factors similar to those that constitute the ROME puzzle:

Society is organized to contribute to the development of road laws. Society is informed about the laws, and about their influences. Various organizations and mechanisms are created to manage the implementation of

the rules, and the impact of the rules on society (the police force, legal organizations, and so forth).

The effectiveness of the laws is largely dependent on internal regulation of behavior on the part of road users.

- 136 -

The effectiveness of the rules is further enhanced by social ‘evaluation’ of adherence to the laws—if individuals do not adhere to the laws, they are ostracized.

Finally, the laws are enforced by ‘external evaluators’—law enforcement agents who monitor road behavior and prosecute those who do not abide by the laws.

The road laws are institutionalized through the interaction of all these elements. If one or more of the elements are missing, the laws would not be institutionalized (they would not drive behavior as desired). Consider, for example, if law development only involved telling society to abide by certain regulations—with no management agents to administer the laws and no evaluation or enforcement mechanisms. It is unlikely that the laws would be effective as people would easily be able to ignore them. Or consider a situation in which laws are developed centrally but society are not informed about them. No matter how much the rules are enforced and managed, they will not be adhered to because society does not know what they are.

8.1.a. ROME and the institutions approach

Moving governments from an input orientation to a results orientation involves a process of institutionalization similar to that in which road laws are created. This process involves the introduction of rules and norms shaping and guiding behavior towards a concentration on results instead of a concentration on process and inputs. The process of institutionalization is complex and involved, and requires attention to a number of elements—much like the development of road laws requires social involvement, information, management, and evaluation. In ROME, the important elements fall in three sections—the basics of ROME, results-oriented management, and results-oriented evaluations—which, when applied in concert, facilitate true results-oriented government.

Most results-based interventions do not present reform as an interaction of elements, however, concentrating usually on one of ROME’s sections (or even on one of the pieces or elements in an individual section. In particular, reforms tend to pay great attention to results-oriented management (and in that section, the concentration is on performance-based budgeting) without considering the importance of the other elements in ROME—the basics of ROME and results-oriented evaluation. This is one of the main reasons why reforms have not had much effect in practice. One cannot institutionalize a new way of thinking by providing a budgeting mechanism based on such thinking alone, and ignoring the influence of other factors.

Figure twenty-one shows the pieces of ROME interacting to generate a results outcome. The first two elements are derived from the section ‘participatory decentralization’. They are the voice of society, creating a demand for results, and the re-structuring of government, through decentralization and participation, to allow for a closeness to citizens and an increased ability to listen to and consider demands. The third section is taken from the results-oriented management section of ROME in which the public organization is focused on results and mechanisms are developed to provide the capacity to produce results. The fourth element emanates from the third section of ROME,

- 137 -

results-oriented evaluation. It involves the evaluation of results and the impact such evaluation has on management—effective evaluation and follow-up should create management incentives to produce results.

Figure twenty-one. ROME as a model of results-oriented reformParticipatory

decentralizationParticipatory

decentralizationResults-oriented

managementResults

1. Social demand for results is evident

2. Government structures are close to citizens, able to consider demand for results

3. Organization is structured around results demand, and has the capacity to manage for results

ImprovedResults

Results-oriented evaluation

4. Results are evaluated, and the evaluations provide incentives for government to continue (or begin) focusing on results.

When these elements are combined, they institutionalize a focus on what government does—its results—and the expected impact is an improvement in results. When one of the elements is left out, the results-orientation is not as strong, and it is not certain that results will improve. Consider, for example, a situation where social demand is not evident, or governments are not decentralized. Performance-based budgets and other management tools focused on ‘results’ in such situations are developed as top-down management tools rather than citizen-oriented devices focused on the results that matter—those that citizens are interested in. Or consider the case where social demands are evident, government is structured to respond to these, and management has the tools to facilitate effective response, but is not evaluated in terms of the results that are generated. There is no pressure to respond to demand and produce results, therefore. Even though the appropriate budget and cost systems may be in place, they are not guaranteed to lead to improved results.

8.2. Implementing ROME

In order to institutionalize a results-orientation in government, therefore, reform interventions need to combine the three sections of ROME. This requires planning for the implementation of a number of different elements of reform, something which has not been done in reforms partly because of the lack of a model arguing for such and partly because this kind of combination is incredibly complex. This said, partial implementation of the model does not guarantee any success in facilitating a results-orientation.

- 138 -

The first step in implementing ROME involves developing an understanding of how the various pieces of the model interact. This understanding is informed by three chapter conclusions in chapters 5 through 7, which discuss how individual elements in the three central pieces of ROME interact. It is also informed by figure twenty-one, which shows the important interaction of the pieces themselves. In terms of a plan of implementation, figure twenty-two provides a graphic of how the elements build on one another.

Figure twenty-two. The basis of ROME implementation

Reform facilitation:When managers start to focus on results, change internal institutions and provide tools necessary for results provision.formally monitor reform progress to ensure continued incentives for results focus.

Reform focus: Bottoms-up creation of social institutionsand social evaluationsto motivate for results-orientedgovernance, and institutional structures for effective, responsive local government

The figure shows that the first section, participatory decentralization, provides the blocks on which the other sections are built. The first two pieces of the basics of ROME, providing citizens with a voice and informing that voice, are of particular importance. Citizen demand forms the basis of results demand in the ROME model, which is bottoms-up in every sense. This citizen demand is often poorly considered in reform, and takes a while to develop—with reforms at this stage involving the active organization of society, and the dissemination of information on government production.

- 139 -

Focusing management systems on results

Developing results-oriented roles

Providing objective oversight through external evaluations, to create incentives

for results oriented reform

Organizing governments around citizens, facilitating citizen-oriented incentives in government

Stimulating a results focus by requiring relevant and regular internal evaluations

Decentralizing governments, establishing access and accountability to citizen voice

Involving citizens in results-based evaluations

Providing citizens with a voice, creating a source of demand for results

These pieces form the foundation on which government can re-structure itself, decentralizing its operations and involving citizens (now empowered) in its planning and production processes. Only once these re-structuring activities have been completed, should government develop active results-oriented management programs. Results-oriented evaluations should be developed in tandem with these management interventions to ensure a constant incentive to pursue a results-orientation.

8.2.a. Sequencing implementation

Having made sense of the basic pieces of ROME, one can consider how the reforms should be sequenced in implementation. It is not possible to develop all of the pieces at once, and some of the pieces should only be developed after others are in place. Our model of implementation is not meant to be a rigid mode, just as the specific practices we discuss are open to variation. We do, however, suggest that there be at least two phases of reform implementation—the ROME starting point, involving changes to the institutional environment in which governance processes are located, and the ROME tools stage, where results-oriented rules and tools are introduced to facilitate results oriented management and to ensure continued incentives to produce results.

The starting point: considering the institutional environment

The first step in pursuing the development of a results-oriented government is adjusting the institutional environment for results oriented government. This step comprises implementing a number of elements presented in the participatory decentralization section as well as elements from the results-oriented management and results-oriented evaluations sections. This stage of implementation is designed to provide a foundation for results-oriented reform. At this stage, governments should:

Concentrate on developing the voice of citizens, and structuring themselves to respond to that voice. This means developing social action plans to mobilize society, and providing information to society.

Concentrate on structuring political processes to facilitate representative government where political representatives are held to some form of ‘social contract’.

Concentrate on decentralizing governance structures in the way presented earlier in this paper.

Promote and mandate the adoption of tools that allow an outward orientation among managers, facilitating administrator-citizen interaction. Tools like Total Quality Management should be developed to this end.

Relationships and roles should be re-considered at this early stage, as much to facilitate a consciousness about the process orientation of old models of government as to develop new results-oriented organizations.

A coordinating agent is very important to ROME and needs to be developed before major results-oriented plans are set in motion. This coordinating authority could come from central agencies or could be independent.

- 140 -

Finally, on the evaluations side, the early stage should involve some mobilization of social evaluation (as citizens are empowered and their demand for results is facilitated by organization and information). An external evaluations agent should also be raised up in this early period, to provide advice and assistance in the move to a results-orientation and to evaluate the reform progress as well.

Changing rules and tools

Building on the elements developed in the early reform period, there should be a period in which plans are set and capacities are developed for managing towards results. This period of reform concentrates on the specifics of developing results-oriented contracts in government, through budgets and other devices. It also looks at involving citizens directly in government planning, and in developing official evaluation requirements and rules. The main results oriented management section comes into its own here, with the tools discussed forming the foundation of the reform facilitation at this point. Importantly, these tools should be provided in response to demand, not on the basis of plain ‘reform supply’.

8.3. ROME, the institutional environment and reform space

Most governments are continually in the process of implementing reform. Reforms currently being implemented tend to resemble the results-genre, like ROME. Generations of reforms preceding these have failed to improve governance processes and results, yielding a question: “What factors should be considered when implementing ROME, to give it a greater chance of success?” The “usual suspects” of reform failure are ‘poor capacity’ and a ‘lack of political will’ (Ramakrishnan, 1998; World Bank, 1998). While reports emanating from the development community regularly emphasize these concepts, however, it is difficult to find publications that explain what they are. Indeed, the concept of ‘capacity’ is so large as to prohibit definitive explanation (Grindle, 1997; Polidano, 2000) and ‘political will’ is a term sufficiently loose to mean both everything and nothing at all. As such, while the two concepts provide general excuses for reform failure, they don’t constitute an avenue for effective explanation of such failure, especially in ‘industrialized’ countries where reforms also have a poor record but the two suspects are arguably less present.

Shah provides a new, more comprehensive, explanation of “why the road to reform remains a field of dreams” (Shah, 1998). His model of the public sector institutional environment is the basis of this chapter, which argues that countries pursuing ROME-type reforms need to pay attention to the institutional environment in which reform itself occurs. Three sets of factors require particular attention: mission, value and goal-based acceptance, authority and capacity. For ROME reforms to genuinely improve governance, reformers must concentrate on harmonizing the participatory, results-oriented philosophy with “the public sector’s mission and values, its authorizing environment and its operational capacity” (Shah, 1998: 7). Figure twenty three illustrates the basic model.

- 141 -

Figure twenty three. Reform and the public sector institutional environment model

The three factors, acceptance, authority and capacity (or ability) interact to create an institutional setting for governance, which is shown in the graph as a ‘reform space’. Governments could have the capacity and authority to govern according to a specific model (or to implement a reform), for example, but if the mission and goals of the governing model or reform clash with those of the public organization, the model or reform are likely to have less than an impressive impact. In terms of the figure, the ‘reform space’ exists at the point of interaction between all three factors. Explaining reform experience and developing reforms that impact on governance outcomes requires paying attention to all three factors individually and, probably more importantly, examining how they interact to create ‘reform space’. The size of this reform space, and thus the appropriateness of the institutional setting for reform, has a great influence on chance of ROME success.

The three factors identified in Shah’s model reflect both ‘supply’ and ‘demand’ sides of governance and reform (Coston, 1998). Capacity is obviously focused on the supply side, related to “the capacity to carry through policy reforms” (Coston, 1998: 481). The authorizing environment, and values, missions and goals reflect the kinds of pressures exerted on governments (some from within governing organizations themselves) to behave in certain ways. The interaction of all three factors manifests in different abilities to reform (on the supply side), as well as different incentives and demands regarding reform (on the demand side). Each factor is itself the sum of a number of sub-factors.

Shah argues that a “closer look” at these factors (and indeed their interaction) will assist understanding “why the public sector is dysfunctional” and “difficult to reform” (Shah, 1998: 6). The current article provides such a “close look” with the next three sections focusing on factors individually and the final section discussing how the factors interact

- 142 -

to influence reform experience. All sections apply evidence from studies on reform to detail the model and illustrate its validity.

8.3.a. The factors affecting reform experience: Capacity

Weak operational capacity is often argued as the reason why reforms fail to achieve their potential, with studies identifying situations where abilities are “not consistent with the (reform) task at hand” (Shah, 1998: 7). Coston extends the argument in saying that “governance problems” in general “may derive from a lack of capacity” (1998: 480). Capacity is considered a core issue in India, for example, “posing serious issues of capacity even for the Indian federal state”(Polidano, 1998: 374). Even the New Zealand experience (usually the model for reform success) was at times frustrated by limits in ability (Aitken, 1997). Two aspects of capacity recur as factors in the literature and are the key areas for concentration when asking if governments have necessary ‘ability’ to reform (Bell et al., 1999; Watson, 2000): (1.) Personnel capacity, and (2.) Process and systems capacity.

Personnel capacity

Recent research suggests that the key to any significant reform is the people involved (Bell, et al. 1999). Personnel capacity influences service delivery, performance in executing primary functions and the ability to reform service delivery processes (Mentz, 1997). In countries like New Zealand it is apparent that reform-based change generates new human resource demands (Aitken,1997: 46). Personnel capacity was developed in New Zealand to meet new demands. This response is credited as a primary reason why reforms succeeded in New Zealand and other countries like Singapore (Ball and Dale, 1998; Jones, 1998; Schick, 1998). Other countries, like Tanzania (Therkildsen, 2000) and Malawi (Adamolekun, et al., 1997), have been less successful at developing personnel capacity for reform, however, and human resource limitations remain major reform constraints. In South Africa budget reforms have had limited influence on spending quality, partly because of low personnel capacity—leading to calls for government to spend more on personnel (Fölscher,2001).

Important aspects of personnel capacity required for reform include:

The need for core reform competencies: Recent research identifies core competencies to include abilities related to the preparation and interpretation of financial statements, the formulation of budgets under resource constraints, the implementation of information systems, and the development and use of performance measures (vital in ensuring effective budget implementation) (Aitken, 1997; Bell, et al. 1999). Where competencies are lacking there will be an inability to adopt reforms. This is apparently the case in a number of countries, including Malawi where “the implementation of measures aimed at improving personnel management functions has been slow” leading to “capacity overload” and reform stagnation (Adomelekun, et al., 1997: 216-218). In other governments (like New, Singapore and the state of

- 143 -

Virginia), the development of strategic competencies has been a contributing factor to reform success (Aitken, 1997; Schick, 1998; Hill and Shook, 1998; Foltin, 1999).

The need for managerial competence: Developing strategic competence for reform requires a concentrated effort “in the managerial, professional and technical fields” (Cohen 1993: 11 - 16). Bureaucratic literature related to most public organizations (particularly in the developing world) emphasizes the motivational and technical role played by such managers in organizational change (Scott, 1998). This is the case in reforms at the local government level (Vidlakova, 1993; Watson, 2000) and at higher levels (Adomelekun, et al., 1997; Aitken, 1997). Mentz (1997) calls this the ultimate level of personnel capacity because highly able managers are expected to bring out the best in their staff, developing personnel abilities and facilitating successful reform (see also Sanders, 1998).

Political representatives and civil society need capacity: Experience also shows that effective reform requires competent executive budget agencies and parliamentary support, as well as civic organizations (Hill and Shook, 1998; Peters, 1998; Foltin, 1999; Krafchick, 2001). Personnel shortfalls are usually most evident in legislative budget agencies and civic organizations, where support staff often lacks the necessary skill in analyzing budgets and in relating budget information to legislators. This was the case in New Zealand, where political decision-making capacity was enhanced to facilitate performance-based reform (Aitken, 1997: 43), and is also seen in a variety of countries from South Africa (Krafchick, 2001) to Tanzania where the lack of non-bureau capacity limits the ability of parliamentary representatives (and even executive heads) to engage in budgetary dialogue or to impact policies (Therkildsen, 2000: 65).

Process and systems capacity

Process and systems capacity is also central to reform implementation. Processes and systems order the way in which governance operations are undertaken. Where processes and systems are lacking or poorly suited to the direction of new reforms, they do not facilitate the introduction of reforms, which are themselves purposeful adaptations to existing systems. Vital processes and systems pertain to revenue and expenditure management and budgeting information provision. There are instances where reforms have succeeded only after responding to systems capacity concerns. Singapore is an example, where the Singapore Government Management System (SIGMA) facilitated reforms by providing detailed reform-based costs of costs and programs to managers and parliamentarians (Jones, 1998: 287). Other examples include Florida and Bangladesh (Khan et al., 2000). In Florida performance-based budgeting (PBB) reform was slowed by capacity problems centered on the pre-existing accounting system (OPPAGA, 1997; Cornett, 1998; Willoughby and Melkers, 2000). Performance-based reforms only had an effect when the legacy accounting system was replaced. In Bangladesh accounting information systems had to be improved to facilitate a ‘modernization’ program, because they institutionalized an outdated process in public organizations.

- 144 -

Recent comparative studies like these suggest the importance of a number of systems, including:

Budgeting and accounting systems: Hilderbrand and Grindle (1997: 46) found that problems with both kinds of system hampered reform in Bolivia, Ghana, Morocco and Tunisia. The authors found weaknesses in the areas of budgetary proposal development, reporting mechanisms, and information provision. Studies on performance-based budgeting in the United States emphasize the importance of basic budgeting systems that structure decision-making, facilitate program identification and establish commonly available budgeting databases in which information (on performance, or medium-term assets and liabilities, for example) is readily available, in appropriate formats, to a variety of users (Carter, 1994; Hill and Shook, 1998). This kind of budget information database is also emphasized in studies on developing countries like South Africa (National Treasury, 2001).

Performance-management systems : Given the fact that ROME is a results-oriented reform in which citizens demand results and results information from government, it is apparent that governments need some ability to provide such information. Indeed the earlier chapters of ROME laid great emphasis on the necessity of information provision to citizens. But many governments have no information themselves. It appears that many states in the United States, for example, are still in the process of evaluating technology needs related to performance-based budgeting adoption, and many lack systems relating performance data to varying user needs, or to the financial management processes underlying management, allocation, and incentive decisions. Some authors identify the main problem as the influence of established systems (as in Florida) while others argue that the problem is not the influence of existing systems, but a lack of financial resources required to develop new systems that relate to the old ones (Carter, 1994; OPPAGA, 1997; GASB, 1999).

Information systems: Honadle provides a list of other important systems, including those that “process information; change priorities, programs and procedures; provide feedback; and modify behavior on the basis of evaluation” (1986: 11). A specific example is the importance of intra-organizational (non-budgeting) communications systems, required for the dissemination of information about reform concepts and progress. If communication systems are lacking or poorly functioning, it is highly likely that reforms will fail at some stage of implementation because of communications failures. Where such systems are effectively evident, they facilitate reform by “ameliorating processes, simplifying expedients, and relating tasks to processes” as was the case in the INSS agency in Spain (Parrado-Diez and Ruiz-Lopez, 1997: 63).

Reform monitoring systems : Although reporting and monitoring systems are often considered products of reform initiatives, reform-oriented monitoring systems are often also necessary in governments attempting significant change. Without such systems it is difficult to analyze the degree of reform implementation or to identify where implementation might be faltering (Andrews, 2001a). These systems are

- 145 -

centrally important if governments are to learn from their reform experiences (Peters, 1998: 99). Financial management systems are difficult and costly to develop, and many governments lack the resources necessary to set up systems required for implementing complex reforms (Sabatier and Mazmanian, 1981). This results in such reforms failing in implementation. Governments in such situations, often at the sub-national level trying to copy national-level interventions, should match their systems abilities with reform types (or to reform timing), to ensure that the two are consistent (Shah, 1998: 7).

8.3.b. The factors affecting reform experience: Public sector values, mission, goals

As his second factor affecting reform performance, Shah emphasizes examining the values providing “points of reference for public sector mandates” (1998: 6). de Montricher (1998: 113) also speaks of such values, arguing that “the public sector is structured around a certain number of common values that are shared by all its organizations.” Shah opines that broad ‘societal values’ should drive governance behavior, but that such values in developing countries are “rarely addressed…because the orientation of the public sector remains ‘command and control’” (Shah, 1998: 6). The important points of reference for public values in such closed governing structures are political and managerial motivations. Effective reform implementation requires reform ‘acceptance’ within the context of the values, missions and goals of politicians, managers and entrenched incentive systems.

In ROME, because of the emphasis on civic participation, this ‘acceptance’ must of necessity center on the willingness of politicians and administrators to respond to social voices. Such ‘acceptance’ arises under certain conditions, and with certain adjustments, to the following:1. Political values, missions and goals,2. Managerial values, missions and goals, and3. Organizational and social incentives facing political representatives and managers.

Political values, missions, goals and reform acceptance

Political representatives have an interesting influence on reforms. On the one hand the last decade has seen a number of political regimes promoting neo-liberal policies and reforms because of their attractiveness in the global and business communities (Rockman, 1998). On the other hand there is evidence that political Such behavior reveals that political representatives support reform types (and degrees of reform implementation) because of what reforms contribute to political missions and goals. Reforms are ‘accepted’ as long as they further political goals, but when reforms threaten such goals, they are ignored or resisted.

Internationally accepted reforms are attractive to globally minded politicians, as long as they don’t create too many changes to ‘business as usual’. This means that reforms will be introduced (to gain global funds or acceptance) until reform consequences create incentives for political officials to impede them, either by resisting them or by ignoring

- 146 -

their potential. Coston (1998) calls this a problem of volition, while the general literature terms it poor ‘political will’ (Savoie, 1998). This problem arises for a number of reasons, mostly because reforms can clash with the values, missions and goals of politicians and political systems. This is reflected in Thompson’s (1998) discussion of public sector change, where the problem of political acceptance in reforms is described in terms of the need for congruence between the details of reform and the values of political leaders.

Political values are shaped by a number of factors:

The complexities of electoral success: Reforms can clash with political values when they make politicians vulnerable to their electorate and threaten the potential for political electoral success (a key political value, mission and goal) (Savoie, 1998). Reforms may impose periods of fiscal austerity on jurisdictions or create threatening ‘transparency’ regarding the way in which funds are allocated, for example, which both increase the exposure of politicians to their constituencies. Such reforms may be ‘introduced’ to gain political favor in neo-liberal circles, but are often poorly implemented because of the threat of constituent disfavor.

The need to influence decisions : Similarly, reforms can also threaten the ‘political’ aspect of decision-making by changing decision rules and reducing political discretion over budgetary allocations. Furthermore, by increasing the information density of decision-making, reforms test political abilities to handle new information and processes and could shift budgeting influence towards more technically-minded administrators (Therkildsen, 2000).

The timing of political benefits and costs: The threats reforms can provide into political decision-making are seldom countered by tangible benefits (apart from the initial benefit of being able to say “we’re doing something!”). Reform benefits are often located in the future, which is a problem for short-sighted politicians who support programs (like reforms) on the basis of their short-term gains (because political values, missions and goals tend to be short-term). This issue is magnified in situations where the “need for positive news about government service is urgent” (Carter, 1994: 28). In these situations the issue of political acceptance is particularly evident, reflected in Carter’s idea of a shrinking window of reform opportunity: “as the window of opportunity shrinks for lawmakers facing term limits, so might the chances for performance budgeting reforms. Changing systems and institutions takes time, which means that the promise of improved government service is years away. Elected officials cannot wait” (Carter, 1994: 28).

The argument is simply that politicians will resist reforms (either actively or passively) if the reforms, when judged according to political values, missions and goals, provide more potential threats than gains. This lack of political acceptance can thwart reform implementation as politicians refuse funding to reform projects or ignore the reform processes, information or other products, sponsoring the ‘business as usual’ budget and allocations processes instead. The problem of political acceptance is being tackled in some governments by linking reform motivation with political success (especially where

- 147 -

there is a social demand for reform), ensuring that performance measures matter to politicians (what Governor Barnes of Georgia called ‘kitchen table measures’), and developing critical competencies and appropriate technology such that the use-value of information is maximized for politicians (Kline, 1997; Foltin, 1999). In other initiatives, reformers are attempting to structure interventions so that they provide immediate returns to political representatives: “Without short-term wins, too many people give up and actively join the ranks of those people who have been resisting change” (Kotter, 1995 from Khan, 2000).

All these efforts aside, political decision-making remains largely untouched by reforms: in the United States legislators are the missing party in many performance-based initiatives, while studies in other countries tell similar stories (Adamolekun et al. (1997: 213) finds a similar lack of political will in Malawian budget reforms, as do Hulme and Sanderatne (1997) with regard to budget reform in Sri Lanka). The precise manifestation of low political acceptance differs from case to case: in some instances it shows as active political resistance to reform, in others it reflects as political disinterest in reform. Both types of manifestation threaten reform effectiveness.

Managerial values, missions, goals and reform acceptance

The literature shows that managerial acceptance is also critical to reform success. In recent commentary regarding the re-invention movement in the United States, Sanders finds that reform-accepting managers are a crucial ingredient to reform implementation, describing such managers as “catalysts who actually sparked the change process; effectively linked it to some larger, exogenous agenda; and managed the transformation to its conclusion” (Sanders, 1998: 33). Managers that fail to accept reforms can just as easily thwart its progress, resisting the change process, closing the organization off to external influences, and focusing their management decisions and activities away from reform.

The general issue regarding managerial acceptance centers on the need for reforms to relate to the values of administrators: if managers view reforms as a threat to them that is not countered by benefits (reforms don’t make jobs easier or allow some flexibility in managerial budgets, for example) they will not support the reforms. This is revealed in the ways in which reforms are channeled by administrative managers, usually towards their established interests, and not towards the goals of redressing failures in budgeting processes (that often arise because of administrative interest maximization).

In this light, reformers need to consider the following when implementing reform:

The effect of reforms on administrative discretion: Management acceptance will be limited where administrative entities feel that reforms constrict agencies more often than reward them. It is typical for civil servants to be anxious about changes because of the potential reforms bring for increased workload and increased job uncertainty (Hallyar, 1994; Therkildesn, 2000: 63). Administrators will resist reform because of this anxiety, as well as the fear that reforms will erode any influence they have

- 148 -

(through information asymmetries and such) in the public organization. This resistance rationale is particularly expected in developing countries where non-reformed bureaucracies are often closed and facilitate bureaucratic power structures and influence (Shah, 2000).

Reform uncertainty: Managerial resistance is also expected in situations where political reform ‘champions’ are expected to depart from their pro-reform position or to lose power. Related to what Savoie (1998) calls the ‘disbelief culture’, managers will wait to see the strength of political support for reforms before pledging their commitment, so that they don’t find themselves mid-way through reform when political dynamics change. In this light Carter (1994: 28) asks: “will agencies participate wholeheartedly and challenge the status quo when they know lawmakers might not be around to reward or penalize?” The experience with performance-based budgeting in Minnesota is further instructive in this regard: “Agency staff…said that agencies have viewed performance measurement as an idea that was not relevant to decision makers” and they thus declared its failure saying that it “would not outlive each existing administration”(Minnesota,1994).

Reform opportunities for managers: The literature suggests that managers “may support reform initiatives that create new opportunities for them” (Therkildsen, 2000: 63). In response to this line of thought some states allow agencies to develop their own measures, hold strategic briefings to show how performance information and other aspects of reform relate to the accepted mission of an administrative entity, or mandate aspects of the reform in everyday management (to get managers to accept reforms through repeated use of them) (Carter, 1994; Melkers and Willoughby, 1998). These kinds of interventions are designed to enhance managerial acceptance of reform by focusing aspects of the reform on values and goals and missions shared by managers.

Experience from financial reform in Bangladesh illustrates the threat factor reforms pose to managers: “most people at operational levels viewed the project as a threat to their established culture, tradition and the way they were used to doing things” (Khan, et al., 2000: 36). This kind of sentiment made managers reluctant supporters of reform in Bangladesh, and threatens managerial reform in many other countries. In Tanzania, for example, Therkildsen (2000: 63) writes that “there is substantial resistance in many of the affected line ministries to changes in power relations that will result” from reform. The issue is further reflected in Lam’s (1997: 414) comment on reform experience in governments like Hong Kong: “success of public sector reform is unlikely if staff regard themselves as being involuntarily pledged to externally imposed standards” and feel threatened by such.

In most reforms it seems that managers are more readily affected than are politicians (with managers using performance data in the states more than politicians, for example), but insufficient acceptance of reforms is still proving a problem for reform adoption (Kline, 1999). This is revealed in the development of reforms to reflect managerial interests, rather than the broader political or social interests. Politicians in some states in

- 149 -

the United States, for example, find that administrative heads are using performance information requirements to bulk their budget requests up even more than before, thus making budgets harder to navigate and evaluate and entrenching the managerial information advantage. In the case of local government reforms in South Africa it appears that municipal managers are pursuing planning-programming-based reforms with a similar agenda (Andrews, 2001b). For ROME to succeed, managers must have the incentives to pursue reform effectively or else it will fail.

Incentives, values, missions, goals and reform acceptance

The third way in which values, missions and goals influence reforms is through established incentives. This is of central importance in ROME, where civic influence and evaluations are seen to be distinct institutional sources of incentives for political representatives and managers to pursue results-based reform. These incentives often clash, however, with other incentives pervading public organizations. Politicians and administrators alike have standing incentives to use specific types of information, and to behave in certain ways. These incentives are shaped by established processes and norms, which can differ with those required for successful ROME adoption. In such situations it is possible that incentives reflecting non-reform values, missions and goals will continue to drive behavior (Mummert, 1999).

The literature suggests, for example, that the process-oriented, input-focused budgetary systems in place in most government create incentives that run counter to a results-oriented approach, and shape the information use of administrators and politicians. As another example, many governments around the world who are trying to develop performance-based budgeting are expected to find their initiatives undermined by constant managerial incentives to focus on process (because it is process that auditors evaluate, not performance) (Andrews, 2001a). As long as organizational incentives are incompatible with reforms, it is unlikely that reforms will be effectively adopted.

Essentially, whether one is dealing with incentives shaping behavior as a result of established processes or because of entrenched social norms, these incentives must be considered when reforms are developed. When incentives are considered as such they are generally seen to motivate people to action that runs counter to new reforms, requiring reform designers to actively develop pro-reform incentives to counter pre-existing incentives and encourage reform adoption. In relation to this, Willoughby and Melkers (2000: 107) argue that there is a lack of leadership and acceptance in many performance-based reforms as well as “a lack of incentives or an inappropriate use of disincentives related to the conduct of performance measurement,” which “often shelve such reform attempts prematurely.” These authors call for an “incentive strategy for the use of performance-based information.” Similarly, Khan et al. (2000: 36) find that organizational incentives associated with established norms and practices impeded the adoption of financial management reforms in Bangladesh. They conclude that, “to be successful, the focus of change must…be at the group level for influencing their norms and roles in the project implementation process.” This sentiment is evident in Frederickson and Perry (1998) who emphasize promoting reform by shaping

- 150 -

organizational reward and incentive structures to ensure the continuity of change initiatives.

ROME requires that two institutional areas are strong sources of incentives:

Civic society: The institutional environment must be one conducive to participation and a high degree of social organization. Only in these situations will citizens be able to pressure governments into producing results and pursuing results oriented reforms. The important institutions underlying this environment are discussed in chapter five and relate to political, information and administrative structures in developing countries. Reformers should build on positive manifestations of such institutions and should work fervently to correct negative institutional parameters.

Evaluations: ROME also posits an important role for evaluations in creating incentives for ongoing results oriented reform in government. For this reason, as with institutional developments in civil society, it is proposed that evaluations development be a starting point in the ROME implementation program.

8.3.c. The factors affecting reform experience: Public authorizing environment

The third factor in Shah’s model is the public authorizing environment. The authorizing environment involves the laws, processes and lines of interaction that structure what organizations and organizational members can or cannot do. Such authorizing factors pertain particularly to the following key aspects of governance and reform: revenue authority (and autonomy), authority over conditions of service (and hence ability to forecast such expenses), process authority, decision-making authority and authority over the kinds of services provided.

The authorizing environment influences reform implementation by shaping the authority government entities have to adopt (or resist adopting) certain practices, providing the organizational objectives (or values), structuring interaction, creating incentives, and stipulating decision-rules (Sabatier and Mazmanian, 1981: 12). If governments, or users in governments, are not authorized to adopt new ROME reforms, such adoption will be hindered. Similarly, if reforms are entrenched in influential and enforced law, governments are forced to reform (they have no authority to resist). There are two important aspects of authority: legal authority and procedural and organizational authority.

Legal authority

Formal governance processes are tightly legislated. This legislation is generally the focal point of f institutional research (Poterba, 1996; Alesina and Perotti, 1996; Alesina et al., 1999; Bales and Tieslau, 2000). New reforms (or reform elements) cannot be adopted if they conflict with existing legislation, but new reforms are more likely to be adopted and successfully implemented if they are congruent with, or enforced by, legislation (Campos and Pradhan, 1996).

- 151 -

The influence of legal authority on reform adoption is not to be underestimated, as cultures within government organizations are based almost entirely on law, with good reason (O’Leary and Straussman, 1993). Legal limits and definitions are meant to keep public organizations in check, and as such legislation is always going to be a factor to consider when altering financial management and budgeting approaches. Legislation lays down the basis of the institutional environment in which governments exist, defining capacities and roles of such governments, and in a very real way legislation also shapes government values (or at the very least reflects socio-political values regarding governance models) (Riker, 1980; Peters, 1995; Mueller, 1999).

Changing the institutional character of a government through reform requires an intimate knowledge of the pre-existing legal frameworks (and indeed the values reflected by such frameworks). Not only is there reason to believe that reforms will be influenced by legislation because of path dependency (the rules created in the past tend to influence rule creation in the future), but laws also provide valuable information about social values, which should inform reform development. Authors like Bush (1994), Blau (1996) and Uslaner (1998) provide interesting perspectives on the issue of institutional change and the relationship between laws and values. Without locating common ground between them, or suggesting a simple relationship between these concepts, it can be said that there is interaction between laws (especially those deeply nested in constitutions or in social systems) and values, which complicates any changes to (or within) legal structures. Legal authority, and the social legitimacy reflected in such, is argued to be a vital reform influence.

Procedural and organizational authority

Laws are not the only authority mechanisms influencing reform implementation. Reforms are often grafted into existing budget processes characterized by established procedures that act as rules binding budgeting behavior, which also influence reform adoption (Wildavsky, 1988). Indeed, the importance ascribed to formal legislation can easily be extended to procedural regulations, whether formal or informal. March and Olsen (1995: 31) develop their ‘logic of appropriateness’ theory on the understanding that behavior is shaped by a wide variety of institutional arrangements, including formal and informal procedures, that determine what is “appropriate, exemplary, natural or acceptable behavior.” As with laws, procedures are closely related to organizational values (both reflecting values and shaping values) and determine what is acceptable: if reforms are ‘acceptable’ within established organizational processes, then they are adopted; if not, they will be opposed (Xavier, 1998).

- 152 -

The required degree of compatibility between reforms and organizational procedures will vary from place to place (Mummert, 1999). At minimum, however, the literature does suggest that reforms need to have some link with pre-existing social values (or with changing social values), such that reforms have appeal within social authority structures (like procedures) developed from such value bases (Tool, 1979; Bush, 1994). This kind of link suggests that reform should be incremental and not sweeping, causing “minimal dislocation” rather than radical change and underlies the argument that ROME should begin with the transformation of social institutions. To paraphrase Bush (1994), “new reforms should have a minimally disruptive effect on the instrumental patterns of behavior and procedures the community relies on to sustain its regular interactions.” The kinds of procedures that should be considered when reforms are developed include those regulating budget management, reporting, decision-making and allocations. These typically emphasize an input and control bias and require role players to adhere to process above all else (White, 1994). These procedures typically centralize authority and limit discretion or openness in the public budgeting process, complicating reforms based on introducing decentralized, transparent budgeting processes.

8.4. Conclusion: Implementing ROME by increasing ‘reform space’

This chapter has discussed the idea of institutionalizing ROME as well as the institutional factors that require attention in implementation. The chapter’s argument can be summed in the idea that ROME implementation requires expansion of a reform space to facilitate effective results oriented tool and rule adoption. This is shown in figure twenty four.

Figure twenty four: Expanding the reform space for ROME reform

- 153 -

The top figure shows the three factors affecting the public sector institutional environment, as in figure twenty three. The reform space is identified by the small letter ‘a’ in this figure. The second figure shows what happens to the reform space when the three factors are expanded. Expanding all three factors yields a reform space equal to the area a,b,c,d,e,f, and g. This is achieved by expanding ability, acceptance and authority in the reform process.

The first step in the ROME implementation process involves expanding this reform space, particularly by enhancing citizen activity in the governance process and creating incentives for governments to respond to citizens. This amounts to a concentration on expanding the acceptance of the reform (enlarging the top left circle). At the same time, attention must be given to authority structures, particularly in decentralized government structures, that either facilitate or hinder results oriented responses by public entities. Once acceptance and authority have been expanded, experience in areas like Malaysia and Rajasthan suggest that governments will move to expand their own abilities, and will have a large reform space into which tools like performance-based budgeting can be located.

- 154 -

CONCLUSION:ROME: AN EFFECTIVE PUBLIC SECTOR REFORM

FOR THE DEVELOPING WORLD

This monograph has presented a new reform model for the developing world. This model, the participatory Results Oriented Management and Evaluation model (ROME), is essentially a program based on citizen-led, results-oriented government.

ROME tackles the problems of government by adjusting the institutional environment in which public organizations are located. This involves institutionalizing citizens as the core principals, clients and evaluators of government. It makes sense that this will lead to a results-orientation in government because citizens demand results (not inputs). It makes further sense that this approach should be localized because citizens can influence local governments better than they can national governments.

Some don’t believe that a results-oriented decentralized system can work in the developing world (Schick, 1998). They prefer a reform model that starts with creating bureaucratic controls. While they believe such controls facilitate greater accountability, the ROME perspective holds that it is citizens and social institutions that do so. Surely governments are accountable when people hold them accountable? Surely more bureaucratic controls lead to less government by the people?

ROME says, let the people govern themselves by creating institutions conducive to such. Paraphrasing Turner and Hulme (1997), “the issue of governance reform is far too important to be left out of the people’s hands.”

- 155 -

SELECTED REFERENCES

Abdul Karim, Muhammad Rais Bin (Malaysian Administrative Modernization and Management Planning Unit, Prime Minister’s Department, Kuala Lumpur, Malaysia) Public-Private Sector Cooperation for Development in Malaysia, Asian Review of Public Administration, July-December 1997.Adamolekun, L, Noel Kulemeka, and Mouftaou, Laleye (1997). Political transition, economic liberalization and civil service reform in Malawi. Public Administration and Development, Vol. 17, 209-222.Aitken, Judith (1997). Managing the new organization: some problems of institutional transition—a New Zealand perspective. Public Administration and Development, Vol.17, 41-48.Alford, John, and Jeanette Baird (1997) Performance Monitoring in the Australian Public Service: a Government-Wide Analysis, Public Money and Management, 49-57.Ammons, David N. (2000). Benchmarking as a performance management tool: Experiences among municipalities in North Carolina. Journal of Public Budgeting, Accounting & Financial Management, Vol. 12, No. 1, 106-124.Assibey-Mensah, George O. (1997) Budgeting in Ghana: Innovations, Problems, and Prospects. Public Budgeting & Finance Vol.17, No.3, 89-111.Atkinson, Sarah, Regianne Leila Roilm Medeiros, Paulo Henrique, Lima Oliviera, Ricardo Dias de Almeida (2000). Going down to the local: incorporating social organization and political culture into assessments of decentralized health care. Social Science and Medicine, Vol.51, 619-636.Bahl, Roy (1999). Fiscal Decentralization as Development Policy. Public Budgeting & Finance, Vol.19, No.2, 59-75.Bale, Malcolm, and Tony Dale (1998). Public Sector Reform in New Zealand and Its Relevance to Developing Countries. The World Bank Research Observer, Vol.13, No.1, 103-121.Ball, Ian (1994). Reinventing government: Lessons learned from the New Zealand Treasury. The Government Accountants Journal, Vol.43, No.3, 19-28.Ball, Ian, Tony Dale, William D. Eggers, and John Sacco (1999) Reforming Financial Management In the Public Sector: Lessons U.S. Officials Can Learn From New Zealand. Reason Public Policy Institute Policy Study No. 258. Available at: http://www.rppi.org/ps/ps258.htm.Batley, Richard (1999a.). The New Public Management in Developing Countries. Journal of International Development, Vol.11, 755-760.Batley, Richard (1999b.). Urban Water in Zimbabwe: Performance and Capacity Analysis, Paper 33 of the Role of Government in Adjusting Economies, University of Birmingham. Available at: http://www.bham.ac.uk/IDD/rog33.htm. (June 13, 2000).Beal, Jo (1997). Social Capital in Waste—A Solid Investment? Journal of International Development, Vol.9, No.7, 951-961.Bebbington, Anthony (1991). Sharecropping Agricultural Development. Grassroots Development, Vol. 15, No.2, 20-30.Bendahmane, Diane B. (1990) New Perspectives on Evaluation. Grassroots Development, Vol. 14, No.1, 35-37.

- 156 -

Bendahmane, Diane B. (1991) Performance Review for NGOs. Grassroots Development, Vol. 15, No.2, 31-37.Blair, H. (1997). Democratic Local Governance in Bolivia. CDIE Impact Evaluation No.6, United States Agency for International Development: Washington, D.C.Blair, Harry (2000). Participation and Accountability at the Periphery: Democratic Local Governance in Six Countries. World Development, Vol. 28, No. 1, 21-39.Brandl, John E. (1998). Money and Good Intentions are not enough: or, why a Liberal Democrat Thinks States Need both Competition and Community. Washington, D.C.: Brookings Institution.Breslin, Patrick, and Duncan Campbell (1999). Giving a community the authority to distribute public funds creates a dynamic of efficiency and accountability. Grassroots Development, Vol. 22, No.1, 28-30.Brignall, Stan (1997). Performance Measurement and Change in Local Government: A general Case and a Childcare Application, Public Money and Management, 23-31. Brinkerhoff, Derek W. (2000). Democratic Governance and Sectoral Policy Reform: Tracing Linkages and Exploring Synergies. World Development. Vol. 28, No.4, 601-615.Brown, Richard E., Mark J. Myring, and Cadillac G. Gard (1999) Activity-Based Costing in Government: Possibilities and Pitfalls. Public Budgeting & Finance, Vol.19, No.2, 3-21.Campos, Jose Edgardo, and Hadi Salehi Esfahani (2000). Credible Commitment and Success with Public Enterprise Reform. World Development, Vol. 28, No.2, 221-243.Campos, Nauro, F. and Jeffrey B. Nugent (1999). Development Performance and the Institutions of Governance: Evidence from East Asia and Latin America. World Development, Vol.29, No.3, 439-452.Chiu, Ng Kam (1997). Service Targets and Methods of Redress: the impact of accountability in Malaysia. Public Administration and Development, Vol.17, 175-180.Christensen, Linda F. and Douglas Sharp (1993). How ABC Can Add Value to Decision Making. Management Accounting, May 1993, 38-42.Christensen, Tom, and Per Laegreid (1999). New Public Management—design, resistance, or transformation? : A study of how modern reforms are received in a Civil Service System. Public Productivity and Management Review, Vol.23, Issue 2, 169-193.Colon Carlo, Ileana M. (1997). The role of government auditors in combating corruption—The Puerto Rican experience. The Government Accountants Journal, Vol.46, No.3, 30-34.Cooper, Robin, and Regine Slagmulder (1999). Strategic Cost management. Management Accounting, January 1999, 16-17.Coston, Jennifer M. (1998). Administrative Avenues to democratic Governance: the Balance of Supply and Demand. Public Administration and Development, Vol.18, 479-493.Dale, B.G. (1994). A Framework for Quality Improvement in Public Sector Organizations: A Study in Hong Kong. Public Money and Management, April-June, 31-36.Davies, Ken, and Peter Hinton (1993). Managing Quality in Local Government and the Health Service. Public Money and Management, January-March 1993.Davies, Marlene, and Elaine Shellard (1997). The value of performance measurement in the United Kingdom. The Government Accountants Journal, Vol.46, Issue 3, 48-51.

- 157 -

De Souza, Herbert (1995). Enlisting the Power of Citizenship to Fight Poverty. Grassroots Development, Vol. 22, No.1, 49-50.Desai, Vandarda, and Rob Imrie (1998). The new managerialism in local governance: North-South dimensions. Third World Quarterly, Vol.19, No.4, 635-650.Efficiency Unit, Hong Kong Government (2000). News, Information and Views on Public Sector Reform. Available at http://www.info.gov.hk/eu/psrhk/hist_con.htm (June 22, 2000).Ellis, Roger and Dorothy Whittington (1994). Health care Quality Assurance: Techniques and Approaches. Public Money & Management, April-June, 23-29.Ewer, Sid R. (1999). Managerial cost accounting: A step toward accountability and reliable costing of federal programs. The Government Accountants Journal, Vol.48, No.1, 48-53.Fielding Smith, James (1999). The Benefits and Threats of PBB: An Assessment of Modern Reform. Public Budgeting and Finance, Vol.19, No.3, 3-15.Foltin, Craig (1999). State and local government performance: It’s time to measure up! The Government Accountants Journal, Vol.48, Issue 1, 40-46.Friedman, Mark (1995). From Outcomes to Budgets. Center for the Stduy of Social Policy, Available at: http://www.cssp.org/kd13.htm (August 1, 2000).Garrison, John W. II, and Leilah, Landim (1995). Harvesting the Bounty of Citizenship: The Fight Against Hunger and Poverty in Brazil. Grassroots Development, Vol. 19, No.2, 38-48.Geiger, Dale R. (1995). Motivating contingencies at early adopters of federal cost management accounting systems. The Government Accountants Journal, Vol.44, No.1, 17-29.Geiger, Dale R. (1998). Practical issues in managerial cost accounting. The Government Accountants Journal, Vol.47, No.2, 48-53.Geiger, Dale R. (1999). Practical issues in cost driver selection for managerial costing systems. The Government Accountants Journal, Vol.49, No.3, 32-39.Gillibrand, Angela, and Brain Hilton (1998). Resource Accounting and Budgeting: Principles, Concepts, and Practice—The MoD Case. Public Money & Management, April-June, 21-28.Goddard, Andrew and Kean Ooi (1996). Activity-Based Costing and Central Overhead Cost Allocation in Universities: A Case Study. Public Money and Management, January-March, 31-38.Gopakumar, K. (1997). Public feedback as an aid to public accountability: reflections on an alternate approach. Public Administration and Development, Vol.17, 281-282.Granof, Michale H., David E. Platt, and Igor Vaysman (2000). Using Activity-Based Costing to Manage More Effectively. The PricewaterhouseCoopers Endowment for The Business of Government Grant Report, Available at: endowment.pwcgolbal.com (June 30, 2000).Grindle, Merilee S. (1997). Divergent Cultures? When Public Organization Perform Well in Developing Countries. World Development, Vol.25, No.4, 481-495.Haggard, Stephen (1999). Governance and Growth: Lessons from the Asian Economic Crisis. Asian-Pacific Economic Literature, Vol. 13, Issue 2, 30-42.

- 158 -

Hanna, Nagy (1999). The Annual Review of Development Effectiveness: Towards a Comprehensive Development Strategy, Report No. 19905, Washington, D.C.: World Bank.Haque, N.U. and J. Aziz (1999). The quality of governance: ‘second-generation’ civil service reform in Africa. Journal of African Economies, Vol.8, Supplement 1, 68-106.Henry, Laurie, J. and Panu Attavitkamtorn (1999). Governmental Accounting and Auditing in Thailand: An overview and some suggestions for improvement. The International Journal of Accounting, Vol. 34, No. 3, 439-454.HM Treasury (1988). Policy Evaluation: A Gudie for Managers. HMSO, Annex C.Hubbard, Michael, Simon Delay, and Nick Devas (1999). Complex management Contracts: The Case of Customs Administration in Mozambique, Public Administration and Development, Vol. 19, 153-163.Hubbard, Michael, and Marisol Smith (1999). Does New Public Management work in Reforming the State’s Role in Agricultural Marketing in Developing Countries? Journal of International Development, Vol. 11, 785-790.Hulme, David, and Nimal Sanderatne (1997). The Toothless and the Muzzled: Public Accountability, Public Expenditure Management and Governance in Sri Lanka. Public Policy and Management Working Paper Series, Paper No. 2. Institute for Development Policy and Management, University of Manchester.Huther, Jeff and Anwar Shah (1998). A Simple Measure of Good Governance and its Application to the Debate on the Appropriate Level of Fiscal Decentralization. World Bank Policy Research paper Series No.1894, March 1998. Washington, D.C.: World Bank.Inter-American Development Bank (2000) ensuring Transparency and Accountability in Hurricane Reconstruction and Transformation. Available at: http://www. Iadb.org/regions/re2/consultative_group/groups/transparency_workshop_3.htm (August 24, 2000).Jenkins, Rob, and Anne Marie Goetz (1999). Accounts and Accountability: theoretical implications of the right-to-information movement in India. Third World Quarterly, Vol. 20, No.3, 602-622.Jones, David Seth (1998). Recent Budgetary reforms in Singapore. Journal of Public Budgeting, Accounting & Financial Management, Vol. 10, No.2, 279-310.Jones, Rowan (1998). The Conceptual framework of Resource Accounting. Public Money & Management, April-June, 11-16.Kaplan, Robert S. and David P. Norton (1992a). The balanced Scorecard—Measures That Drive Performance. Harvard Business Review, January-February 1992, 71-79.Kaplan, Robert S. and David P. Norton (1992b). Linking the balanced scorecard to strategy. California management Review, Vol. 39, No.1, 53-79.Kaufmann, Daniel, Aart Kraay, and Pablo Zoido-Lobaton (1999). Aggregating Governance Indicators. World Bank Policy Research Working Paper 2195, Washington, D.C.: World Bank.Kaufmann, Daniel, Aart Kraay, and Pablo Zoido-Lobaton (2000). Governance Matters: From Measurement to Action. Finance & Development, Vol.37, No.2.Kaul, Mohan (1997). The New Public Administration: management innovations in government. Public Administration and Development, Vol.17, 13-26.

- 159 -

Khan, Zakir Ahmed, Nigel Thornton, and Mike Frazier (2000). Experience of a Financial Reform Project in Bangladesh. Public Administration and Development, Vol. 20, 33-42.Khumawala, Saleha B., and Steve Buchheit (1999). Costs and benefits of International Governmental Accounting Standards: The Case of Lesser Developed Countries. Public Fund Digest, Vol. X, No. 1, 19-34. Kline, James J. (1997). Local government outcome based performance measures and customer service standards: Has their time come? The Government Accountants Journal, Vol.45, Issue 4, 46-49.Kousmin, Alexander, and Judy Johnston (1998). An international symposium on auditing Australian public sector innovations: From public administration to public management. Public Productivity & Management Review. Vol.21, Issue 4, 344-350.Lam, Jerman T.M. (1997). Transformation from public administration to management: Success and challenges of public sector reform in Hong Kong. Public Productivity and Management Review, Vol.20, Issue 4, 405-418.Larbi, George A. (1998). Institutional Constraints and Capacity Issues in Decentralizing Management in Public Services: The case of Health in Ghana. Journal of International Development, Vol. 10, 377-386.Lee, Robert D, Jr., and Robert C. Burns (2000). “Performance Measurement in State Budgeting: Advancement and Backsliding from 1990 to 1995,” Public Budgeting and Finance, Vol.20, No.1, 38-54.Likierman, Andrew (1998). Resource Accounting and Budgeting—Where are we Now? Public Money & Management, April-June, 17-20.Lipe, Marlys Gascho, and Steven E. Salterio (2000). The Balanced Scorecard: Judgmental Effects of Common and Unique Performance Measures. The Accounting Review, Vol.75, No.3, 283-298.Lu, Haoran (1998). “Performance Budgeting Resuscitated: Why is it still inviable?” Journal of Public Budgeting, Accounting & Financial Management, Vol.10, No.2, 151-172.McGraw, Dickenson (2000). Governmental Effectiveness: Competition, Community, and Innovation. Public Productivity and Management Review, Vol.23, Issue 3, 412-416.McNabb, David E., and F. Thomas Sepic (1995). Culture, climate, and total quality management: Measuring readiness for change. Public Productivity and Management Review, Vol.18, Issue 4, 369-382.MacManus, Susuan A. (1996). Democratizing the Capital Budget Planning and Project Selection Process at the Local level: Assets and Liabilities. Public Budgeting & Financial Management, Vol.8, No.3, 406-427. Martin, Lawrence L. (1997). Outcome Budgeting: A New Entrepreneurial Approach to Budgeting. Journal of Public Budgeting, Accounting & Financial Management, Vol. 9, No. 1, 108-126.Matheson, Alex (1998). Governing Strategically: the New Zealand experience. Public Administration and Development, Vol. 18, 349-363.Melkers, Julia, and Katherine Willoughby (1998). “The State of the States: Performance-Based Budgeting Requirements in Forty-Seven Out of Fifty,” Public Administration Review, Vol.58, No.1, 66-73.Mistry, M.D. (2000) Influencing the Accountability of Government to the Poor: participatory Budget Analysis in Western India. Presentation at the Participation thematic

- 160 -

Group and the South Asia Social Development Unit, The World Bank, Thursday July 27, 2000.Mitchell, Mike (1996). Activity-Based Costing in UK Universities. Public Money and Management, January-March, 51-57.Mohamad, Mustapha (Public Service Department, Malaysia) Public Service Management: Management Appraisal, Incentives and Sanctions, Asaan Review of Public Administration, January-June 1997Mohan, Giles, and Kristian Stokke (2000). Participatory development and empowerment: the dangers of localism. Third World Quarterly, Vol.21, No.2, 247-268.Mountfield, Robin (1997). Organizational reform within government: accountability and policy management. Public Administration and Development, Vol.17, 71-76.Mullins, Daniel R. and C. Kurt Zorn (1999) Is Activity-Based Costing Up to the Challenge When it Comes to Privatization of Local Government Services? Public Budgeting & Finance, Vol.19, No.2, 37-57.Niesner, Helen (1999). Local government auditing—improving the performance of government in the next century. The Government Accountants Journal, Vol.48, Issue 4, 32-38.Nickson, Andrew (1999). Does the NPM work in Less Developed Countries? The Case of the urban water supply sector. Journal of International Development, Vol. 11, 777-783.Office of Program Policy Analysis and Government Accountability, State of Florida (1997). Performance-Based Program Budgeting in Context: History and Comparison. Tallahassee: State of Florida.Padilla, Agnes D. ( Civil Service Commission, Republic of the Philippines), An Enabling Government, Asian Review of Public Administration, July-December 1997.Parrado-Diez, Salvador, and Joaquin Ruiz-Lopez (1997) The Path of Quality in a Spanish Autonomous Agency. Public Productivity and Management Review, Vol.21, Issue 1, 56-69.Pelling, Mark (1998). Participation, Social Capital and Vulnerability to Urban Flooding in Guyana. Journal of International Development, Vol.10, 469-486.Pendlebury, Maurice, and Yusuf Karbhari (1998). Resource Accounting and Executive Agencies. Public Money & Management, April-June, 29-34..Perrot, Bruce E. (1996). Managing Strategic issues in the Public Service. Long Range Planning, Vol.29, No.3, 337-345.Peterson, Stephen B. (1998). Saints, demons, wizards, and systems: why information technology reforms fail or underperform in public bureaucracies in Africa. Public Administration and Development, Vol.18, 37-60.Polidano, Charles (1998). Public Sector Reform in Developing Countries: The State of Practice. Journal of International Development, Vol. 10, 373-375.Polidano, Charles (2000). Measuring Public Sector Capacity. World Development, Vol. 28, No.5, 805-822.Putnam, Robert D. (1993). Making democracy Work: Civic Traditions in Modern Italy. Princeton University Press: Princeton, New jersey.Reinikka, Ritva (1999). Using surveys for public sector reform. World Bank Premnotes Number 23, Washington, D.C.: World Bank.

- 161 -

Recascino Wise, Lois (1999). The use of innovative practices in the public and private sectors: The role of organizational and individual factors. Public Productivity and Management Review, Vol. 23, Issue 2, 150-168.Rivenbark, William C. and K. Lee carter (2000). Benchmarking and Cost Accounting: The North Carolina Approach. Journal of Public Budgeting, Accounting & Financial Management, Vol.12, No. 1, 125-137.Robinson, Marc (1998). Capital charges and capital expenditure decisions in core government. Journal of Budgeting, Accounting & Financial Management, Vol.10, No.3, 354-374.Rodgers, Jacci L., S. Mark Comstock, and Karl Pritz (1993). Customize your Costing System. Management Accounting, 31-32.Rodriguez, Justine Farr (1995). The usefulness of cost accounting in the federal government. The Government Accountants Journal, Vol.44, No.1, 31-35.Russell, Steven, Sara Bennett, and Anne Mills (1999). Reforming the Health Sector: Towards a Healthy New Public Management. Journal of International Development, Vol. 11, 767-775.Schick, Allen (1998). Why Most Developing Countries Should Not Try New Zealand Reforms. The World Bank Research Observer, Vol.13, No.1, 123-131.Schneider, Hartmut (1999). Participatory Governance for Poverty Reduction. Journal of International Development, Vol. 11, 521-534.Shand, David (1998). Budgetary Reforms in OECD Member Countries. Journal of Public Budgeting, Accounting & Financial Management, Vol.10, No.1, 63-88.Shih-Jen, Kathy Ho, and Linda Achey Kidwell (2000). A survey of management techniques implemented by municipal administrators. The Government Accountants Journal, Vol.49, Issue 1, 46-52.Sim, Pah Suk (1999). Globalization of human resource management: A cross-cultural perspective for the public sector. Public Personnel Management, Vol. 28, Issue 2, 227-243.Simpson, Wane K. and Michale J. Williams (1996). Activity-based: Costing, management and budgeting. The Government Accountants Journal, Vol.45, Issue 1, 26-28.Stratton, William O. (1993). ABC: An All-Purpose Solution for Financial Reporting. Management Accounting, May 1993, 44-49.Therkildsen, Ole (2000). Public Sector Reform in a Poor, Aid-dependent country, Tanzania. Public Administration and Development, Vol. 20, 61-71.Tierney, Cornelius, E. (1994). Cost for government: What Costs: How much accounting? The Government Accountants Journal, Vol.43, No.1, 5-9.United Nations Development Programme (1999). Programme Bolivia 21, Summary. Available at: http://www3.undp.org/c21/prog/bolivia.htm (August 17, 2000). United Nations (2000). Local Government in Asia and the Pacific: A comparative analysis of fifteen countries. Available at: http://www.unescap.org/huset/lgstudy/comparison1.htm. (July 27, 2000).Valenzuela, David (1999). From Conflict to Cooperation: Local Development in Nejapa. Grassroots Development, Vol. 22, No.1, 4-10.Vance, Marion Ritchey (1993). Grassroots Development Results—Widening the Lens. Grassroots Development, Vol. 17, No.1, 42-43.

- 162 -

Weintraub, Daniel M. (1993). Budgeting for results.(California’s performance-based budgeting). State Legislatures, Vol.19, No. 6, 19-21.Williams, Clif, and Ward Melhuish (1999). Is ABCM Destined for Success of Failure in the Federal Government? Public Budgeting & Finance, Vol.19, No.2, 22-36.Willoughby, K., and Julia Melkers (2000). “Implementing PBB: Conflicting Views of Success,” Public Budgeting and Finance, Vol.20, No.1, 105-120.World Bank (2000a). Accountability, Transparency and Corruption in Decentralized Governance. Available at: http://www1.worldbank.org/publicsector/decentralization/accountability.htm. (August 24,2000).World Bank (2000b). Bolivia—Decentralization Reform APC. Available at: http://www.worldbank.org/pics/pid/bo68134.txt (August 24, 2000).Yang, Gilbert Y., and Roger C. Wu (1993). Strategic Costing & ABC. Management Accounting, May 1993, 33-37.Younis, Talib, and Tony Boland (1997). Towards Total Quality management in a Supportive Employment Workplace. Public Money & Management, January-March, 55-62.Xavier, John Anthony (1998). Budget Reform in Malaysia and Australia compared. Public Budgeting and Finance, Vol.18, No.1,99-118.

- 163 -