the future demand for employability skills: a new

THE FUTURE DEMAND FOR EMPLOYABILITY SKILLS:

A NEW DIMENSION TO LABOR MARKET FORECASTING

IN AUSTRALIA

Alexis Esposto

Swinburne University of Technology

and

G.A.Meagher

Centre of Policy Studies, Monash University

1. INTRODUCTION

In recent years there has been considerable interest in employability skills, both in Australia and

overseas. In 2001, the Australian Chamber of Commerce and Industry and the Business Council

of Australia undertook a major research project designed to provide the Department of

Education, Science and Training with a detailed understanding of the employability skills needs

of industry. The research was published by the Commonwealth of Australia in 2002 as the

report Employability Skills for the Future.

In the report, employability skills are defined as “skills required not only to gain employment,

but also to progress within an enterprise to achieve one’s potential and contribute successfully to

enterprise strategic directions”. The report also identifies an Employability Skills Framework

which incorporates eight key skill groupings:

• communication skills,

• team work skills,

• problem-solving skills,

• initiative and enterprise skills,

• planning and organising skills,

• self-management skills,

• learning skills, and

• technology skills.

A striking feature of the report is its implicit contention that the employability skills required for

the future can be determined in the absence of any detailed view of the industrial and

occupational structure of the economy. Thus, the report argues that employability skills should

equip the workforce to meet the “challenges facing Australian industry”, especially insofar as

those challenges have to do with “globalisation and the knowledge economy”, but no specific

connections are made. Further, “the required technical and job-specific skills, being subject to

ongoing change, are not readily predictable. What is important, therefore, is the capacity to

continually adapt and upgrade with the application of core or generic employability skills that

can be transferred across different settings."

The idea seems to be that workers should be provided with a set of generic skills which will

enable them to move from one job to another as the need arises. But suppose that “technical and

job-specific skills” can be associated with the occupations of the Australian Standard

Classification of Occupations (ASCO). While all eight skills in the Employability Skills

Framework may be required to some extent in all ASCO occupations, some skills will be more

relevant for some occupations than others. For example, Technology skills would seem to be

very important for 1224 Information technology managers, but not particularly important for

9932 Fast food cooks. Hence, if it is thought that more Information technology managers than

Fast food cooks will be needed to meet the future challenges facing Australian industry, the

training resources devoted to generic skills should be allocated appropriately. Generic

employability skills may well be more transferable than job-specific skills, but not (or, at least,

not obviously) to the extent that structural change can safely be ignored when determining future

requirements.

Furthermore, while it is certainly true that future employment by occupation is “not readily

predictable”, it is not true that it cannot be predicted at all. Many industrialised countries

routinely produce forecasts of employment by industry and occupation1. In Australia, the Centre

of Policy Studies (CoPS) at Monash University has done so for more than ten years.

The position adopted in Employability Skills for the Future and similar reports would appear to

owe more to necessity than to virtue. It seems likely that the link between the structure of the

1 See Neugart and Schomann (2002) for a recent survey.

2

economy and the demand for particular generic skills is typically ignored, not because the link is

thought to be unimportant, but because there has been no suitable analytical tool for its

elaboration. It is this deficiency which the current paper seeks to redress.

In 1998, the U.S. Department of Labor introduced the Occupational Information Network.

Commonly referred to as the O*NET, it is a comprehensive database linking worker attributes

(or employability skills, in terms of the above discussion) and job characteristics (or job-specific

skills). Moreover, the links are specified in both qualitative and quantitative terms. Esposto

(2005) has adapted the O*NET to the Australian labour market and created a means whereby the

Monash occupational forecasts can be extended to employability skills. The paper presents the

first such forecasts and details the methodology involved.

The rest of the paper is organised as follows. Section 2 provides a review of the Australian

literature on employability skills. Section 3 describes the O*NET and its adaptation to

Australian data. Section 4 describes the Monash Forecasting System and its foreshadowed

extension. Section 5 presents forecasts of employment for various kinds of employability skills.

Section 6 contains concluding remarks.

3

2. Employability Skills: A Review

The motivation for developing, defining and identifying employability skills has a historical

background that goes back to the beginning of the 1980s brought about by a desire by various

industry groups, policy makers and educators to remain competitive and to gain ground in local

and global markets (Allen Consulting, 2004; ACER, 2001; and DEST, 2002). The need to

compete internationally has forced industry and enterprises to adapt and change, while the skills,

knowledge and abilities required by individuals to remain competitive in the labour market have

had to be realigned in order to meet a new global competitive reality. Thus, the challenge for

industry, educational bodies and policy makers in Australia has been to identify the skill sets that

make individuals employable in a rapidly changing economy and labour market. To date, the

most important efforts to try to establish the skill sets that are required to enter, or access and

remain in the world of work have been the Karmel Report (1985), the Finn Report (1991), the

Mayer Report (1992) and the Employability Skills for the Future Report (2002).

The Karmel Report in 1985 stressed the requirement of the secondary school sector to support

the attainment by graduates of educational standards that would lead to long term employability.

The Finn Review Committee of 1991 was required to report on “… national curriculum

principles designed to … develop key competencies (Australian Education Council, Finn

Review, p. 2). The report stressed among other policy matters the inclusion of curriculum

principles that supported the development of key competencies (DEST, 2002, p. 21) and

recommended six key areas of competence, including Language and Communication,

Mathematics, Scientific and Technological Understanding, Cultural Understanding, Problem

Solving and Personal and Interpersonal Skills (ACER, 2001, p. 13). In 1992, the Mayer

Committee “…used its own expertise, consulted with industry and with educators in the school

and Vocational, Educational and Training (VET) sectors, and to a lesser extent with the higher

education sector, and finally undertook a validation exercise which involved further

consultations with industry” (ACER, p. 13). The committee recommended a set of

competencies, which are detailed in Table 1. The Committee created three levels of performance

for each competency “… which differentiated the levels of competency necessary to undertake

the activity, manage the activity, or evaluate or revise an activity undertaken” (DEST, 2002, p.

22).

4

In 2002, Department of Education, Science and Training (DEST) commissioned the

‘Employability Skills for the Future Report’, which provided advice on the following 5 key

areas:

• possible new requirements for generic employability competencies that industry requires,

or will require, in the foreseeable future, since the Mayer key Competencies were

developed;

• clear definitions of what Australian and leading business enterprises mean by

‘employability’ skills and the consistency or otherwise between the various terms

similarly studied;

• a proposed suite of employability skills, including outlines of assessment, certification

and reporting of performance options that suit both industry and education;

• industry (small, medium and large business) reactions to the proposed suite and reporting

options;

• a report on the case studies involving 13 large enterprises; and

• a report on focus group research with small and medium-sized enterprises. (DEST, 2002,

p. 2).

The study was based upon a literature review by Australian Council for Educational Research

(ACER) (2001), case study research with large firms, and focus groups/interviews with small

and medium firms. As mentioned above, the report developed an Employability Skills

Framework consisting of personal attributes, skills and elements. The Report includes three key

terms that are described in Table 2 below and which reflect the views of employers operating in

small, medium and large sized enterprises. These have been developed in order to assist in both

the enterprise’s views and the framework.

The framework of the Employability Skills for the Future is underpinned by a number of critical

factors. Firstly, it is closely linked and builds on the Mayer Key competencies. Secondly,

employers (regardless of enterprise size) recognise the link between ‘Employability Skills’ with

the Mayer Competencies, and, thirdly, small, medium and large firms identify the same critical

mix of skills as being necessary for employability and continued employment for employees.

5

Table 1. List of Mayer Key Competencies

Key Competencies Descriptors

1. Collecting, analysing and

organising information

The capacity to locate information, sift and sort the

information in order to select what is required and

present in a useful way, evaluate both the information

itself and the sources and methods used to obtain it.

2 Communicating ideas and

information

The capacity to communicate effectively with others

using a whole range of spoken, written, graphic and other

non-verbal means of expression.

3. Planning and organising

activities

The capacity to plan and organise one’s own work

activities, including making good use of time and

resources, sorting out priorities and monitoring

performance.

4. Working with others and

in teams

The capacity to interact effectively with other people

both on a one-to-one basis and in groups, including

understanding and responding to the needs of others and

working effectively as a member of a team to achieve a

shared goal.

5. Using mathematical ideas

and techniques

The capacity to use mathematical idea, such as number

and space, and techniques, such as estimation and

approximation, for practical purposes.

6. Solving problems The capacity to apply problem-solving strategies in

purposeful ways, both in situations where the problem

and the desired solution are clearly and evident, and in

situations requiring critical thinking and a creative

approach to achieve an outcome.

7. Using technology The capacity to apply technology, combining the

physical and sensory skills needed to operate equipment

with the understanding of scientific and technological

principles needed to explore and adapt systems.

Source: (Australian Education Council. Mayer Committee, 1992, pp. 8-9).

6

Table 2 Terminology used in explaining the Employability Skills Framework

Term Explanation

Personal attributes Term used to describe a set of non skill-based behaviours

and attributes that employers felt were as important as

the employability skills and other technical job-specific

skills.

Skills Term used to describe the learned capacity of the

individual. Skills has been used instead of competencies,

reflecting the language of enterprises interviewed and to

avoid any definitional confusion with the different ways

competencies is used.

Elements The elements are the facets of skill that employers

identified as important.

The mix and priority of these elements would vary from

job to job.

The list of elements is not exhaustive but rather reflects

the information provided by the interviewees in this

study.

The list of elements is indicative of the expectations of

employers.

The level of sophistication in the application of the

element will depend on the job level and requirements.

Source: DEST, 2002, p. 36.

7

3. The O*NET

1.1 A Suggested Approach to Defining and Measuring Skill

One way of achieving a good understanding of the skill needs of a changing economy and labour

market is through the Occupational Information Network (O*NET). O*NET is an extensive and

comprehensive database that describes the attributes and characteristics of occupations and

workers. Its first version was launched in 1998 and is known as O*NET 98. The O*NET was

designed to replace the DOT and is considered to be the most comprehensive standard source of

occupational information in the US. An advantage of O*NET is that it offers statistical

information that can be applied to the Australian context to analyse labour market change.2 The

O*NET was developed by a consortium led by the Department of Labor in the US. It was built

to replace the Dictionary of Occupational Titles (DOT), which was conceived in the 1930s and

was last published in 1991. The DOT was developed in an industrial economy during a time that

emphasised blue-collar occupations. It was updated periodically and proved very useful for

more than six decades. Its usefulness, however, faded as economic changes shifted towards

information and services away from heavy industries and other blue-collar work. Because of

these changes, the need for more up to date occupational information that reflects modern jobs

spurred the creation of the O*NET. The O*NET, which was developed on the foundations of the

DOT, was a response to a growing demand for an expanding public employment service and was

intended to assist, among other things, in job placement activities. It follows a different strategy

from DOT, providing very detailed information on about 1,120 occupations rather than more

limited information on over 12,000 occupations. The O*NET is continually updated to reflect

the dynamic and ever-changing nature of employment. For example, the latest version is the

O*NET 8.3

2 For example, Pappas (2001) used the DOT to analyse the causes of increasing earnings inequality in Australia,

while Sheehan and Esposto (2001) used the O*NET to study the characteristics of Australian jobs. The information

that follows is mainly descriptive and draws heavily on the work conducted by Peterson et al. (1995, 1997, 1999,

2001) who were responsible for developing the O*NET. Information is also drawn from the O*NET 98 Viewer (US

Department of Labor, 1998) and the O*NET 98 training materials developed and distributed by the National

Occupational Information Coordinating Committee (NOICC).

3 The O*NET data used in this analysis is from the O*NET 98 version. The reason for this is that newer versions of

O*NET have not upgraded all the data on the variables used here for analysis, which are contained in the first

version of the O*NET 98.

8

The occupational information in the O*NET is organised in a relational database which

identifies, defines and describes the comprehensive elements of job performance. It contains

hundreds of information items on job requirements, worker attributes and the content and context

of work, capturing what people do in their day-to-day activities.

The origins of the O*NET go back to the 1993 report to the US Secretary of Labor’s Advisory

Panel on the DOT (APDOT). The report noted that the framework underlying the then current

version of the DOT was not adequate for analysis and description of occupations. This was

because the descriptions of occupations in the DOT were predominantly based on industries of

an earlier era. These descriptors were deficient in dealing with the large shift towards services

and the emerging information industries in the US economy. Furthermore, it was highlighted

that the DOT did not meet the demands of the emerging labour force of the 21st century

(Mumford and Peterson, 1999, p. 22).

The DOT’s shortcomings indicated that it could not be used for rapid assessment of skills,

abilities and other characteristics required by a job family or for showing how skill levels are

changing or may be related, for example, across occupations.

The O*NET was developed with the aim of becoming an information system consisting of a

framework made up of a variety of components. Firstly, it possesses occupational information

that allows jobs to be described in terms of more general descriptors that reflect the modern

labour market. Secondly, the O*NET is closely linked to labour market data which are updated

on a continual basis. Although it was not originally intended for this purpose, an advantage of

the O*NET is that the detailed data collected on worker characteristics and job requirements can

be used to examine and analyse changes in the labour market. Data are gathered on over 200

occupations each year, with the aim of totally upgrading the database every five years (O*NET

Consortium, 2004). As distinct from the DOT, the O*NET contains cross-occupation descriptive

information that includes the kind of work, conditions under which it is done, and the

requirements imposed on the people doing the work. Finally, the O*NET takes into account the

variety of applications of the information that is collected. All the occupations in the database

are related to a common framework that describes job requirements and worker attributes, as

well as the content and context of work using nearly 300 descriptors. All the information

collected in the database is organised into a framework known as the Content Model.

9

Table 3. Similarities between O*NET and employability skills

Mayer Key

Competencies

Employability

skills

framework

O*NET Skills O*NET

Knowledge

O*NET Worker

Activities

Collecting,

analysing and

organising

information

Personal

Attributes

Content skills Business and

Management

Looking for

and receiving

job-related

information

Communicating

ideas and

information

Planning and

organisation

Process skills Manufacturing

and Production

Identifying

/evaluating job-

relevant

information

Planning and

organising

activities

Communication Social skills Engineering

and Technology

Information

/data

Processing

Working with

others and in

teams

Planning and

organising

Complex

problem

solving skills

Mathematics

and Science

Reasoning

/decision

making

Using

mathematical

ideas and

techniques

Teamwork Technical skills Health Services Performing

physical and

manual work

activities

Solving

problems

Problem

solving

Systems skills Education and

Training

Performing

complex

/technical

activities

Using

technology

Technology Resource

Management

skills

Arts and

Humanities

Communicating

/interacting

Learning Law and Public

Safety

Coordinating

/developing

/managing

/advising others

Self-

management

Transportation Administering

Source: Author’s compilation.

10

3.3 The O*NET Content Model

The Content Model is the conceptual foundation of the O*NET. It was developed by Mumford

and Peterson (1995b), using research on job and organisational analysis, and embodies a

framework that reflects the character of occupations (i.e. using job-oriented descriptors) and

people (i.e. using worker-oriented descriptors). The Content Model also allows occupational

information to be applied across jobs, industry sectors (by using cross-occupational descriptors)

and within occupations (using occupation-specific descriptors). The Content Model classifies

data into six domains that provide detailed information related to the attributes of occupations

and to the characteristics required of people who actually do the job. It includes the specific

domains and elements in the O*NET database that might be used to describe jobs. These

components are based on psychological and job analysis research carried out by the O*NET

consortium.

Figure 1 summarises each of the six domains and their corresponding components. The six

domains are Worker Characteristics, Worker Requirements, Experience Requirements,

Occupation Requirements, Occupation Characteristics and Occupation Specific Information.

The organisation of the Content Model allows the user to concentrate on relevant information

that details the attributes and characteristics of jobs and workers.

3.4 O*NET Variables Used in Forecast Analysis

As noted, distinctive feature of the O*NET’s Content Model is that it provides a detailed

description of occupations in terms of cross-job descriptors. Furthermore, the O*NET contains

data that can assist in measuring and comparing the content of occupations and how these have

changed over time.

Esposto (2005) showed that many of the proxies of skill used in econometric work are not very

effective and leave much to be desired. For example, numerous studies have used years of

education or experience as a proxy for skill, but these two concepts are not necessarily the same

thing. A number of studies of manufacturing have used the ratio of production to non-production

employees as a measure of skill, but replacing skilled tradesmen with salesmen or general office

11

Figure 1. Six domains of the O*NET Content Model

O*NET

Experience Requirements Training

Experience

Licensure

Occupation Specific Requirements

Occupational Knowledges

Occupational Skills

Tasks

Duties

Machines, Tools and Equipment

Occupational Requirements

Generalised Work Activities

Work Context

Organisational Context

Worker Requirements

Basic Skills

Cross-Functional Skills

Knowledges

Education

Worker Characteristics

Abilities

Occupation Values and

Interests

Work Style

Occupation Characteristics

Labor Market Information

Occupational Outlook

Wages

Source: Mumford and Peterson (1999, p. 25, Figure 3.2).

staff may reduce rather than increase the overall level of skill in a firm. Other studies have used

earnings as a proxy for skill. Although this has some advantages, it does little to throw light on

the role of skill changes in the pattern of earnings, or to increase our understanding of the

changing nature of skill. In summary, economies and labour markets in the industrialised world

have experienced dramatic changes in the last three decades. As a result, the skill composition of

many occupations may also be changing rapidly, and the measures and proxies for skill used to

date may not provide us with enough insights to comprehend these changes.

12

3.5 O*NET Data Collection Method4

Data on the content model’s descriptors were collected from a variety of sources. The most

important source of information was questionnaires collected from job incumbents. A variety of

rating scales was used in order to describe the attribute of the occupation under examination. In

trying to ensure that the data was valid, a wide variety of ratings were utilised, such as level,

importance and frequency. Where possible, moreover, behavioural based rating scales were

employed (Hubbard, et al., 2000).

The data were obtained from a variety of sources. One such source employed incumbent-based

data collection methods in order to minimise the costs of using the analyst based approach.

Second, the O*NET measures were developed in order to extract as much information as

possible from incumbents. This allowed incumbents to provide reliable and accurate

information. The body of knowledge contained in the field of industrial psychology was also

used in order to assess a wide range of job, occupational and organisational characteristics based

on highly reliable questionnaires and other job related information data gathering tools (Peterson,

et. al. 2001).

Information for the model domains was collected by a series of questionnaires completed by job

incumbents. The respondents to O*NET data collections were guided through a series of

questionnaires designed to solicit ratings of jobs or workers represented by the following Content

Model elements, including, worker characteristics, worker requirements, occupational

requirements, occupational specific requirements, and occupational characteristics. For the data

used in this analysis, the questionnaires used several types of rating scales for measuring

frequency, level and importance.

3.6 O*NET Descriptors Used for this Analysis

This section describes in more detail the three O*NET indicators used to investigate changes in

the content of Australian occupations. The three indicators of skill, knowledge and GWAs form

4 For a detailed discussion on data collection methods for the O*NET see Peterson et.al., 1999, Chapters 3 and 4.

This section draws heavily on their work.

13

part of the domains of worker characteristics, worker requirements and occupational

requirements described in the previous section.

3.6.1 Skills

The approach taken by O*NET to define skill is that of Mumford and Peterson (1995a). They

define skill as a set of general procedures that underlie the effective acquisition and application

of knowledge in different areas of endeavour (chap. 3, p. 4). The implication of this definition is

threefold. Firstly, skills are innately linked to knowledge, learning, practice, education and

experience. For example, a person cannot acquire or apply skills without learning, practising,

being exposed to education, experiencing or by acquiring knowledge. Secondly, skills can be

seen as general procedures that are necessary for the performance of multiple tasks. These tasks,

however, must form part of a given domain of skills such as social skills, basic skills or problem

solving skills. Finally, skills are not constant attributes of individuals that remain unchanged

over time. These attributes can be acquired (sometimes they can be lost) and developed as a

result of new learning, experience or newly acquired knowledge.

Given the above, Mumford and Peterson (1995a) argue that skills are not one-dimensional and

require a variety of taxonomies. They divide the taxonomy of skill into two broad categories.

The first is referred to as basic skills. These are defined as the developed capacities that facilitate

learning or the attainment of new knowledge. Basic skills are subdivided into two further

categories described as content and process skills. These are made up of six and four skill

variables respectively, out of a total of 46 skills that comprise the complete O*NET skill

taxonomy. Content skills can be broadly defined in terms of those capabilities that allow people

to acquire information and convey it to others. They represent the structures required to work

with and acquire other skills. This category includes skills such as reading, writing, listening,

speaking, mathematics and science. These skills are also widely seen as fundamental in the

provision of any sound educational system.

Process skills, on the other hand, are seen as those skills that facilitate the acquisition of content

across domains. The ability to think critically is thus part and parcel of process skills. This skill

is closely related to a second kind of general learning skill, referred to as active learning.

Another process-oriented skill takes the form of learning strategy. This uses a variety of

14

approaches when learning new things. Finally, monitoring represents an ongoing appraisal of

the success of an individual’s efforts because it assists them in assessing how well they are

learning something or doing a particular task. The second classification of skills is defined in the

O*NET as the capacities that facilitate individuals to perform effectively in a variety of job

settings. This skill definition is also known as cross-functional skills and in the O*NET Content

Model is based on the notion of socio-technical systems theory.

3.6.2 Knowledges

The O*NET defines knowledge as a collection of discrete but related facts, information and

principles about a particular area of work. Knowledge is acquired through formal education

and/or training or can be built upon a collection of a variety of experiences. Knowledge is

acquired incrementally and can be used to acquire further knowledge and to develop skills. It is

closely related to skills because it can assist in the development of skills, while, at the same time,

skills can aid the process of knowledge acquisition.

In the O*NET taxonomy, the 33 knowledges can be regarded as belonging to general categories.

Some of these categories are general and are regarded as being essential elements in the

successful performance of occupational tasks. Others are narrower and can only be applied to a

fine range of occupational groups, while others can be seen as being occupation specific.

In identifying a taxonomy of knowledge indicators, Fleishman et al. (1995) developed a

knowledge classification by analysing job descriptors and looking for tasks and behaviours that

were representative of underlying knowledges in occupations. These underlying knowledges had

to be general enough so that they could be transferable across different occupations. Lists of

knowledge descriptors were created and these were further divided into categories or clusters of

knowledges. Fleishman et al. (1995) arrived at a parsimonious set of 33 knowledges.

3.6.3 Generalised work activities (GWAs)

GWAs in the O*NET provide a taxonomic structure that can be used to describe the work that

people do in their particular occupations. They describe the work activities that are needed or

required for successful job performance. The distinction between GWA, knowledge and skills

15

lies in the fact that our work activity measures are closely linked to the requirement of

occupations, whereas knowledge and skill are measures that are closely related to requirements

of individuals who perform a given job.

GWAs are useful in the O*NET system because they are closely linked to the knowledge and

skill of individuals and can provide useful strategies when matching jobs to people.

Furthermore, the data that describe and measure each of the 42 GWAs can be used to analyse

changes in the labour market in terms of occupation specific requirements. This information can

also be used to analyse changes in the composition of employment in terms of occupational

requirements and can be compared to changes in the skill and knowledge composition of

Australian occupations.

In the O*NET a GWA is defined as the aggregation of similar job attributes that brings about or

promotes the accomplishment of major work functions and tasks.

3.6.4 An illustrative example of O*NET occupational data

The three O*NET measures described above contain a total of 121 variables which are assigned

to a total of 340 Australian occupations in ASCO 2nd

edition. Conceptualisation of how the data

are organised in the O*NET and what they represent can be confusing without further

explanation. To demonstrate how the O*NET data are organised for each of our measures, we

provide an example using GWA information contained in O*NET5

The GWA taxonomic structure of the O*NET is hierarchically organised and is very detailed.

Each descriptor defines a GWA, such as getting information to do the job, identifying objects,

actions and events, judging quality of things, analysing data or information and thinking

creatively. For a given occupation, each of these areas of GWAs is ranked on a scale of 0 to 100,

for both its importance to the occupation and for the level in that area required in that occupation.

With a rank of zero in the importance scale, the descriptor of GWA shows that the work activity

is of no consequence to that occupation, while at a ranking of 100, the descriptor is considered to

be extremely important in performing that particular job. Thus the ‘importance’ indicator for a

5 In the O*NET, information on skill and knowledge scores is organised in the same way as for GWAs.

16

particular GWA refers to how important it is to the performance of the job in question.

Similarly, the ‘level’ indicator for a particular area of GWA refers to the degree or quality which

that work activity requires in the occupation under investigation.

Thus, in the example provided in Table 1, for the occupation ‘generalist medical practitioner’ the

importance of the GWA descriptor ‘monitor processes, materials, surroundings’ is ranked 100,

and the level of ‘monitor processes, materials, surroundings’ required is high, at 89. For

‘librarian’ the same GWA is not as important (score 42 in importance) and the quality required is

even less demanding (score 29 in level). Their most important GWA relates to ‘communicating

with persons outside the organisation’ (importance score 92, whilst its level has a score of 62).

3.6.5 Link between O*NET and Australian Occupations

To analyse changes in the GWA knowledge, and skill intensity of Australian occupations,

Australian employment data were linked to the O*NET occupational data. This was done

because, unfortunately, concordances between ASCO and the O*NET occupational codes do not

exist. But even though in the US concordances are available between the O*NET and

occupational employment data series, the number of occupations covered in the O*NET is

greater than in the employment series. For example, by comparison with about 1,120 O*NET

codes, the most detailed occupational employment data in the US are the Department of Labor’s

Occupational Structure data which cover 820 occupations. Thus, the ‘most appropriate’ O*NET

code was assigned to a particular occupation in the employment data. More specifically, each

Australian occupation was assigned an O*NET occupational code on an individual basis. In

assigning the ‘most appropriate’ O*NET occupation code to an ASCO 2nd

edition occupation

code, it was assumed that there was a similarity in the description of US and Australian

occupations, as the great bulk tend to have similar or close definitions and descriptions.6

6 This assumption has also been made by Pappas (2001) using the DOT and by Sheehan and Esposto (2001) using

the O*NET.

17

Table 4. Generalised Work Activity, Importance and Level

Generalised work activities / Occupation Generalist Medical

Practitioner

Librarian Bricklayer

O*NET 32102A O*NET 31502A O*NET 87302

ASCO 2311 ASCO 2292 ASCO 4414

Importance Level Importance Level Importance Level

Monitor processes, materials, surroundings 100 89 42 29 67 26

Identifying objects, actions and events 95 83 58 43 63 29

Analysing data and information 95 89 63 43 54 26

Making decisions and solving problems 95 83 54 38 38 21

Getting information to do the job 95 91 79 60 71 31

Assisting and caring for others 95 91 67 38 8 12

Implementing ideas, programs, etc. 85 74 67 38 67 29

Documenting/recording information 85 57 79 48 4 10

Communicating with persons outside

organisation.

80 71 92 62 8 14

Source: US Department of Labor (1998a) and author’s calculations.

18

The assignment of an O*NET occupation to an ASCO 2nd

edition occupation was done by

matching the occupation title from the O*NET to the ASCO 2nd

edition, using the list of

definitions and titles of occupations as described in both the ASCO (ABS cat. no. 1220) and the

O*NET Dictionary of Occupational Titles (1998 edition). The assignment of occupations

involved searching for an occupation title in the ASCO 2nd

edition and then finding its

corresponding title in the O*NET database. This was done for a total of 340 ASCO 2nd

edition

occupations.

3.6.6 Applications of O*NET in Labour Market Research

As noted, the O*NET is an occupational database which offers an important resource that can be

applied in the study of labour market change. Given its recent and continuing development a

number of studies have used the O*NET to study and analyse the labour market, to identify

occupational skill requirements and to provide forecast of short-term demand for skills.

One of the first studies to use the O*NET database was conducted by the Minnesota Department

of Economic Security (MDES). The report titled “Minnesota’s Most Marketable Skills”, (1999)

identified the occupational skill requirements considered to be most marketable. Marketable

skills were defined as those occupational requirements that are meant to be associated with high

wages and/or employment growth. The report found that out of 57 occupational skill

requirements that measure the knowledge, ability and skill dimensions of an occupation, 18 were

found to be extremely ‘marketable’. The finding of this study were used by the State of

Minnesota to apply to areas of policy related both to the labour market and to align the skill

requirement of occupations to education curricula.

In 2002, the MDES conducted a forecast of the demand for skills in the short term, using 46

skills, 52 abilities, 33 knowledge and 38 generalised work activity measures found in the O*NET

database. The report summarised research into the feasibility and appropriate methodology for

developing short-term, skill-based forecasts in order to direct public resources more effectively.

The report concluded that the O*NET was an effective data system that could be applied for the

analysis of forecasts of short-term demand for skills.

Esposto (2005) applied the O*NET database to study labour market change in Australia.

Drawing on earlier work conducted by Sheehan and Esposto (2001), Esposto (2005) investigated

19

labour market change using the O*NET by addressing the following issues. Firstly, the study

focused on how skill should best be measured in addressing the issue of skill-bias in the demand

for labour; secondly, he showed that the Australian labour market had experienced a long-term

process of skill-bias in the demand for labour; and thirdly having confirmed skill-bias, the study

showed using O*NET measures of skill and knowledge evidence that this increasing relative

demand for higher skill labour is an important explanatory factor in the rise in earnings

inequality in Australia.

Another application of the O*NET measures was used in a study designed to identify

occupations that are comparable in earnings to primary and post-primary teachers in the US.

Using the generalised work activities, basic and cross-functional skills and educational level

scales embodied in the O*NET, Milanowski (2003) performed a variety of cluster analysis

methods, and was able to identify comparable work activity and skills for US occupations.

Rotundo and Sackett (2004) using the O*NET database conducted a job-level evaluation of

whether specific skills or abilities could be identified that were most strongly linked to wage or

whether broad skill/ability factors accounted for a majority of wage variance. The authors found

that a majority of the wage variance explainable by skills/abilities could be attributed to a general

cognition factor.

4. The Monash Forecasting System

The demand for labour depends on many factors. It depends on the state of macroeconomic

health of the domestic economy and of the economies of trading partners. It depends on the

amount of capital investment and on its allocation between industries. It depends on the rate of

technical change and on changes in government policy. Moreover, all these factors are

interconnected. Developments in one industry affect the demand for labour in other industries.

The Monash Forecasting System (MFS) incorporates all these factors in a set of formal

economy-wide forecasts for labour demand.

The elements of the MFS are set out in Figure 2. As a formally specified system, its role is to

supply a framework for incorporating relevant data into the forecasting framework. Published

20

data accessed by the system include the national accounts, input/output tables, State accounts,

population censuses, foreign trade statistics, capital stock statistics, and income and expenditure

surveys. Additional unpublished material is prepared by the Australian Bureau of Statistics

especially for the system. Moreover the MFS requires all its data to be consistent. If any

inconsistencies do exist in the primary sources, they must be reconciled before the data can be

included. This consistency requirement makes the system especially powerful as a framework

for organising data.

As well as data about the past, formal or model-based forecasts usually rest upon informed

opinion about future changes in variables that are exogenous to (i.e., determined outside) the

system. The MFS is quite adaptable in this regard. It incorporates the views of numerous expert

bodies and can accommodate more detailed exogenous forecasts as they become available. The

sources of exogenous forecasts identified in Figure 2 are:

the private forecasting agency, Access Economics (which contributes information about the

future state of the macro economy),

The Australian Bureau of Agricultural and Resource Economics (export prices and volumes

for primary products),

The Tourism Forecasting Council (prospects for tourism),

The Productivity Commission (changes in protection implied by government industry

policy), and

The Centre of Policy Studies (changes in technology and consumer tastes).

The system can also produce alternative forecasts corresponding to competing views about the

future. Just as for historical data, all opinions formally incorporated in a particular forecast must

21

Macro

Forecasts

(ACCESS)

Industry

Policy (PC)

Exports

(ABARE &

TFC)

MONASH

simulations

2004-05

to

2012-13

Structure of

Technical & Taste

Changes (CoPS)

Output & Employment Forecasts for

158 Industries

Occupational Share

Effects (CoPS)Labour Market Extension I

Employment Forecasts for

340 Occupations

Qualification Share

Effects (CoPS)Labour Market Extension II

Employment Forecasts for

8 Qualification Levels, 72

Qualification Felds

Figure 2. The Monash Forecasting System

22

be consistent with each other. A forecaster using the MFS must either seek a consensus between

the expert bodies involved in forecasting the exogenous variables or impose his/her own

judgement to resolve any outstanding differences before the forecast can proceed. In other

words, the MFS provides a framework for coordinating both historical data and expert opinion

about the future that bear on the future demand for labour.

An MFS forecast of the demand for labour proceeds in five stages. It begins with a

macroeconomic scenario derived from the Business Outlook published quarterly by Access

Economics7. Table 5 compares the current scenario with recent history for a selection of macro

variables. During the eight years of the forecast period (2004-05 to 2012-13), Access expects

annual GDP growth to be 0.37 percentage points lower than the 3.61 per cent achieved on

average during the preceeding eight years. Furthermore, the contribution of exports to GDP

growth is expected to increase at the expense of private consumption and investment. The

balance of trade deficit is forecast to decline.

The second stage in the process is to convert the forecasts for GDP and its components into

forecasts of output and employment8 by industry. The structural forecasts supplied by the expert

bodies indicated in Figure 2 are incorporated at this stage. In particular, the array of exogenous

information is treated as a set of constraints which governs a simulation using the MONASH

applied general equilibrium model in forecast mode. Results of the simulation for 17 industries

are shown in Table 6. Among the industries with good output growth prospects, Mining,

Wholesale Trade and Education are strong participants in the forecast rapid growth in exports.

Wholesale Trade also benefits because it provides margins on the exports of other industries.

Finance and insurance and Property and business services, on the other hand, are favoured by

changes in technology which result in their outputs being used more intensively by other

industries. The poor growth prospects for the Construction industry reflect a slump in

residential construction forecast for 2009-10 by Access Economics. Personal Services suffers

because it is very labour intensive and has little scope for improvements in labour productivity.

Hence, as real wage rates are forecast to increase at about 0.4 per cent per annum, the output of

the industry tends to become relatively expensive and households switch their expenditure in

7 The forecasts presented in the paper are based on the September 2005 edition. 8 In what follows, we implicitly assume that employment is demand determined and refer to employment when,

strictly speaking, we mean the demand for labour.

23

Table 5. Average Annual Growth Rates, Selected Macro Variables

Australia, Per Cent Per Annum

Historical Forecast

Variable Data

1996-97 2004-05

to to

2004-05 2012-13

1 Private consumption 4.14 2.86

2 Private residential construction 5.78 1.30

3 Private non-residential construction 6.20 2.11

4 Total private investment 6.87 2.59

5 Government investment 4.91 2.97

6 Total investment 6.68 2.62

7 Government consumption 3.19 2.45

8 Exports 3.06 7.61

9 Imports 8.05 4.82

10 Gross domestic product 3.61 3.24

favour of other industries. Government expenditure, on the other hand, is expected to shift in

favour of Public Administration and Defence during the forecast period. Hence the prospects

for Health and community services and Cultural and recreational services are commensurately

reduced.

The output and employment forecasts are related by production functions which determine the

increase in output associated with given increases in inputs (capital and labour) and a given rate

of primary factor saving technical change. The influence of capital growth and technical change

can produce quite different output and employment forecasts for some industries. The change in

capital inputs depends critically on whether an industry was under- or over-capitalised in the

base period of the forecast (i.e., on whether the rate of return in the industry was above or below

the average across industries). An industry with a relatively high rate of return attracts

24

Table 6. Forecast Changes in Output and Employment, ANZSIC Divisions, Australia, 2004-05 to 2012-13

Output (Constant Prices) Employment (Persons)

Growth Rate Growth Share Growth Rate Growth Share

(per cent per annum) (per cent) (per cent per annum) (per cent)

A00 Agriculture forestry and fishing 3.13 3.35 0.30 0.86

B00 Mining 5.10 8.44 2.72 2.49

C00 Manufacturing 2.60 9.93 0.78 6.77

D00 Utilities 2.88 2.37 -1.05 -0.60

E00 Construction 1.56 3.38 -0.15 -0.99

F00 Wholesale trade 3.92 7.37 1.70 6.20

G00 Retail trade 2.74 5.50 1.58 19.46

H00 Accommodation cafes restaurants 2.73 2.01 0.83 3.37

I00 Transport and storage 3.52 6.92 1.33 4.96

J00 Communications 3.29 3.42 -1.59 -2.12

K00 Finance and insurance 4.09 12.28 0.72 2.07

L00 Property and business services 4.03 16.12 2.10 19.92

M00 Public administration and defence 2.87 4.13 1.57 5.81

N00 Education 3.64 5.82 2.88 16.86

O00 Health and community services 2.50 5.55 1.15 9.35

P00 Cultural and recreational services 2.40 1.51 0.55 1.15

Q00 Personal and other services 2.33 1.87 1.39 4.43

999 All industries 3.24 100.00 1.25 100.00

25

investment and enjoys a relatively high rate of capital growth. For a given rate of output growth

and technical change, this implies a relatively low rate of employment growth. Similarly an

industry with a relatively rapid rate of technical change will tend to have a relatively low rate of

growth in employment. Thus Communications, which has better than average prospects for

output growth, has poor employment prospects because it is expected to experience particularly

rapid growth in labour productivity.

At the third stage, the national forecasts for output and employment are converted into regional

forecasts using the MONASH regional equation system (MRES), a derivative of the ORANI

regional equation system (Dixon et al., 1982, Chapter 6). The regionalisation process takes

account of:

differences in industrial structures,

region-specific industry effects, such as mine closures,

population movements,

expected expenditures by regional governments, and

local multipliers.

Regional forecasts are produced at two levels of aggregation, namely, eight State and Territories

and 56 Statistical Divisions.

At the fourth stage, the employment forecasts are converted from an industry basis to an

occupational basis. National results for nine occupations are reported in Table 7. As the table

indicates , employment growth (measured in persons) for a particular occupation can be

decomposed into:

• a component due to the growth in aggregate employment (measured in hours),

• a component (the industry share effect) due to changes in the distribution between industries,

• a component (the occupational share effect) due to changes in the distribution of

employment between occupations within industries, and

• a component due to changes in the number of hours per worker.

26

The forecast for aggregate employment in hours is 1.14 per cent per annum. The industry share

effects are computed from the growth rates in employment by industry using an industry-by-

occupation matrix derived from the Population Census and the Labour Force Survey. The

occupational share effects are treated as a type of technical change and are forecast by

extrapolating historical trends in the occupational mix in each industry. The method is described

in detail in Meagher (1997). Changes in the number of hours per worker in an occupation are

also derived by extrapolating past trends.

At the final stage, the forecasts for employment by occupation in persons are used to determine

thew employment outlook for workers identified by age, sex, qualifications and hours worked

per week. The method is analogous to that used to determine the occupational forecasts from

the industry forecasts.

27

Table 7. Contributions to Employment Growth, Australia, ASCO Major Groups, 2004-05 to 2012-13, Per Cent Per Annum

Growth Industry Occupational Total Hours Total

Effect Share Share Growth Effect Growth

(hours) Effect Effect (Hours) (persons)

1 Managers & Administrators 1.14 -0.43 1.46 2.17 0.17 2.34

2 Professionals 1.14 0.57 0.30 2.01 0.14 2.15

3 Associate Professionals 1.14 -0.01 0.61 1.74 0.29 2.03

4 Tradespersons & Related Workers 1.14 -0.53 -0.86 -0.25 0.09 -0.15

5 Advanced Clerical & Service Workers 1.14 0.08 -1.71 -0.49 0.36 -0.12

6 Intermediate Clerical, Sales & Service Workers 1.14 0.04 0.12 1.30 0.09 1.40

7 Intermediate Production & Transport Workers 1.14 -0.01 -0.45 0.68 0.14 0.82

8 Elementary Clerical, Sales & Service Workers 1.14 0.15 -0.85 0.44 0.33 0.77

9 Labourers & Related Workers 1.14 -0.12 -0.92 0.10 0.08 0.19

10 All Occupations 1.14 0.00 0.00 1.14 0.11 1.25

28

5. Forecasting Employability Skills

As adapted by Esposto, the O*NET assigns an “importance” indicator between 0 and 100

(inclusive) to each of 46 skill groups for each of 340 occupations (i.e., the ASCO unit groups).

To extend the MFS forecasts of employment by occupation (measured in hours) to apply to the

O*NET skill groups, we assume that the number of hours worked in an occupation is distributed

between the skill groups in proportion to the importance indicator for that occupation. More

formally, let Aik be the value of the importance indicator for skill type i in occupation k. Then

į1 Ȉi Aik = Ek ,

where Ek is total employment in occupation k and į1 is a scale factor. That is, employment of

skill type i in occupation k is given by į1 Aik when measured in hours.

We further assume that the number of efficiency units delivered per hour by labour of skill type i

in occupation k is proportional to the corresponding O*NET skill “level” indicator. Let Bik be the

value of the level indicator for skill type i in occupation k. Then

į2 Ȉi Bik (į1 Aik ) = Wk ,

where Wk is total employment (measured in efficiency units) for occupation k and į2 is a scale

factor. That is, employment of skill type i in occupation k is given by į1 į2 Aik Bik when

measured in efficiency units. Wk might reasonably be set equal to the wage bill in occupation k

but the size of an efficiency unit is essentially arbitrary.

Similar assumptions are adopted for the 32 O*NET knowledges and the 42 O*NET work

activities.

Clearly, the methodology for forecasting the demand for employability skills (as represented by

the O*NET skills, knowledges and work activities) is somewhat less robust than that for the

other variables of the MFS. In particular, the conversion of occupational forecasts into forecasts

29

for employability skills is not based on specific information (such as that collected in the Labour

Force Survey) about the relationship between the two, but on opinion (embodied in the O*NET

importance and levels indicators) about what that relationship might be. However, the method

is plausible, the opinion involved is expert opinion , and there is no obvious alternative method

by which it can be taken into account. Certainly, it is not possible to convincingly accommodate

matrices of the size required (i.e., 340 by 46 in the case of the O*NET skills) by means of

intuition.

Table 8 presents results for the 46 O*NET skill groups. The forecasts of employment growth in

hours range from 0.57 per cent per annum for 601 Visioning to 1.59 per cent for 703

Management of material resources. This range (1.02 percentage points) is substantially less than

that for the nine major ASCO occupations shown in Table 7 (2.42 percentage points). At the

ASCO unit group level (340 occupations), the range is much larger still, extending from –11.11

per cent per annum for 4943 Footwear tradespersons to +6.66 per cent for 2429 ESL Teachers.

This is consistent with the generic character of the O*NET categories which implies a growth

rate which is an average of the growth rates for the relevant occupations. Nevertheless, the

growth rate forecast for Management of material resources is nearly three times that for

Visioning and substantial variation remains.

The average annual growth rate of 1.59 per cent for Management of material resources translates

into a total growth rate of 13.45 per cent over the eight years of the forecast period. Table 9

shows the contributions made by each of the nine ASCO major groups to total employment

growth. From the first row of the table, the contribution of Managers and Administrators is

given by 0.197 x (18.74 + 1.60) or 4.01 percentage points. A substantial part of the hours spent

performing the skill Management of material resources is contributed by workers belonging to

the occupation Managers and administrators (19.7 per cent in the base period). Over the

forecast period, the employment of Managers and administrators is forecast to increase more

rapidly than average employment (18.74 per cent as compared to 9.52 per cent). Specifically,

the growth of the occupation contributes (0.197 x 18.74) or 3.69 percentage points to the growth

of the skill group (the shift effect). Moreover, Management of material resources is forecast to

increase its share of employment within the occupation Managers and administrators at the

expense of other skill groups, contributing an additional (0.197 x 1.60) or 0.32 percentage points

(the share effect). Because the O*NET “importance” indicators are assumed to remain constant

30

Table 8. Average Annual Growth Rates, O*NET Skill Groups

Australia, Hours, Per Cent Per Annum

Historical Forecast

Skill Group Data

1996-97 2004-05

to to

2004-05 2012-13

100 Content Skills 1.56 1.11

101 Reading and comprehension 1.50 1.08

102 Active listening 1.61 1.11

103 Writing 1.58 1.16

104 Speaking 1.68 1.20

105 Mathematics 1.54 1.07

106 Science 1.38 0.98

200 Process Skills 1.65 1.19

201 Critical thinking 1.71 1.24

202 Active learning 1.70 1.23

203 Learning strategies 1.63 1.21

204 Monitoring 1.57 1.11

300 Social Skills 1.68 1.20

301 Social perceptiveness 1.68 1.20

302 Coordination 1.69 1.14

303 Persuasion 1.76 1.29

304 Negotiation 1.78 1.27

305 Instructing 1.62 1.23

306 Service orientation 1.61 1.12

400 Complex Problem Solving Skills 1.61 1.15

401 Problem identification 1.59 1.14

402 Information gathering 1.58 1.14

403 Information organisation 1.46 1.06

404 Synthesis/Reorganisation 1.63 1.17

405 Idea generation 1.70 1.21

406 Idea evaluation 1.66 1.18

407 Implementation planning 1.67 1.20

408 Solution appraisal 1.62 1.17

…continued

31

Table 8 (continued). Average Annual Growth Rates, O*NET Skill Groups

Australia, Hours, Per Cent Per Annum

Historical Forecast

Skill Group Data

1996-97 2004-05

to to

2004-05 2012-13

500 Technical Skills 1.37 0.91

501 Operations analysis 1.62 1.14

502 Technology design 1.46 0.97

503 Equipment selection 1.38 0.88

504 Installation 1.37 0.78

505 Programming 1.57 1.12

506 Testing 1.33 0.93

507 Operation monitoring 1.34 0.96

508 Operation and control 1.28 0.87

509 Product inspection 1.25 0.88

510 Equipment maintenance 1.38 0.97

511 Troubleshooting 1.11 0.62

512 Repairing 1.27 0.81

600 Systems Skills 1.87 1.29

601 Visioning 1.10 0.57

602 Systems perception 1.95 1.38

603 Identifying downstream consequences 2.04 1.46

604 Identification of key causes 2.04 1.45

605 Judgement and decision making 1.93 1.34

606 Systems evaluation 1.93 1.30

700 Resource Management Skills 2.11 1.43

701 Time management 2.20 1.53

702 Management of financial resources 1.99 1.35

703 Management of material resources 2.34 1.59

704 Management of personnel resources 1.99 1.30

9999 All Skill Groups 1.63 1.14

32

over time, the share effects for the 340 ASCO unit groups are zero. The share effects identified

here reflect a change in the mix unit groups within each major group.

When the occupational contributions are compared for the skill groups Management of material

resources and Visioning, it is clear that the former owes its relatively good employment

prospects to its concentration in the managerial and professional occupations, and the latter owes

its relatively poor prospects to its concentration in the trades and labouring occupations.

Tables 10 and 11 present the forecasts for the O*NET knowledge groups and work activities.

They can be understood in the same way as Table 8. The final table, Table 12, compares the

forecasts in hours and efficiency units.

33

Table 9. Contributions to Employment Growth, Australia, Selected O*NET Skill Groups, 2004-05 to 2012-13

Shift Effect (Per Cent) Share Effect (Per Cent) Contribution

Skill Group / Occupation Employment

hare

(Percentage

nts) S Poi

2004-05 Average Total Average Total

Annual Annual

703 Management of Material Resources

1 Managers & Administrators 0.197 2.17 18.74 0.20 1.60 4.01

2 Professionals 0.214 2.01 17.23 0.26 2.10 4.14

3 Associate Professionals 0.232 1.75 14.88 0.28 2.25 3.97

4 Tradespersons & Related Workers 0.068 -0.24 -1.90 -0.25 -1.99 -0.26

5 Advanced Clerical & Service Workers 0.048 -0.48 -3.80 0.67 5.52 0.08

6 Intermediate Clerical, Sales & Service Workers 0.115 1.31 10.94 -0.31 -2.43 0.98

7 Intermediate Production & Transport Workers 0.075 0.68 5.55 -0.04 -0.32 0.40

8 Elementary Clerical, Sales & Service Workers 0.034 0.44 3.55 -0.23 -1.80 0.06

9 Labourers & Related Workers 0.017 0.10 0.83 0.56 4.54 0.09

10 All Occupations 1.000 13.45

602 Visioning

1 Managers & Administrators 0.059 2.17 18.74 -0.36 -2.84 0.94

2 Professionals 0.087 2.01 17.23 -0.36 -2.84 1.24

3 Associate Professionals 0.077 1.75 14.88 -0.50 -3.90 0.85

4 Tradespersons & Related Workers 0.281 -0.24 -1.90 -0.10 -0.77 -0.75

5 Advanced Clerical & Service Workers 0.010 -0.48 -3.80 -0.88 -6.80 -0.11

6 Intermediate Clerical, Sales & Service Workers 0.081 1.31 10.94 0.86 7.07 1.46

7 Intermediate Production & Transport Workers 0.165 0.68 5.55 -0.28 -2.19 0.56

8 Elementary Clerical, Sales & Service Workers 0.079 0.44 3.55 0.23 1.89 0.43

9 Labourers & Related Workers 0.162 0.10 0.83 -0.07 -0.54 0.05

10 All Occupations 1.000 4.66

34

Table 10. Average Annual Growth Rates, O*NET Knowledge Groups

Australia, Hours, Per Cent Per Annum

Historical Forecast

Knowledge Groups Data

1996-97 2004-05

to to

2004-05 2012-13

100 Business and Management 1.78 1.29

101 Administration and management 2.00 1.42

102 Clerical 1.44 1.12

103 Economics and accounting 1.82 1.38

104 Sales and marketing 1.87 1.34

105 Customer and personal service 1.76 1.20

106 Personnel and HR 1.86 1.30

200 Manufacturing and Production Skills 1.28 1.00

201 Production and processing 1.38 1.04

202 Food production 1.14 0.93

300 Engineering and Technology 1.45 0.90

301 Computers and electronics 1.55 1.16

302 Engineering and technology 1.48 0.92

303 Design 1.43 0.88

304 Building and construction 1.55 0.73

305 Mechanical 1.25 0.78

400 Mathematics and Science 1.61 1.12

401 Mathematics 1.67 1.19

402 Physics 1.40 0.91

407 Chemistry 1.42 0.93

404 Biology 1.35 0.98

405 Psychology 1.88 1.31

406 Sociology and anthropology 1.81 1.28

407 Geography 1.65 1.13

500 Health services 1.73 1.18

501 Medicine and dentistry 1.62 1.09

502 Therapy and counselling 1.83 1.27

600 Education and Training 1.90 1.42

…continued

37

Table 10. (continued). Average Annual Growth Rates, O*NET Knowledge Groups

Australia, Hours, Per Cent Per Annum

Historical Forecast

Knowledge Group Data

1996-97 2004-05

to to

2004-05 2012-13

700 Arts and Humanities 1.63 1.17

701 English language 1.70 1.23

702 Foreign language 1.63 1.15

703 Fine arts 1.55 1.09

704 History and archaeology 1.59 1.16

705 Philosophy and theology 1.56 1.13

800 Law and Public Safety 1.72 1.17

801 Public safety and security 1.66 1.09

802 Law, government and jurisprudence 1.79 1.24

900 Communications 1.66 1.20

901 Telecommunications 1.57 1.13

902 Communications and media 1.75 1.26

1000 Transportation 1.51 1.08

9999 All Knowledge Groups 1.63 1.14

38

Table 11. Average Annual Growth Rates, O*NET Work Activities

Australia, Hours, Per Cent Per Annum

Historical Forecast

Work Activity Data

1996-97 2004-05

to to

2004-05 2012-13

100 Looking for/Receiving Job Related Information 1.63 1.13

101 Getting information needed to do the job 1.61 1.11

102 Monitor processes, material, surroundings 1.66 1.16

200 Identifying/Evaluating Job Related information 1.62 1.08

201 Identifying objects, actions and events 1.63 1.10

202 Inspecting equipment, structures, material 1.51 0.96

203 Estimating needed characteristics 1.71 1.16

300 Information/Data Processing 1.64 1.19

301 Judging quality of things, service, people 1.64 1.15

302 Evaluating information against standards 1.57 1.14

303 Processing information 1.59 1.19

304 Analysing data or information 1.77 1.28

400 Reasoning/Decision Making 1.77 1.27

401 Making decisions and solving problems 1.73 1.23

402 Thinking creatively 1.71 1.28

403 Updating and using job related knowledge 1.72 1.25

404 Developing objectives and strategies 1.95 1.43

405 Scheduling work activities 1.87 1.32

406 Organising, planning and prioritising 1.69 1.19

500 Performing Physical and Manual Work Activities 1.40 0.91

501 Performing general physical activities 1.45 0.82

502 Handling and moving objects 1.32 0.82

503 Controlling machines and processes 1.28 0.83

504 Operating vehicles or equipment 1.60 1.25

600 Performing Complex Technical Activities 1.49 1.03

601 Interacting with computers 1.40 0.97

602 Drafting and specifying technical devices, etc 1.56 0.99

603 Implementing ideas, programs, etc 1.63 1.15

604 Repairing and maintaining mechanical equipment 1.20 0.75

605 Repairing and maintaining electrical equipment 1.36 0.90

606 Documenting/Recording information 1.58 1.18

…continued

39

Table 11 (continued). Average Annual Growth Rates, O*NET Work Activities

Australia, Hours, Per Cent Per Annum

Historical Forecast

Work Activity Data

1996-97 2004-05

to to

2004-05 2012-13

700 Communicating with Others 1.75 1.26

701 Interpreting meaning of information to others 1.78 1.31

702 Communicating with other workers 1.79 1.26

703 Communicating with persons outside organisations 1.72 1.26

704 Establishing and maintaining relationships 1.73 1.24

705 Assisting and caring for others 1.62 1.14

706 Selling or influencing others 1.86 1.35

707 Resolving conflicts, negotiating with others 1.91 1.37

707 Performing for/working with public 1.62 1.17

708 Coordinating work and activities for others 1.61 1.11

800 Coordinating/Developing/Managing/Advising 1.62 1.10

801 Developing and building teams 1.51 0.96

802 Teaching others 1.57 1.14

803 Guiding, directing and motivating subordinates 1.73 1.23

804 Coaching and developing others 1.95 1.43

805 Provide consultation and advice to others 1.45 0.82

806 Performing administrative activities 1.60 1.25

900 Administering 1.60 1.19

901 Staffing organisational units 1.63 1.15

902 Monitoring and controlling resources 1.58 1.18

9999 All Work Activities 1.63 1.14

40

Table 12. Average Annual Growth Rates, O*NET Categories

Australia, Per Cent Per Annum

O*NET Category Hours Efficiency

Units

Skill Groups

1.31 100 Content Skills 1.11

1.52 200 Process Skills 1.19

1.51 300 Social Skills 1.20

1.47 400 Complex Problem Solving Skills 1.15

0.90 500 Technical Skills 0.91

1.59 600 Systems Skills 1.29

1.71 700 Resource Management Skills 1.43

9999 All Skill Groups 1.14 1.39

Knowledge Groups

100 Business and Management 1.29 1.64

200 Manufacturing and Production Skills 1.00 1.12

300 Engineering and Technology 0.90 0.76

400 Mathematics and Science 1.12 1.34

500 Health services 1.18 1.44

600 Education and Training 1.42 1.93

700 Arts and Humanities 1.17 1.53

800 Law and Public Safety 1.17 1.40

900 Communications 1.20 1.44

1000 Transportation 1.08 1.16

9999 All Knowledge Groups 1.14 1.37

Work Activities

100 Looking for/Receiving Job Related Information 1.13 1.41

200 Identifying/Evaluating Job Related information 1.08 1.21

300 Information/Data Processing 1.19 1.51

400 Reasoning/Decision Making 1.27 1.64

500 Performing Physical and Manual Work Activities 0.91 0.81

600 Performing Complex Technical Activities 1.03 1.18

700 Communicating with Others 1.26 1.56

800 Coordinating/Developing/Managing/Advising 1.10 1.19

900 Administering 1.19 1.20

9999 All Work Activities 1.14 1.34

41

6. Concluding Remarks

The Australian economy is currently experiencing a period of tight labour markets.

Because of population ageing, the future growth in labour supply is likely to be relatively

slow. In these circumstances, the allocation of training resources becomes particularly

important. They should be allocated in favour of the skills, be they job-specific or generic,

which are otherwise most likely to be in short supply.

Quantitative labour market forecasting is fraught with uncertainty. However, the question

of allocation is in essence a quantitative question. Eventually, someone has to decide what

courses should be provided and that decision, of its nature, should be informed by a view

about the future. Qualitative analyses, such as the Employability Skills for the Future

report, may inform such a view but eventually the qualitative ideas must be assigned

concrete form. If the quantification of the allocation decision is not based on formal labour

market forecasts, it will of necessity be based on informal or intuitive forecasts and there is

no reason to suppose the former will be more uncertain than the latter. Indeed, formal

methods provide a framework for bringing large amounts of relevant information to bear in

a coherent manner, an achievement that lies outside the range of informal methods. In any

case, the two are not mutually exclusive and a decision maker can consider quantitative

forecasts as simply one of a number of information sources to be taken into account.

When it comes to generic or employability skills, the level of quantitative uncertainty

increases. Generic skills are generally not as tightly defined as job-specific skills, and

hence their economic role cannot be easily elicited by the usual method of surveying

market participants. However, by surveying expert opinion, the U.S. O*NET purports to

specify important quantitative aspects of that role in the form of the “importance” and

“level” indicators it associates with its range of generic skills. As there is no alternative

data source against which the O*NET assignments can be compared, their efficacy must be

judged on the basis of the methodology adopted. The O*NET approach seems to be

appropriate to the task and to have been thoroughly implemented. It can reasonably be

argued, therefore, that the indicators embody important information with which to

supplement the quite extensive qualitative discussion of employability skills in Australia

and elsewhere.

42

One purpose of economic research is to develop analytical tools with which to furnish

better information to economic policy makers. The modelling system described in the

paper meets this criterion. The O*NET indicators have been interpreted in a way that is

both in the spirit of their definition and allows existing Australian forecasts of employment

by occupation to be extended to employability skills. The exercise suggests that the

structural details of the future state of the economy do indeed have important implications

for the relative demands for various types of employability skills, and that general

qualitative considerations provide only an incomplete basis for allocating training

resources between those skills. The forecasting methodology has been adapted from a U.S.

database but is amenable to future refinement to incorporate the views of relevant

Australian experts.

43

References

Allen Consulting Group, 2004, Employability Skills: Development of a Strategy to Support the Universal

Recognition and Recording of Employability Skills: A Skills Portfolio Approach, Report prepared to

the Australian National Training Authority.

Allen Consulting Group, 1999, Training to Compete: The Training Needs of Industry, Report prepared to the

Australian Industry Group.

Australian Education Council. Finn Review Committee, 1991, Young People’s Participation in Post-

Compulsory Education and Training, Report of the Australian Education Review Committee,

Canberra, AGPS.

Australian Council for Education Research (ACER), 2001, Employment skills for Australian Industry:

Literature Review and Framework Development, Report to BCI and ACCI.

Department of Education and Training (DEST), 2002, Employability Skills for the Future, Commonwealth of

Australia.

Department for Employment and Training, 2005, Skills for Jobs Growth: Queensland’s Proposed Responses

to the Challenges of Skills for Jobs Growth, A Green Paper, Queensland Government.

Drucker, P., 1994, ‘Jobs and Employment in a Global Economy’, Atlantic Monthly, 1, 132-136.

Dixon, P.B., B.R.Parmenter, J.Sutton, and D.P.Vincent (1982), ORANI: A Multisectoral Model of the

Australian Economy, North-Holland, Amsterdam.

Esposto, A., 2005, Dimensions of Earnings Inequality in Australia, PhD Thesis, Victoria University of

Technology.

Hubbard, M., McCoy, R., Campbell, J., Nottingham, J., Lewis, P., Rivkin, D., Levine, J., 2000, ‘Revision of

O*NET Data Collection Instruments’ National O*NET Consortium. Available at

http://www.onetcenter.org/dl_files/Data_appnd.pdf, accessed 20 November, 2005.

Meagher, G.A., (1997), “Changes in the Demand for Labour in Australia”, in Changing Labour Markets:

Prospects for Productivity Growth, Industry Commission, Melbourne.

Milanowski, A., 2003, ‘Using Occupational Characteristic Information from O*NET to Identify Occupations

for Compensation Comparisons with K-12 Teachers’, Working Paper no. TC-03-06, Consortium for

Policy Research in Education, University of Wisconsin-Madison.

Minnesota Department of Economic Security, (MDES), 2002, ‘Forecasting Short-Term Demand for Skills’,

http://www.mnworkforcecenter.org/lmi/pub1/mms/,

http://www.deed.state.mn.us/lmi/__shared/assets/forecasting_short_term15579.pdf accessed

10/03/2005.

Minnesota Department of Economic Security, (MDES), 2000, ‘Minnesota’s Most Marketable Skills’,

http://www.mnworkforcecenter.org/lmi/pub1/mms/, accessed 05/11/2005.

Neugart, M., and K.Schomann (eds), 2002, Forecasting Labour Markets in OECD Countries, Edward Elgar,

Cheltenham, UK.

Occupational Information Network, O*NET Consortium, 2006, O*NET Data Collection,

http://www.onetcenter.org/dataCollection.html, accessed 28/02/2006.

Occupational Information Network, O*NET Consortium, 2004, O*NET Data Collection,

http://www.onetcenter.org/dataCollection.html, accessed 05/04/2004.

44

OECD (Organisation for Economic Cooperation and Development, 1996, Technology, Productivity and Job

Creation, Paris.

Pappas, N., 2001, ‘Earnings Inequality and Skill’, in Borland, J., Gregory, B., and Sheehan, P. (eds), Work

Rich, Work Poor: Inequality and Economic Change in Australia, Centre for Strategic Economic

Studies, Victoria University, Melbourne.

Peterson, N., Borman, W, Jeanneret, P., Mumford, M., Fleishman, E., Levin, K., Campion, M., Mayfield, M.,

Morgeson, F., Peralman, K., Gowing, M., Lancaster, A, Silver, M., and Dye, D, 2001,

Understanding Work Using the Occupational Network (O*NET): Implications for Practice and

Research, Personnel Psychology, vol. 54, pp. 451-492.

Peterson, N., Borman, W., Jeanneret, P., Fleishman, E., Levin, K., Campion, M., Mayfield, M., Morgeson, F.,

Pearlman, K., Gowing, M., Lancaster, A., Silver, M and Dye, D., 2001, ‘Understanding Work Using the

Occupational Information Network (O*NET): Implications for Practice and Research’ Personnel

Psychology, 54, pp. 451-492.

Peterson, N., Mumford, M., Borman, W., Jeanneret, P. and Fleishman, E. (eds) 1999, An Occupational

Information System for the 21st Century: The Development of the O*NET, Utah Department of

Employment Security, Salt Lake City.

Peterson, N., Mumford, M., Borman, W., Jeanneret, P. and Fleishman, E. (eds) 1997, O*NET Final

Technical Report (Vols 1 and 2), Utah Department of Employment Security, Salt Lake City.

Peterson, N., Mumford, M., Borman, W., Jeanneret, P. and Fleishman, E. (eds) 1995, Development of

Prototype Occupational Information Network (O*NET) Content Models (Vols 1 and 2), Utah

Department of Employment Security, Salt Lake City.

Reich, R., 1992, The Work of Nations: Preparing Ourselves for 21st Century Capitalism, New York,: Vintage

Books.

Rotundo, M., and Sackett, P., 2004, ‘Specific Versus General Skills and Abilities: A Job Level Examination

of Relationships with Wage’, Journal of Occupational and Organizational Psychology, 77, 127-148.

Sheehan, P. and Esposto, A., 2001, ‘Technology, Skills and Earnings Inequality: A New Approach to

Understanding the Characteristics of Jobs’, in Borland, J., Gregory, B. and Sheehan, P., (eds) Work Rich,

Work Poor: Inequality and Economic Change in Australia, Victoria University, Melbourne.

US Department of Labor, 1998a, O*NET Viewer, Version 1, CD-ROM, Washington DC.

45