Manajemen | Fakultas Ekonomi Universitas Maritim Raja Ali Haji 2003 1 (28)

SUPPLY-SIDE AND DEMAND-SIDE
EXPLANATIONS OF DECLINING
APPRENTICE TRAINING RATES:
A CRITICAL OVERVIEW
PHILLIP TONER*

O

ver the last decade numerous academic, industry and government studies have
suggested that Australia has experienced a sustained decline in apprentice training
rates and that this is contributing to shortages in core vocational occupations. This
article redresses significant deficiencies in these studies by providing new data on
long-run apprentice training rates by broad occupational group. This data confirms the
existence of a sustained break in the long-run apprentice training rate from the early
1990s to the present. The article also provides an overview and critical assessment of the
key explanations of this decline and policy recommendations to redress the decline. It is
argued that these explanations may be classified into supply-side and demand-side
approaches. The article concludes firstly, that, in general, demand-side explanations
provide a superior understanding of declining training rates. Secondly, given the
potential importance of trade skill shortages additional research is required both to
quantify the effect of the various demand-side contributions to reduced training rates

and to reorient current policies to better address these demand-side causes.

INTRODUCTION1
Over the last decade it has been argued that there has been a significant reduction
in the overall apprentice training rate. The training rate is the ratio of apprentices
in-training to employed tradespersons. Apprentices in-training is the total stock
of employed apprentices at a point in time. It is the conceptual equivalent of total
employed tradespersons. Concerns have been expressed at the implications of
this decline for the long-run supply of skilled trades for production and maintenance activity. This issue of declining apprentice training rates is not isolated
to Australia, but has been a feature of many countries which have apprenticeship
systems including Great Britain and the US. It has also been the subject of
considerable international research (Ball 1988; Rainbird 1991; Gospel 1993, 1994;
Gann and Senker 1998; Ryan 2000). This study seeks to contribute to this
literature firstly by redressing deficiencies in previous Australian estimates of
apprentice training rates. These deficiencies precluded definitive conclusions
regarding long run trends in training rates. The new data confirm that in the
* Senior Research Fellow, Australian Expert Group in Industry Studies, University of Western
Sydney, Level 8/263 Clarence Street, Sydney NSW 2000. Email: ph.toner@uws.edu.au

THE JOURNAL OF INDUSTRIAL RELATIONS, VOL. 45, NO. 4, DECEMBER 2003, 457–484

458

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

1990s and early 2000s a marked trend decline occurred in employer investment
in apprentice training in the principal trade occupations.
Secondly, the article provides a critical overview of the key explanations
and policy recommendations regarding trends in apprentice training rates in
Australia. Whilst there is a large literature on this topic, it is highly fragmented
with many articles focusing on only one or two potential causes of the decline,
or this decline constitutes only a small part of the subject of the articles or reports.
The present study is the first attempt to provide a systematic, if brief, overview
of the key arguments regarding the decline.

Thirdly, the study argues that the numerous explanations offered for the decline
in apprentice training rates may be usefully divided into two opposing approaches
to labour market and economic analysis. These competing theories may be
characterised as supply-side and demand-side approaches. The task of categorising
and outlining the respective arguments is important, as it is often the case that
in the literature these arguments are juxtaposed indiscriminately which leads to
a lack of conceptual clarity. By classifying the arguments into the two broad
approaches the distinct policy implications of the arguments and approaches can
be more clearly appreciated.
The study concludes that, in general, demand-side explanations provide
a superior understanding of declining training rates. Secondly, given the
potential importance of trade skill shortages, additional research is required
both to quantify the effect of the various demand-side contributions to
reduced training rates and to reorient current policies to better address these
demand-side causes.

SUPPLY

AND DEMAND-SIDE APPROACHES

These two approaches have a long history in economic theory and broader social
sciences. The two approaches have distinct analyses of the bases of production,
exchange and competition. Applied to the issue of apprenticeship training,
the supply-side approach emphasises human-capital type arguments such as
excessive apprentice wages (reducing the demand for apprentices); inflexibilities
on the supply-side of the training system (restrictive curriculum, delivery modes,
and regulatory complexity) and declining apprentice applicant quality, to explain
employers’ declining investment in apprentice training. The supply-side approach
regards the level of investment in apprenticeship as the outcome solely of
rational decision-making based on cost-benefit analyses on the part of individual
employers and apprentices. Adjustment of prices and quantities are assumed to
ensure that any under-supply will be temporary.
The demand-side approach focuses on the structural and institutional
foundations of vocational training and changes in the economy which have
reduced employer capacity to invest in such training.2 The demand-side
approach considers decisions about the level of apprentice training to be determined, in large part, by particular economic characteristics of firms and industries,
and historical and institutional factors. This structural or demand-side approach
is based firstly on the empirical observation of persistent and large differences
in the intensity and propensity of training in general, and apprenticeship

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

459

training in particular, across different economic entities.3 There are marked differences in training propensity and intensity between public and private sectors,
between full-time and causal employees, and across firm sizes and industries. Shifts
in the relative distribution of economic entities with these characteristics will
shift the aggregate propensity and intensity of training. Over the last two
decades a number of changes have occurred which are argued to be adverse
for apprentice training rates. These changes include: corporatisation and
privatisation of public sector entities; reductions in average firm size; increased
competition; growth of casual and part-time employment and growth of
outsourcing and labour hire. Secondly, the demand-side approach argues for
the ‘cultural specificity of many skill-supply mechanisms and their location within
broader systems of production, industrial relations, inter-firm networks, industrial
capital, corporate governance and politics’ (Keep & Mayhew 1999 p. 5).4 Given
the emphasis on structural and institutional factors the demand-side approach
rejects the a priori assumption that the economy will so adjust in the long run
as to equilibrate supply and demand, either at an aggregate level or in specific
activities, such as ensuring an adequate supply of skilled trades.

These differing broad interpretations have significantly different policy
implications for the Vocational Education and Training (VET) system. The
supply-side approach recommends measures to ensure the market for training
more accurately approximates a competitive model through further deregulation
of labour markets and training systems; breaking the historical connection
between vocational training and employment; increasing firm-specific training;
and shifting the cost burden of training from firms to individuals. The policy
responses from the demand-side perspective involve a broad range of specific
measures targeted to offset specific structural impediments to lifting apprentice
training rates.

DATA

ON APPRENTICESHIP TRAINING RATES

Numerous Australian studies have argued that a sustained decline in apprenticeship training rates and/or intake occurred over the last decade (Marshman 1996,
1998; OTFE 1998; Smith 1998, 1999; Toner 1998, 2000a, 2000b, AiG 2000;
BVET 2001). The view of these studies is that, following the 1991–92 recession,
there was a sustained break in the long-run trend apprentice training rate.
However, there was uncertainty about these claims due to data limitations which

restricted analyses to the period from 1986 onwards. This was due to the introduction by the Australian Bureau of Statistics (ABS) in 1986 of a new occupational
classification (ASCO First Edition), which was largely incommensurable with the
previous classification system, the Classification and Classified List (CCLO). The
CCLO had been used from the 1960s. Consequently, it was difficult to calculate
long-run training rates by broad occupational trade group as comparable data
on pre-1986 trades employment were not readily available. (Trades employment
is, of course, the denominator in the calculation of the training rate. This
data is derived from the ABS Labour Force Australia survey. The numerator
is apprentices in-training derived from historical DEET/COSTAC data and

460

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

Table 1

Apprentice training rates, Australia, 1974–2001

June

Metal

1974
1975
1976
1977
1978
1979
1980
1981
1982
1983
1984
1985
1986
1987

1988
1989
1990
1991
1992
1993
1994
1995
1996
1997
1998
1999
2000
2001

20.4
21.1
22.0
20.6
20.8

21.8
23.1
23.9
25.2
25.4
22.7
20.7
18.1
11.0
11.3
11.2
12.9
13.0
12.8
11.1
9.9
9.2
10.4
10.3
10.2

10.0
8.5
8.2

Electrical/ Construction Print Vehicle Food
Electronic
13.8
12.9
13.4
11.7
12.3
11.9
11.5
12.9
14.1
13.2
12.9
11.7
10.0
12.0
11.2
12.7
12.4
12.8
12.3
10.2
9.9
8.9
9.4
9.3
9.0
9.0
9.2
9.3

12.8
13.0
12.9
11.1
11.3
10.9
10.7
10.7
11.5
12.1
10.3
9.6
9.3
9.8
11.2
10.5
12.5
13.0
11.7
9.4
9.9
9.9
9.5
10.1
8.9
9.4
10.4
10.4

13.4
13.2
12.0
10.3
10.9
10.1
10.8
8.7
8.9
8.0
9.4
10.1
9.3
9.5
11.5
7.4
9.5
9.6
8.4
5.4
5.2
7.2
9.6
8.2
6.9
5.8
5.5
9.0

5.8
6.0
6.2
6.1
6.4
6.7
5.7
5.1
5.0
5.1
4.7
4.9
4.7
16.0
17.2
17.2
18.8
17.3
15.8
13.7
13.6
14.1
15.3
15.5
17.9
17.8
18.8
16.4

7.6
8.2
8.2
7.9
8.7
9.3
8.9
10.2
10.2
10.5
10.0
11.2
12.2
12.7
13.8
12.3
12.6
13.8
12.6
12.1
12.2
12.5
13.3
14.2
13.6
15.2
15.1
14.4

December 2003

Other

Total

11.1
11.9
12.3
12.2
12.9
13.9
14.5
13.4
15.1
16.0
14.2
15.0
14.8
17.4
16.9
17.0
17.2
16.3
15.9
14.4
14.8
12.9
13.2
12.2
10.1
11.4
11.4
11.6

12.4
12.7
13.0
11.9
12.4
12.6
12.7
13.0
13.9
13.9
12.7
12.2
11.3
12.3
12.5
12.3
13.3
13.3
12.6
10.8
10.7
10.1
10.7
10.7
10.4
10.7
10.9
10.8

Source: Data for trades employment was derived from Labour Force Australia, Historical Summary 1966–1984
(ABS Cat. No. 6204.0) and Labour Force Australia (ABS 6203.0, various issues). Conversion of pre-1986
CCLO occupational classification to ASCO (First edition) is based on the ABS concordance (ABS Cat.
No. 2182.0). Data on apprentices in-training was derived from COSTAC/DEET Apprenticeship Statistics
for data from 1974 to 1993, from 1994–1998 NCVER (1998) Apprentices and Trainees in Australia
1985–1997 and NCVER unpublished data 1998-2001 inclusive. The NCVER data from 1999 to 2001
are based on ASCO (Second Edition) Major Group 4 New Apprentices in-training, Australia (December
2001 estimates).
Note: in 1987 Automotive Electrical apprentices were allocated from Metal to Vehicle. ASCO (Second
Edition) introduced in 1996 allocated Chefs from Trades to Associate Professionals. Data on trades
employment from 1996 inclusive has been corrected for this.

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

461

more recent National Centre for Vocational Education Research [NCVER] data).
Given the relatively short span of the time series used in these studies, from 1986
onwards, it could have been the case that low training rates evident over the last
decade were not unusual if a longer-run perspective had been used. This doubt
was reinforced by the fact that the period of the mid-to-late 1980s experienced
the highest level of apprentice commencements since records were first published
from the 1960s. In other words, the training rates of the 1990s could simply have
appeared low in comparison with the high rates of the latter 1980s without this
reflecting necessarily any structural break with previous longer-run trends.
This study overcomes this problem by providing data on training rates from
1974 to 2001, and therefore incorporates the effect of several business cycles.5
The problem of incommensurable occupational classification systems was overcome by using an ABS concordance, which permitted the construction of a
consistent time series of Trade occupations (see notes to Table 1). Aside from
the total apprentice training rate, the selected trade occupations for which a
consistent time series over this period could be constructed were Electrical/
Electronics; Construction, Printing; Food and Other.
This data confirms the existence of a sustained break in the long-run apprentice
training rate following the severe recession in the early 1990s to the present
(Table 1 and Fig. 1). Over the 19 years between 1974 and 1992 the training rate
averaged 12.7 per cent. From 1993 to 2001 the total annual average training rate
was 10.6 per cent. This represents a decline of 16.3 per cent. The change between
the two periods 1974–1992 and 1993–2001 is highly statistically significant.6
Trends across occupational groups
Although there was a reduction in the aggregate apprentice training rate
after 1992 there was also considerable variation in training rates across the
Figure 1

Annual apprentice training rates, Australia, 1974–2001

462

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

major occupational groups (Table 2). The Metal training rate declined by nearly
19 per cent after 1993 compared to the latter 1980s.7 Electrical and Electronic
trades training rate declined by close to a quarter after 1993. The training rate
for Construction fell by close to 15 per cent, even though the industry recorded
a significant increase in trades employment over the 1990s (Table 4) and sustained
a record level of real construction output over much of the 1990s and early 2000s
(Toner 2000a). Vehicle had a modest decline of around one-third the total fall
in training rates. Food experienced an increase of just over 5 per cent during the
last decade. The training rate for Printing was volatile over the last decade given
the relatively small numbers of apprentices and tradespersons employed, though
the overall rate fell by 25 per cent. The Other category also experienced a large
slump in the training rate from 1993. (This is due in large part to the decline in
hairdressing apprentices, which comprise the bulk of the Other category.)
An examination was also made of the contribution of the various occupational
groups to the total reduction in the annual average level of apprentices intraining (Table 3). Metal and Electrical contributed 31.5 per cent and 21.7 per
cent respectively to the total decline in the annual average level of apprentices
in-training over the two periods. Construction accounted for 9.3 per cent of the
total decline.
These three broad occupational groups combined represented 50 per cent
of total apprentices in-training over 1993–2001, but contributed 65 per cent of
the decline in annual average level of apprentices in-training over the period.
The other principal contributor to the decline was Other.
As will be described below, the very large variation in the performance of
the apprentice occupations in terms of training rates and numbers in-training
over the period has important implications for assessing the validity of competing
explanations of declining training rates. It is suggested that these findings
are more consistent with the demand-side approach, which focuses on industry
specific structural and institutional factors stimulating or impeding apprentice
training.

Table 2

Apprentice training rates, Australia, 1987–1992 and 1993–2001
Average training rate Average training rate Percentage change
1987–1992
1993–2001
1987–92/1993–2001

Metal
Electrical/Electronic
Building
Printing
Vehicle
Food
Other
Total
Source: Derived from Table 1.

12.0
12.2
11.5
9.3
17.1
13.0
16.8
12.7

9.8
9.4
9.8
7.0
15.9
13.6
12.4
10.6

-18.9
-23.5
-14.7
-25.1
-6.7
+5.1
-25.9
-16.3

Table 3

Contribution to the decline in the total level of apprentices in-training

Source: Derived from Table 1.

Average level of apprentices
in-training 1993–2001

Contribution to decline
per cent

27195
21473
29129
4070
24317
13817
28780
148780

20054
16540
26405
2514
21790
16927
21869
126099

31.5
21.7
12.0
6.9
11.1
-13.7
30.5
100.0

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

Metal
Electrical/Electronic
Building
Print
Vehicle
Food
Other
Total

Average level of apprentices
in-training 1987–1992

463

Trades employment and percent change in trades employment and apprentices in-training, Australia
Average trades
employment
1993–2001

Change in average
trades employment
1987–1992 to 1993–2001 (%)

Change in average
apprentices in-training
1987–1992 to 1993–2001 (%)

226833
175567
254950
43850
142783
106900
218217
1169100

206100
176833
270322
36622
137456
124011
209111
1160456

-9.1
0.7
6.0
-16.5
-3.7
16.0
-4.2
-0.7

-26.3
-23.0
-9.3
-38.2
-10.4
22.5
-24.0
-15.2

I N D U S T R I A L R E L AT I O N S

Source: Labour Force Australia (ABS 6203.0) and Toner 2003.

OF

Average trades
employment
1987–1992

THE JOURNAL

Metal
Electrical
Building
Print
Vehicle
Food
Other
Total

464

Table 4

December 2003

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

IMPLICATIONS

465

OF THE DECLINE IN TRAINING RATES

There are numerous implications flowing from reduced training rates for the
trades. Only two are briefly mentioned here: the effect on skill shortages and
technical innovation.
Training rates and skill shortages
The most obvious implication of the decline in training rates is that the
reduction in the supply of new tradespersons from domestic training sources
is contributing to current skill shortages in trades occupations. The principal
sources of qualified tradespersons are the domestic apprenticeship system
and skilled migration, with the former accounting for around 80–85 per cent
of the flow of trades skills over recent years.8 Numerous surveys and reports
from employer associations, government and academics have identified persistent
skills shortages, notably in metal, electrical and construction trades (AiG 1999;
DETYA 2000a,b, 2002; Construction Training Australia 2001; Worland &
Doughney 2001). A combination of reduced apprentice intake over the 1990s,
continuing high wastage rates from the trades (that is, trades employees
electing to work in other occupations) and economic growth over the last
decade has resulted in significant trade skill shortages. The economic implications
of these shortages are severe. ‘Skill shortages, if extensive and sustained, can
limit investment and growth opportunities, give rise to upward pressure on
earnings and, thereby, dampen the pace of economic and jobs growth and
make it more difficult to reduce unemployment’ (DEETYA 1999a, p. 2). The
National Skills Shortage List produced by the Department of Employment,
Workplace Relations and Small Business which is used for a range of purposes
including the targeting of occupations for the Skilled Migration Program,
identifies shortages in metal, electrical and construction trades (DEWRSB
2002).
Higher vocational skills and innovation
Higher-level vocational skills are central to the development of innovation,
quality product design and production processes and higher productivity.9 In turn,
these skills are crucial to success in world trade in innovation-intensive manufactures. These are products, which ‘require a well-qualified workforce capable
of rapid adjustment in the work process and continual product innovation’
(Finegold & Soskice 1988, p. 21).
A number of studies have identified the links between national differences
in VET performance and national differences in innovation and productivity
(Prais 1995; Anderton & Schultz 1999). Some of the specific linkages between
higher level VET skills and innovation and productivity include lower rework
and defect rates; improved maintenance permitting higher machinery utilisation
rates; improved quality assurance; flatter management structures and more
efficient operation of flexible specialisation equipment which permits closer
tolerances of work and improved quality, product customisation and reduced
cycle times (Toner et al. 2002).

466

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

This strong nexus between the product-market strategies of firms and the
quality and quantity of vocational skills available to them raises a major
potential concern. The current failure to adequately reproduce trade skills
essential in the production and maintenance of sophisticated manufactures in
Australia, and especially in areas such as metal and electrical trades, could lead
managers to develop alternative strategies such as informal upgrading of
employees into positions previously held by people who had undergone
appropriate training. This ‘making do’ with lesser-skilled staff could shift firms’
product-market strategies towards less sophisticated and less innovationintensive products and alter production techniques towards a more intensive
use of less skilled labour.

SUPPLY-SIDE APPROACHES TO THE EXPLANATION OF DECLINING
APPRENTICE TRAINING RATES
The supply-side approach is associated with neoclassical economic theory. The
quantum of training undertaken by employers and employees is the outcome of
the calculus of the discounted time rate of return on such investments (Becker
1964). Strictly speaking, in a perfectly competitive economy, investment in
training will occur up to the point where the marginal cost and marginal
revenue of such training are in equilibrium. From such a view, wage costs and
the efficiency in the supply of training, entailing a range of factors such as price,
flexibility in delivery and relevance of training to the specific needs of the
firm, are of central importance. The following outlines the key arguments in
the supply-side approach to explaining the decline in apprentice training rates
and intake.
Excessive apprentice wages
Sweet (1995, p. 103) has argued that ‘the demand for wage based contractual
training arrangements [apprenticeships] has been declining since the early
1980s, and has continued to decline in to the mid 1990s’. The principal reason
for the ‘failure to expand wage based contractual training arrangements in
Australia lies with inadequate wage structures’ (Sweet 1995, p. 103). The
most important inadequacy is the small relativity between apprentice and
tradespersons’ wages.
As a percentage of the adult rate, apprentice wages appear to differ little from those
applying in classifications to which contractual training is not attached. As a
consequence employers therefore provide minimal training (Sweet 1995, p. 106).

However, to this author’s knowledge there is no evidence in the supply-side
literature that there has been a compression of apprentice and tradesperson
wages over the 1990s which could have led to a substitution of the latter for
the former. One economic analysis of apprenticeship training in Australia
concluded there is ‘No evidence to suggest that wages for apprentices relative
to tradespersons are generally too high’ (Dandie 1996, p. 23).
A recent major research project applied a neoclassical marginal productivity
approach to the determination of employment levels and employer investment

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

467

in apprentice training (Dockery et al. 1996; DEETYA 1997). This study
actually confirmed the Keynesian finding of other research that, change in
the level of a firm’s output is the dominant explanatory variable in an employer’s
decision to hire apprentices (Merrilees 1983). The ‘major single influence on the
number of apprentices employed is the expected level of sales or activity. The
demand for apprentices is thus derived directly from the demand for the firm’s
product’ (DEETYA 1997, p. 31).10 In addition to demand, the study found an
important non-economic factor in firms’ decisions to engage apprentices: ‘firms
appear strongly influenced in their decision to hire apprentices by what may
be termed social or community influences, such as a perceived obligation to
provide training for young people and a sense of obligation to the trade’
(DEETYA 1997, p. 49; Smith 1998, pp. 134–5).
Using data from the 1970s to the early 1990s the study also estimated
the elasticity of employment with respect to apprentice wages. It was found
that the variety of subsidies directed at increasing ‘marginal’ employment of
apprentices, that is, only paid for additional apprentices, had no effect on
total employment. However, other subsidies paid to all apprentices, regardless
of ‘additionality’, were found to have increased the stock of apprentices.
Taking these latter subsidies ‘as a wage reduction equivalent . . . implies an
elasticity in the order of –0.5 to –0.7 ’ (DEETYA 1997, p. 85). The elasticity
is significantly less than unity. A 10 per cent reduction in apprentice wages
would increase the stock of apprentices by between 5 and 7 per cent. To
increase the current level of apprentice intake by 15 per cent, so that the
current training rate would equal the long-run rate prior to 1993, would
require apprentice wages to fall by between 24 and 34 per cent. However, the
study found ‘very little support among employers for the notion of a reduction
in apprentice wages’ (DEETYA 1997, p. 45). This was because such a fall
‘would be accompanied by a decline in the average quality of the apprentice
intake. Second, many [employer] respondents stated that the apprentice
wage should not be reduced as it would be inequitable to do so’ (DEETYA 1997,
p. 48).
In summary, there is no evidence in the supply-side literature of a compression
in apprentice/tradespersons wage relativity over the last decade that could account
for decline in apprentice training rates. Nevertheless, on the basis of the positive
historical experience with the introduction of wage subsidies (see Endnote 5)
and the econometric evidence cited above, it is possible that an increase in
the level of current financial incentives for the employment of apprentices could
have a marginally beneficial effect in redressing the declining training rate.
In theory, a reduction in apprentice wages could have the ‘same’ effect as a
subsidy though, in practice, concerns have been expressed at the effect of such
a reduction on the quality of applicants. There are also anecdotal reports that
the low level of apprentice wages relative to other employment opportunities
for young people acts as a disincentive for more able potential apprenticeship
applicants.
This review finds the supply-side explanation of falling training rates based on
a compression of apprentice/tradesperson wages is not well founded, though a

468

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

case can be made for using adjustments to the level of apprentice wages through
a subsidy as a partial solution to declining training rates.
Inadequate financial incentives to employ apprentices
The current system of financial incentives offered to employers of apprentices
is inadequate in its response to identified skills shortages and, arguably, biased
against the employment of apprentices in favour of trainees. The following sets
out some of the problems.
Commonwealth financial incentives are paid to employers of apprentices
and trainees upon commencement of the person in employment and upon
completion of their training. The same level of payment for the commencement
and completion of New Apprentices applies to both apprentices and trainees
undertaking Australian Qualifications Framework (AQF) III and IV courses.
These payments do not recognise the fact that some AQF III or IV traineeships
can be completed in much less time (even one year) compared to the four year
term of a metal, electrical or construction apprenticeship. As such, the payments
do not recognise the much greater investment of time and effort on the part
of the apprentice employer. In theory, an employer could get four cycles of
commencement and completion payments for trainees in the same time it
takes an employer of an apprentice to get one cycle. The argument is not that
the incentive programme has led to a significant substitution of apprentices for
trainees, the latter only account for around 10 per cent of New Apprentices
classified to ASCO major Group 4 Trades, rather, it is that the incentive programme does not adequately recognise the greater investment of apprentice
employers in training (Toner 2000b).
Secondly, related to the argument above, the incentive programme fails to
adequately target scarce training funds on strategic skill shortages. In particular
it has been found that traineeships are producing an inordinate share of lower
skilled occupations compared to their distribution within the total workforce.
Over one-third of the annual intake of trainees in 2001 consisted of Labourers
and Elementary Clerical occupations, and another third were Intermediate
Clerical and Production workers (Toner 2002b). Between 1995 and 2000 the total
number of New Apprentices increased by 100 per cent from 136 000 to 276 00
but the total number of apprentices in-training was largely static (NCVER 2001,
p. 66).11
There are now well established precedents for the use of strategic targeting
for employer incentives in vocational training. Only minor adjustments to the
programme guidelines would be required to significantly improve the effectiveness
of these schemes.
Flexibility in the training market
The apprenticeship system is also argued to be in decline due to ‘inflexibility’ in
the system, such as mandated course content and time frames for completion of
training; lack of choice over training delivery methods; lack of competition in
the provision of training; and inadequate focus on the specific training needs of
firms.

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

469

Reform of industry training should, in the first instance, be about letting markets
work and making sure that the right kind of training is delivered. Making training
more flexible and responsive to what the market wants will in turn make the
industry more efficient (ACIL 1998, p. 18).

This line of analysis also leads to a view that the historical foundation of the
system, that is, the link between structured training and employment, needs to
be broken or significantly altered. There are two aspects to this line of thought.
Firstly, training could become institutionally based, whereby an individual can
study for a range of vocational occupations at private and public colleges
without contemporaneous employment in the field of study (ACIL 1998, p. 1;
AiG 1999, p. xii). Secondly, broader labour market deregulation is required,
especially direct workplace negotiation between employees and employers as
to the type, delivery, quantum and cost burden of training.
[The] labour market will only become truly efficient and flexible and encourage
meaningful training arrangements between employers and employees when genuine,
freely negotiated contracts between employees and employers are possible (ACIL
1998, p. 7).

Over the last decade there has been a comprehensive implementation of policies
designed to introduce a market for training and remove ‘inflexibilities’ in the
apprenticeship system. These measures include, for example, the removal of age
restrictions for apprenticeships, removing restrictions for certain types of work
to be undertaken only by apprentices or tradespersons, opening up of trade-type
training to traineeships, introduction of competency based training in place of
‘time served’ and making public funds for off the job apprenticeship training
subject to competition between the public and private sectors (ANTA 1998a;
Webster et al. 2001; NCVER 2001, p. 57).
These extensive changes have not restrained the decline in apprentice
training rates.12 In addition, some of the recommendations for increasing
flexibility in apprenticeship arrangements may undermine the traditional
economic rationale for employer investment in apprentice training. For example,
a significant shortening in the length of the apprenticeship, rather than the
traditional time-served indenture could, paradoxically, reduce employer investment in training. This is because the duration of the traditional apprenticeship,
usually four years,
can be viewed not as a calculated period necessary to acquire trade training but as
a set period which enables an employer to gain a return on the low productivity
of the trainee during the early part of the apprenticeship. In other words, the
apprenticeship can be seen as being artificially lengthened to prevent other
employers from ‘poaching’ the newly trained worker until such time as the individual
is regarded as having paid his initial employer for his training (Curtain 1987, p. 18).

A reduction in the period of training before the apprentice receives their trade
qualification will, ceteris paribus, reduce the time in which the employer has to
recoup their investment.

470

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

Inadequate firm specific training
Inadequate orientation of the apprenticeship system towards the provision of
firm specific training is also offered as an explanation of declining apprentice
training. A shift in the balance of training towards firm-specific, as opposed to
general or industry-specific training, is advocated as a means of increasing
employer demand for VET in general, and apprenticeships in particular (AiG
1999, p. xv). The content and delivery methods of training should be tied
more directly to the needs of particular employers. The case for increased firmspecific training is a subset of the more general ‘flexibility’ argument, though
its importance deserves separate consideration.
This explanation of declining apprenticeships may be criticised on a number
of grounds. Firstly, like the argument regarding compression of wage relativities
and demand for increased flexibility in training markets, the stress on firm specific
training does not explain the marked variance in trends in training rates and
levels of employment across the broad apprentice occupations. As described
earlier some large apprentice occupations such as Food have experienced
significant growth in employment and training rates, whilst Vehicle declined
by only one third the aggregate fall in training rates over the last two decades.
As a general explanation then, the argument regarding inadequate provision of
firm specific training is not convincing. Further, advocates of this argument
need to explain why this factor should only have begun to dramatically affect
aggregate training rates over the last decade.
Secondly, flexibility needs to be an important feature of any training system,
though it also needs to be recognised that an undue emphasis on firm-specific
training contradicts a key economic rationale for the provision of public funds
for vocational training. Commonwealth and state expenditure on the VET
system is ‘well over $1 billion per year’ (NCVER 2001, p. 27). A key rationale
for the provision of public funds for VET is that there are significant economic
benefits to society as a whole in providing workers with industry specific or
general skills (Becker 1964). General skills are the broad range of vocational
skills that are necessary for undertaking work either for a particular employer,
for the industry, or even the economy as a whole. Individual employers, it is
argued, will not invest in industry or general skills, as they will not necessarily
reap the returns from such investments as workers can freely contract with other
employers. Another way of expressing this is that there is market failure in the
private provision of general, as opposed to firm-specific, training. The benefits
to society from the investment of public funds in general skills are an improvement in the quality and productivity of the workforce and an increase in the
efficiency of the labour market by improving the intra-firm, intra-industry
and inter-industry mobility of labour. This mobility of labour is essential for the
economy to adjust in the face of structural change and gain the benefits of an
efficient allocation of labour resources.
The Australian labour market for trades has traditionally been one with high
inter-firm labour mobility. This high mobility is based on formal and informal
recognition of the equivalence of trades qualifications and experience by
employers (Curtain 1987, p. 28–29). This essential aspect of the trades labour

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

471

market could be lost if training becomes overly firm-specific (Buchanan &
Callus 1992). This mobile and ‘occupational’ labour market is due to a number
of factors, notably the small size of Australian firms, which has precluded the
development in the private sector of extensive ‘internal’ labour markets (Curtain
1987). The comparatively small size of the Australian market has also prevented
the creation of an intensive division of labour, resulting in smaller firms requiring
more broadly skilled trades labour. In turn, this has been influential in the
maintenance of identifiable trade occupations (Gospel 1994).
Further, the argument for more firm-specific training is based solely on
the interests of the employer and ignores the interests of the apprentice. By
accepting a ‘training wage’ a person undergoing vocational training is also
investing considerable time and forgone earnings in the acquisition of marketable
skills. It is only equitable that individuals investing in such training receive an
appropriate balance between firm-specific and industry/general skills, given that
they can expect to have multiple employers over their working life.13
Declining quality of apprentice applicants
A number of surveys of employers report that the quality of applicants for
apprenticeships is declining (Marshman 1996, p. 24; ACIRRT 2002, p. 35;
DETYA 2002, p. 18). These surveys suggest that an important ‘factor identified
as having a major effect on apprentice recruitment was a supply-side factor,
namely the quality of recruits available, suggesting that many employers would
increase the number of apprentices employed if higher quality candidates
were available’ (DEETYA 1997, p. 31). It is also undoubtedly the case that many
trade fields, such as metals and engineering do have an ‘image problem’ and ‘no
doubt the size and quality of the apprentice applicant pool could always be
improved’ (Hall & Buchanan 2000, p. 8). The key policy recommendation
from this line of analysis is to improve the marketing of apprenticeships
(DETYA 2002).
Sweet (1990, p. 233) notes that, whilst school retention rates to year 12 have
increased markedly, there has not been a corresponding increase in the proportion
of Year 12 apprenticeship commencements. This implies that more academically
able students are remaining in the education system rather than electing to enter
an apprenticeship. Smith (1998, p. ix) cites a number of studies indicating
that apprentices and trainees have ‘inadequate language, literacy and general
reasoning skills’. These inadequacies create significant learning problems,
especially with the move away from face-to-face teaching, and increased use
of computer based and distance learning. The studies cited by Smith, however,
do not demonstrate whether apprentices’ difficulty with off-the-job formal
training is a recent development reflecting declining intake quality or a long
term problem with apprentice intake.
On the other hand, research by the federal department responsible for the
administration of the apprenticeship system found that although there may be
shortages of quality applicants in particular regions or particular trades, overall,
employers reported seven suitable applicants for each vacancy. Group Training
Companies reported three suitable applicants for each vacancy. Smaller firms have

472

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

more of a problem than larger firms in attracting suitable applicants (DEETYA
1998, p. 5).
Overall, it appears highly probable that, due to a broad range of factors, such
as the expansion of tertiary education and associated rapid growth of professional
and paraprofessional employment and the ‘image problem’ of some traditional
trades, there has been some reduction in the quality of applicants. In turn, this
has led employers to reduce their apprentice intake. However, this appears to be
only a minor factor in explaining the decline in aggregate training rates.
Demand-side explanations
In contrast to the supply-side interpretation the demand-side approach emphasises structural shifts in the economy and institutional changes that have reduced
employer investment in training in general and apprenticeship training in particular. These changes include a set of self-reinforcing factors, such as greatly
increased competition leading to increased down-sizing and contracting out;
growth of labour hire companies; increase in the proportion of small firms
and privatisation and corporatisation of public services. Some of these major
structural changes are summarised below.
Reduced demand for trade skills?
It is crucial to note that the decline in training rates and absolute level of apprentices does not reflect a proportionate reduction in demand in the economy for
trade skills. It would be expected that a trend decline in the stock of employed
tradespersons would be accompanied by an equiproportional decline in the flow
of new entrants into the trade.
This is not the case. The average total trades employment between 1987–92
and 1993–2001 fell by under 1 per cent, but the total number of apprentices
in-training fell by over 15 per cent (Table 4, column 4). The fall in the level
of Metal and Vehicle apprentices in-training was nearly three times the decline
in the number of Metal and Vehicle tradespersons employed. The number of
Electrical apprentices in-training declined by nearly one-quarter but the employment of Electrical trades increased by nearly 1 per cent. The average number
of building trades declined by 9.3 per cent over the period but employment of
building tradespersons increased by 6 per cent.
Corporatisation and privatisation of government activities
Toner (1998) found that a major contributor to the reduction in apprentice
numbers over the 1990s was the large-scale withdrawal of all levels of government
from apprentice training. The reason for this withdrawal is largely due to the
corporatisation or privatisation of state and Commonwealth government
activities. Later studies have confirmed this finding (Worland & Doughney 2001;
ACIRRT 2002). Changes in the ownership or legal structure of these entities
led to a shift in their objectives away from a broad range of economic development and social objectives to a singular focus on improving the direct rate
of return on the private or public funds invested in the entity. Prior to corporatisation and privatisation these entities saw it as their role to train for ‘their’

D E C L I N I N G A P P R E N T I C E T R A I N I N G R AT E S

473

industry and engaged a large proportion of their apprentices on the basis
that they would be employed only up to the completion of their training
and then be released to the industry. In the mid-1980s NSW Government
Business Enterprises (GBEs) alone employed 21 per cent of all electrical
apprentices; 10 per cent of building apprentices and 9 per cent of metal
apprentices. By the late 1990s the GBEs share had been reduced by 80 per cent.
It is estimated that the withdrawal of the public sector accounted for around
one-third of the decline in apprentice intake over the last ten years (Toner 1998).
Similar large falls in public sector apprentice intake occurred throughout
Australia (NCVER 2001, p. 75). Crucially, there was no compensating
increase in private sector apprentice employment in the metal, electrical and
construction occupations.
Increased competition
Marshman (1996) and (ANTA 1997, p. 12; ACIRRT 2002; Toner & Wixted 2002)
have argued that factors such as tariff reduction and globalisation have increased
the intensity of competition. Increased competition is especially evident in the
trade-exposed areas of the economy, such as manufacturing, where the effects
of international competition and globalisation have greatly intensified.
Manufacturing is the principal employer of metal trades and to a lesser extent of
electrical trades. One measure of this intensification of competition is that in
1989–90 the manufacturing trade deficit was 12.3 per cent of total manufacturing
turnover; by 1999–2000 it had increased to 19.4 per cent (Toner & Wixted 2002,
p. 33). Another indicator of increased competition is that the profit margins and
rate of return on manufacturing assets have not recovered over the decade from
rates recorded in the deep recession of the early 1990s (ABS, Cat. No. 8140.0,
various issues).
This section highlights four interrelated responses to increased competition.
These responses are the intensification of work, introduction of ‘lean production’
systems, growth of labour-hire and increased outsourcing of functions.
Littler (2003, p. 9) has shown that, internationally, corporate downsizing
and ‘delayering’ (removal of management layers) over the 1990s has led to an
intensification of work, as measured by increased working hours and increased
responsibilities of the remaining workforce. The recent ACIRRT (2002,
p. 33–38) study of Victorian manufacturers found that as a result of reduced
employment the intensity of work in the remaining workforce had increased
to the point where there was simply no surplus labour capacity to disengage
experienced tradespersons from production to train and mentor apprentices.
This lack of surplus labour capacity not only adversely affected the quality
of training for existing apprentices but also made firms reluctant to engage
apprentices. Employers also reported that intensified competition and tighter
margins had reduced their financial capacity to invest in the level of training that
they would like.
Increased competition has resulted in unpredictable and shorter contract cycles.
This has introduced uncertainty about the future, which makes employers
reluctant to enter into the 3-4 year contract involved in the employment of

474

THE JOURNAL

OF

I N D U S T R I A L R E L AT I O N S

December 2003

apprentices. Apart from not having the capacity to carry the additional cost, they
do not want to have to put them off if contracts do not eventuate (Marshman
1996, p. 12). Although not explained by Marshman and others, the move to
shorter contract cycles reflects the introduction of ‘lean production’ production
techniques aimed at reducing a firm’s cost and risk (Harrison 1997). A key
element in lean production is the Just in Time (JIT) production method.
Under JIT, sub-contractors enter into ‘open ended contracts’ with principals, in
which there is no guarantee of a fixed quantity of work over a given period.
JIT offers principals considerable competitive advantage through cost savings,
especially reduced inventory holdings. It is also a method for shifting the risk of
market fluctuations from the purchaser of a good or a service onto the supplier,
as the purchaser is not locked into a fixed long term contract. However, JIT
production methods make it difficult to provide the continuity of work required
for an apprenticeship and they are also a factor in the growth of labour hire. Nearly
50 per cent of employers identified meeting peak production demands as the
main reason for the use of labour hire (ANTA 1998b, p. 26). Other research on
the drivers of training has found that ‘lean production . . . is associated with the
reduction of training activities’ (Smith et al. 2002, p. 8).
Many of the industries that were traditionally the principal employers of
apprentices, such as manufacturing, construction, and electricity gas and water,
are also disproportionately outsourcing production, maintenance and other
services (Hall & Bretherton 2000). The significance of this outsourcing of
trade-based work to labour hire companies is that ‘labour hire firms primarily
rely upon the pool of skilled people in the labour market, and are not large
providers of formalised training of the type involved in the traditional
apprenticeship’ (ANTA 1998b, p. 1). The growth of labour hire, in turn,
reflects increased competition and the need to reduce costs. Related to this is
development of ‘lean production’ systems, based on the ‘downsizing’ of firms with
a smaller ‘core’ full-time workforce and the growth of a ‘peripheral’ workforce
comprised of labour hire employees and other part time or casual employees
(Harrison 1997).
This is a complex issue however, as recent researc