D
Young learners
This appendix presents estimates of the effects of vocational education and training
(VET) reforms that have been, or will potentially be, implemented, as part of the
Council of Australian Governments (COAG) reform agenda. Effects relating to
changes in educational attainment among learners aged 15 to 24 are the focus of
analysis. Estimates for mature learners (aged 25 to 64) are presented in appendix E.
The Productivity Commission has estimated the:
realised and prospective effects of reform in Victoria (section D.1)
effects of contestability in Victoria (section D.2)
prospective effects of reform in South Australia (section D.3)
realised effects of the National Partnership Agreement on Productivity Places
Program (NPAPPP) (section D.4)
potential effects of reforms Australia-wide as defined in chapter 3 (section D.5).
The Commission’s modelling work estimates the effects of changes in qualification attainment attributable to policies enacted under the COAG reform agenda, on economic measures like workforce participation and productivity and on social inclusion. Governments measure their expectations of changes in VET activity as a consequence of reform in terms of increases in training places or enrolments, not qualification completions. The Commission has had to estimate additional completions under the COAG reform agenda from information on places, enrolments and students. The Commission has made many assumptions in estimating the impacts of COAG’s VET reform agenda. These are detailed throughout the report. Sensitivity analysis undertaken indicates that those assumptions are critical. When the assumptions change, so do estimated changes in workforce participation, productivity and broader economic activity, often quite markedly. The results, therefore, should be regarded as only broadly indicative of the possible impacts of the policy initiatives assessed. The estimates and analysis are intended to advance understanding of the scale of benefits that might accrue. The estimates presented are not forecasts of the economic or fiscal impacts of the reforms. YOUNG LEARNERS
83
D.1 Realised and prospective effects of the introduction of a VET entitlement in Victoria Increases in student numbers attributable to the Victorian policy initiative have been translated into qualification completions, by highest level of attainment, before the impacts of policy changes have been assessed. A number of assumptions are used in this exercise. Key among these are that:1
government-funded student numbers in 2008 represent the baseline for calculations of additional training outcomes attributable to the Victorian training entitlement
student numbers increase by 5 per cent during 2012
qualification completion rates for young learners are 36.5 per cent for Diploma and Advanced Diploma students, 35.3 per cent for Certificates III and IV (excluding traditional trade apprenticeships) and 28.8 per cent for courses at and below Certificate II (based on Karmel 2011b)
the completion rate in traditional trades is 56.6 per cent (Karmel 2011a)
people are assumed not to undertake traditional trade qualifications after age 25
students who complete Diploma and Advanced Diploma qualifications do so in the second year of their enrolment. Those who do not, leave the VET system in the first 12 months of study
80 per cent of Certificate III and IV students in qualifications other than traditional apprenticeships leave the VET system (either with, or without completing a qualification) within 12 months of commencing their course of study. The remaining 20 per cent leave within 2 years
10 per cent of students who complete a Certificate I or II go on immediately to complete at least a Certificate III or IV, and 7 per cent of those who complete a Certificate III or IV go on immediately to complete a Diploma or Advanced Diploma. (Transitions into higher education are ignored.)
Over the period 2009 to 2011 (the realised period), the Commission estimates that young learner student numbers will be about 88 000 higher than would have been the case in the absence of policy changes (table D.1).2 Reflecting the rapid growth in enrolments, student numbers in 2012 (the prospective period) are estimated to be about 70 000 above the baseline. 1 A full list of assumptions is presented in a spreadsheet available on request from the study team. 2 Using data on actual student numbers to the third quarter of 2011. 84
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
These student numbers are estimated to translate to 20 000 more qualifications due to the realised increase in student numbers, than would otherwise have been the case. The prospective increase in qualification holding, relative to 2008, is higher, at about 30 000, because a proportion of students who enrolled between 2009 and 2011 do not complete until 2012 (or later in the case of traditional apprentices). The increases in student numbers estimated to translate into increases in Diploma and Advanced Diploma, and Certificate III and IV completions are described below (table D.1). These are the numbers of completions that form the basis for the scenarios described below. Table D.1
Estimates of additional VET qualifications due to Victorian entitlement policies Victorian young learners (aged 15 to 24) Realised increase
Additional students — relative to 2008 Completions Diploma and Ad. Diplomaa Certificates III and IVa Certificates I/II and prevocational courses
Prospective increase
88 182
69 866
18 162
29 100
498
4 010
11 168
20 027
6 496
5 064
a Figures used as targets in the ELMO model. Source: Productivity Commission estimates.
Effects on education and occupational outcomes The increase in qualification completions due to the introduction of an entitlement in Victoria reflects the associated reduction in the costs of VET courses to students, relative to alternatives (tables D.2 and D.3). The increase in the number of people with VET qualifications as their highest level of education is determined by the characteristics of the entitlement (for example, the additional number of places, and eligibility requirements), and is matched by a fall in the number of people in other education groups. The model projects that the percentage of the Victorian young learner cohort with Year 11 and below as their highest level of education falls nearly 1 per cent relative to the base case, and those with Year 12, by 10 per cent as a result of realised reforms. The equivalent projections for the prospective reforms are 2.3 per cent for Year 11 and 22.6 per cent for Year 12.
YOUNG LEARNERS
85
Table D.2
Realised change in the distribution of highest educational attainment due to Victorian entitlement policies Per cent, Victorian young learners (aged 15 to 24)
Year 11 Year 12 Cert. III/IV Diploma Degree Total
Base
Policy
Change
17.98 12.89 25.00 9.97 34.16 100.00
17.83 11.56 26.42 10.03 34.16 100.00
-0.87 -10.27 5.68 0.60 ..
.. less than 0.005 Source: Productivity Commission estimates.
Table D.3
Prospective change in the distribution of highest educational attainment due to Victorian entitlement policiesa Per cent, Victorian young learners (aged 15 to 24)
Year 11 Year 12 Cert. III/IV Diploma Degree Total
Base
Policy
Change
17.29 11.56 26.25 10.74 34.16 100.00
16.88 8.95 28.77 11.24 34.16 100.00
-2.36 -22.61 9.58 4.73 ..
aBase includes realised outcomes. .. less than 0.005 Source: Productivity Commission estimates.
Changes in educational attainment are estimated to increase employment in occupations more commonly entered by people with VET qualifications, such as community and personal services (3 per cent for realised reforms and 5.5 per cent for prospective reforms) and technicians and trades (0.6 per cent for realised reforms and 1 per cent for prospective reforms) (tables D.4 and D.5). Employment in occupations more commonly entered by people with non-VET qualifications is estimated to decrease — for example, for clerical and administrative jobs (around 1.4 per cent reduction for realised reforms and 2.5 per cent reduction for prospective effects).
86
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Table D.4
Realised occupational outcomes from Victorian entitlement policies Victorian young learners (aged 15 to 24) Base
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers Total
Policy
Change
no.
no.
%
95 766 161 238 53 880 48 660 76 517 35 671 19 581 26 985 518 299
96 131 161 230 54 223 50 115 75 464 35 460 19 516 26 810 518 948
0.38 -0.01 0.64 2.99 -1.38 -0.59 -0.33 -0.65 0.13
Source: Productivity Commission estimates.
Table D.5
Prospective occupational outcomes from Victorian entitlement policiesa Victorian young learners (aged 15 to 24)
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers Total
Base
Policy
Change
no.
no.
%
96 154 162 244 53 619 50 435 75 394 35 312 19 388 26 673 519 219
96 766 162 777 54 133 53 214 73 449 34 919 19 135 26 129 520 522
0.64 0.33 0.96 5.51 -2.58 -1.11 -1.30 -2.04 0.25
aBase includes realised outcomes. Source: Productivity Commission estimates.
Effects on participation and productivity The main mechanisms affecting aggregate productivity are summarised in box D.1. Workforce participation by the Victorian young learners cohort is projected to rise by 0.13 per cent, and productivity to rise by 0.02 per cent, as a consequence of the realised changes (table D.6). Workforce participation is projected to increase by 0.25 per cent in the prospective simulation, while productivity is projected to rise by 0.04 per cent (table D.6 and box D.2). Workforce participation increases because there is a positive relationship between education and participation, and the entitlement is estimated to increase general education levels. YOUNG LEARNERS
87
Box D.1
Changes in incentives and substitution between qualifications
An expansion in the number of government-funded places reduces the private costs of vocational education and training (VET) to some individuals. This change in the structure of private incentives can be expected to increase the number of students who undertake VET. The degree to which a potential student responds to changes in incentives depends on the substitutability between different education—work choices, which is regulated by the relative net benefits of the options that they face. The Educational and Labour Market Outcomes (ELMO) model projects changes in the likely education and work choices of individuals as incentives are changed. The vast majority of adjustments to an expansion in government-funded VET places occur between the year 12 and Certificate III/IV groups. In other words, relative to the base case, as the number of government-funded VET places increases, the most common response projected by ELMO is that individuals who might otherwise not have studied beyond Year 12 go on to a Certificate III or IV qualification. The model assumes that it is the more able Year 12 graduates who choose to invest additional time and funds (to the extent that the places are not fully funded) into pursuing a VET qualification. The workforce participation and productivity of these individuals are projected to increase. In particular, their productivity increases by up to the difference in average wages of Year 12 and VET graduates (3.8 per cent for a Certificate III/IV — the vast majority — and up to 8 per cent for a Diploma — a very small number of the group. The relevant wage premiums are in table C.4). People who might otherwise have planned (that is, in the base case) on undertaking a higher education degree, could respond to changes in the incentives before them by pursuing a VET qualification. The relevant population would be that of the less able higher education students, and they would be most likely to switch to Diploma study. The investment in time (typically more than 3 years) and funds that are associated with a degree is significantly larger than the investment required to obtain a VET qualification. This is reflected in the relatively large wage premiums that are typically associated with degrees; in this context, the average wage premium between a degree and a diploma is estimated to be 24.3 per cent (table C.4). In the version of the ELMO model used in this report, the substitution between degrees and VET qualifications is assumed to be limited. To the extent that this is not the case, the results could be different. Assume that an individual substitutes VET for higher education. In the process, they forego the productivity premium associated with a Degree. As a result, his or her productivity would be up to 24.3 per cent lower than in the base case. On the other hand, the fiscal cost of a VET qualification is a fraction of that of a Degree, which has implications for the net social benefit stemming from the individual’s decision. A sensitivity analysis performed with ELMO shows that allowing for some substitutability between higher education and VET qualifications translates into a small number of individuals choosing a VET qualification over a degree. This is likely to reduce estimates of aggregate productivity and increases net social benefit.
88
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Box D.2
Contributors to aggregate productivity
The projected increase in aggregate productivity can be decomposed into two components.
The first component, shows the contribution of changes in individual productivity, holding participation constant. For example, when individuals move from the Year 12 category to the Certificate III/IV category, their productivity increases by the wage premium associated with this move. The contribution of these changes in individual productivity to aggregate productivity amounts to increases of 0.039 per cent for the realised reforms and 0.077 per cent for the prospective reforms.
The second component, the increased participation component, accounts for the effect of increased participation on aggregate productivity, holding individual productivity constant. The increase in educational attainment increases the participation of people who would not otherwise have been in employment. It is assumed that these new entrants are less productive than people already in employment, which reduces aggregate productivity. The participation component is estimated to account for a 0.019 per cent reduction in productivity for the realised reforms and 0.041 per cent for the prospective reforms.
Table D.6
Participation and productivity changes due to Victorian entitlement policies Per cent, Victorian young learners (aged 15 to 24)
Participation Productivity Individual productivity component Increased participation component
Realised
Prospective
0.125 0.020 0.039 -0.019
0.251 0.036 0.077 -0.041
Source: Productivity Commission estimates.
Effects on net social benefit The net social benefit of the realised changes in Victorian young learners’ educational attainment is estimated to be $0.8 billion in net present value terms, and $1.4 billion for prospective changes (table D.7).
YOUNG LEARNERS
89
Table D.7
Estimated effects of Victorian entitlement policies $ million, Victorian young learners (aged 15 to 24)
Outcomes
Gross payments to labour less value of non-market activity forgone by working less money cost of education plus adjustment to capture the value of government revenue plus external benefitsb plus other benefitsc Net social benefit
Changea Realised
Prospective
2 102 683 -37 -4
4 103 1 370 -48 -81
0
0
-638 814
-1 299 1 401
aTotal change in present value terms over 42 years. bZero by assumption. cBenefits of education estimated through model calibration; include any consumption value of education. Source: Productivity Commission estimates.
Net social benefit is calculated from changes in labour income before taxes. Gross labour income is projected to increase by about $2.1 billion and $4.1 billion due to realised and prospective changes, respectively. An estimate of the costs of non-market activities foregone (for example, leisure and caring duties) due to the increase in workforce participation is subtracted from this figure ($0.7 billion for realised changes and $1.4 billion for prospective changes), as is the change in the money costs of education.3 Adjustments have also been made to capture the value of government revenue4 and other benefits of education — for example, the consumption benefits that might accrue to individuals in the course of their study. The modelled policies could counteract or enhance the effects of existing policies (box D.3). Partial fiscal effects The Victorian entitlement project entails additional government expenditure over four years. To the extent that it increases wages and participation, it also increases future tax revenue, assuming current income tax settings. The net effect is estimated to result in an increase in net government revenue of around $18 million for realised changes and $348 million for prospective changes. 3 The time cost of studying is calculated as labour income foregone, and accounted for in gross payments to labour. 4 The value of tax revenue is assumed to be $1.24 per dollar of tax revenue (appendix C). 90
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
This calculation is akin to an investment calculation in which an initial outlay — the additional expenditure associated with increasing the number of subsidised places — is compared with a future stream of revenue — the additional tax revenue that might arise from increases in income induced by the policy. This net effect abstracts from any other effects on tax revenue, such as through increased GST collection that might arise from additional consumption that occurs as income increases, or additional payroll tax, excise tax, etc. This would require detailed tax modelling that is beyond the scope of the Educational and Labour Market Outcomes (ELMO) model.5 An indication of the order of magnitude of the GST revenue not accounted for could be obtained by assuming that all the additional income is consumed and spent on goods and services that are subject to GST — this additional revenue would amount to 10 per cent of the estimated after-tax payments to labour. The fiscal effects also depend critically on the effects on wages. To the extent that some of the partial equilibrium assumptions mentioned in appendix B are not satisfied and wages might be higher or lower than projected, future tax revenue could also be higher or lower than projected. Box D.3
Effects of income taxes and education subsidies
Taxes on labour income reduce the after-tax returns to a qualification. Relative to a situation without taxes (or with lower taxes), this results in an underinvestment in education. Education subsidies reduce the private cost of education and, relative to a situation without subsidies, provides incentives to over-invest in education. To the extent that subsidies bias choices away from the socially optimal mix, this can be a source of distortion. To some extent, the subsidy implied by the expansion of publicly-funded VET places counteracts the effect of the tax on labour income. In the Victoria realised scenario, it could account for about 80 per cent of the estimated net social benefits.a a This figure is obtained as the difference between running the ELMO model with the current tax setting and without it, and allowing for some substitution between degrees and VET.
5 Given the character of the policy, it is difficult to estimate a ‘steady-state’ situation with constant annual expenditure and revenues. That said, the Monash Multi-Regional Forecasting (MMRF) model could be used to generate changes in aggregate annual tax revenues. YOUNG LEARNERS
91
Sensitivity analysis Numerous modelling assumptions have been made regarding parameter values in estimating the realised and prospective effects of Victorian entitlement. Altering these assumptions alters the estimated effects. The sensitivity analysis below illustrates how the estimated effects change in response to changing many of the sets of assumptions that underlie the model (for example, assumptions about the employment and wage premiums, the value of government revenue or the tax rate). The lower and upper bounds of the ranges are estimated as 95 per cent confidence intervals (table D.8). Since some assumptions were not able to be tested through sensitivity analysis, the confidence intervals are likely to underestimate the true uncertainty. The change in gross payments to labour is projected to lie between $1.3 billion and $5.6 billion in the realised simulations and between $2.2 billion and $11.5 billion in the prospective simulations. Projected net social benefits could be positive or negative, depending on the set of modelling assumptions used. The 95 per cent confidence intervals for realised and prospective effects include the possibility that net social benefits and productivity effects are zero. However, a small positive outcome is most likely. Table D.8
Sensitivity analysis of estimated effects of the Victorian entitlementa Victorian young learners (aged 15 to 24) Realised
Prospective
Lowerb
Upperb
Lowerb
Upperb
-0.06 0.05 1 257 -109
0.29 0.36 5 579 2 613
-0.10 0.08 2 185 -152
0.64 0.60 11 480 5 554
Productivity (%) Participation (%) Gross payments to labour ($m) Net social benefit ($m)
aTotal change in present value terms over 42 years. b ‘Lower’ and ‘upper’ define the bounds of the 95 per cent confidence interval. Source: Productivity Commission estimates.
D.2 Possible productivity effects of contestability reforms in Victoria As discussed in chapter 3, the introduction of contestability has allowed private providers to provide some of the places that are funded by the government. There is
92
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
a debate about the likely effects of contestability, especially in terms of its potential effect on the quality of the qualifications produced. In some ‘mixed markets’ where services are supplied by private and public providers, contestability has been introduced under the assumption that private providers can produce better outcomes for the same cost. This would be consistent with conditions in which access to private providers would expand the choices available to students, for example in some regional areas, where there are few providers (‘thin markets’). Others argue that contestability could decrease the quality of the qualifications that the system delivers, as private providers might reduce the quality of teaching to reduce costs. Recent reports (for example, ESC 2011a) indicate that quality is an issue with some providers. The recent inception of the Australian Skills Quality Authority is one initiative designed to improve quality standards. This, along with increased competition and improved information are likely to produce quality over time. Two scenarios are modelled:
improved quality is represented by a 5 per cent increase in the wage and probability of employment premiums that a student can expect from a VET qualification, relative to the baseline, and
reduced quality is represented by a 5 per cent decrease in the corresponding wage and employment probability premiums, relative to the baseline.
The effects described are annual and are assumed to apply to the outcomes of any year during which the contestability regime is in place. Effects on education and occupational outcomes If contestability results in the VET system producing higher quality outcomes, VET qualifications attract higher returns than in the baseline and the share of VET qualifications in the distribution of qualifications increases (table D.9). The converse occurs with a reduction in the quality of the outcomes of VET qualifications relative to the baseline.
YOUNG LEARNERS
93
Table D.9
Change in the distribution of highest educational attainment associated with contestability Per cent, Victorian young learners (aged 15 to 24); annual effects; changes from the baseline Improved quality effect
Reduced quality effect
-0.40 -2.93 1.69 0.27 ..
0.36 3.09 -1.72 -2.71 0.69
Year 11 Year 12 Cert. III/IV Diploma Degree .. less than 0.005 Source: Productivity Commission estimates.
The changes in educational attainment affect the number of people employed in different occupations (table D.10). If contestability improves quality and increases the number of people with VET qualifications, there is an increase in employment in occupations related to VET and a decrease elsewhere. Quality improvements are projected to increase the number of people employed in community and personal services by 0.88 per cent and reduce the number of people employed in professional occupations by 0.02 per cent. Quality reductions are estimated to have larger effects in the opposite direction. Table D.10 Changes in occupational outcomes associated with contestability Per cent, Victorian young learners (aged 15 to 24); annual effects; changes from the baseline
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers
Positive quality effect
Negative quality effect
0.12 0.02 0.36 0.88 -0.37 -0.15 -0.12 -0.27
-0.12 0.27 -0.41 -1.20 0.25 -0.02 0.06 0.21
Source: Productivity Commission estimates.
Thus positive effects on quality can enhance the projected effects of Victorian entitlement policies, and negative effects can detract from the projected effects of the entitlement policies.
94
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Effects on participation and productivity Workforce participation is estimated to rise by 0.063 per cent and productivity by 0.036 per cent due to assumed quality improvements (table D.11). Productivity increases because the individual productivity component is positive and larger than the increased participation component (box D.2). Assumed reductions in the quality of VET outcomes are estimated to decrease workforce participation by 0.043 per cent and productivity by 0.003 per cent. Table D.11 Changes in participation and productivity associated with contestability Per cent, Victorian young learners (aged 15 to 24); annual effects; changes from baseline
Participation Productivity Individual productivity component Increased participation component
Positive quality effect
Negative quality effect
0.063 0.036 0.043 -0.007
-0.043 -0.003 -0.012 0.009
Source: Productivity Commission estimates.
Effects on net social benefit The entitlement is projected to produce a $966 million increase in net social benefit (in present value terms) from assumed improvements in the quality of VET, and a $0.7 billion reduction in net social benefit from assumed reductions in quality (table D.12). Net social benefit is calculated from changes in labour income before taxes. For example, labour income before taxes is estimated to increase by around $1.4 billion in the quality improvement scenario. An estimate of the costs of non-market activity (for example, leisure and looking after family) foregone due to the increase in workforce participation ($0.3 billion), is subtracted from this figure, as is the change in the money costs of education. Adjustments have also been made to capture the value of government revenue and other benefits of education — for example, the consumption benefits that might accrue to individuals in the course of their study.
YOUNG LEARNERS
95
Table D.12 Possible effects of a contestability regime on economic aggregatesa $ million, Victorian young learners (aged 15 to 24); effects for 1 year of contestability regime Outcomes Gross payments to labour less value of non-market activity forgone by working less money cost of education plus adjustment to capture the value of government revenue plus external benefitsb plus other benefitsc Net social benefit
Positive quality Negative quality effect effect 1 439 343 -8 133
-735 -234 129 -89
0
0
-242 996
-3 -723
aTotal change in present value terms over 42 years. bZero by assumption. cBenefits of education not explicitly modelled, but estimated through model calibration. Source: Productivity Commission estimates.
Partial fiscal effects The assumed improvements in the quality of VET are estimated to result in an increase in net government expenditure of $0.5 billion, while the assumed reductions in quality are estimated to increase net government revenue by $0.4 billion. These effects depend on changes in taxes and education expenditure. The model accounts for the additional revenue associated with increased collection of personal income tax over time and for the initial expenditures associated with the initiative. It does not account for the effects of increased wages bills on payroll tax revenue or of additional GST revenue from increased spending.6
D.3 Prospective effects of the South Australian VET entitlement Additional places anticipated in South Australia as a consequence of the COAG VET reforms need to be translated into qualification completions before the impacts of policy changes can be assessed. As discussed in chapter 3, the Commission has assumed that the policy initiative in South Australia leads to an increase in training places of 46 796 in total between 1 July 2012 and 31 December 2014.
6 See section D.1 for a more detailed explanation. 96
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
A number of assumptions are used to estimate the qualification outcomes of the South Australian entitlement policy. Key among these are that:7
the ratio of enrolments to places is 1.5 (based on a comparison of NPAPPP qualification commencement and enrolment undertakings in 2010-11 and 2011-12)
the ratio of enrolments to students is 1.3 (based on a comparison of Victorian data on enrolments by qualification level, with National Centre for Vocational Education Research (NCVER) data for Victoria on students by qualification level)
young learners (the focus of the ELMO model) account for 50.8 per cent of students
qualification completion rates for young learners are 36.5 per cent for Diplomas and Advanced Diplomas, 35.3 per cent for Certificates III and IV and 28.8 per cent for courses at and below Certificate II (based on Karmel 2011b)
the entitlement leads to a small increase in the percentage of students studying at or below Certificate II level.
The conversion of places into additional qualification completions is summarised in table D.13. Table D.13 Estimates of additional VET qualifications due to the South Australian entitlementa South Australian young learners (aged 15 to 24) Prospective increase 1 July 2012 to 31 December 2014 46 796 70 057 53 890 27 626 6 895 326
Additional VET places Converted to enrolments (ENR) ENR converted to students Share of learners aged 15–24 Completions Diploma and Ad. Diplomab
1 771
Certificates III and IVb Certificates I/II and prevocational courses
4 798
a Rounding accounts for small differences between presented figures and the estimates that might be derived from the figures in this section. For example, the ratio of enrolments to places is 1.4971. Use of this figure instead of 1.5 produces the estimate for enrolments of 70 057 (whereas 1.5 returns a figure of 70 194). b Figures used as targets for the ELMO model. Source: Productivity Commission estimates.
7 A full list of assumptions is presented in a spreadsheet available on request from the study team. YOUNG LEARNERS
97
The prospective increase in places is estimated to translate into an additional 27 383 government-funded students aged 15 to 24 between 1 July 2012 and 31 December 2014. This represents significant growth in student numbers, and completions. In 2014, for example, it is estimated that there will be at least 25 per cent more government-funded students in training than in 2011, and an additional 2800 young people will complete a qualification — many, however, at a level below Certificate III. Effects on education and occupational outcomes The mechanisms at work in the prospective South Australian entitlement policy are the same as those described in the Victorian case above. Results differ because of differences in the way the policy is implemented (which leads to the inputs in table D.14) and differences between the Victorian and South Australian populations. The increase in qualification completions due to the introduction of an entitlement in South Australia reflects the associated reduction in the costs of VET courses to students relative to alternatives (table D.14). The percentage of the young learner cohort with Year 11 and below as their highest level of education is projected to fall by 0.7 per cent; the proportion with Year 12, by 7 per cent. Table D.14 Change in the distribution of highest educational attainment due to South Australian entitlement Per cent, South Australian young learners (aged 15 to 24)
Year 11 and below Year 12 Cert. III/IV Diploma Degree Total
Base
Policy
Change
21.99 11.12 27.90 10.31 28.68 100.00
21.84 10.35 28.68 10.46 28.68 100.00
-0.70 -6.92 2.79 1.41 ..
.. less than 0.005 Source: Productivity Commission estimates.
Prospective changes in educational attainment are estimated to increase employment in occupations more commonly entered by people with certificate and diploma qualifications, such as community and personal services (1.3 per cent) and technicians and trades (0.4 per cent) (table D.15). Employment in occupations more commonly entered by people with qualifications other than at a Certificate III 98
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
through to Advanced Diploma level is estimated to decrease — for example, for machinery operators and drivers by 0.5 per cent. Table D.15 Occupational outcomes due to South Australian entitlement policies South Australian young learners (aged 15 to 24)
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers Total
Base
Policy
Change
no.
no.
%
21 521 38 550 21 892 12 773 23 693 9 847 6 847 11 603 146 727
21 522 38 557 21 970 12 939 23 629 9 868 6 811 11 547 146 843
0.00 0.02 0.35 1.30 -0.27 0.21 -0.52 -0.49 0.08
Source: Productivity Commission estimates.
Effects on participation and productivity Workforce participation by the South Australian young learners cohort is projected to rise by 0.08 per cent, and productivity to rise by 0.03 per cent, as a consequence of the entitlement policies (table D.16). Workforce participation increases because there is a positive relationship between education and participation, and the entitlement is estimated to increase general education levels. The contribution to aggregate productivity of individual productivity is positive and participation is negative (box D.2). As in the Victorian case, the increase in participation brings to the labour market some individuals with lower abilities than the incumbents. This is offset by higher individual productivity. Table D.16 Participation and productivity changes due the South Australian entitlement policies Per cent, South Australian young learners (aged 15 to 24) Change Participation Productivity Individual productivity component Increased participation component
0.079 0.026 0.037 -0.011
Source: Productivity Commission estimates.
YOUNG LEARNERS
99
Effects on net social benefit South Australia’s net social benefit is estimated to increase by about $160 million in net present value terms as a consequence of the prospective changes in young learners’ educational attainment (table D.17). Net social benefit is calculated from changes in labour income before taxes. Overall, gross labour income is estimated to increase by about $400 million as a result of the entitlement. An estimate of the costs of non-market activity foregone (for example, leisure and looking after family) due to the increase in workforce participation ($100 million), is subtracted from this figure, as is the change in the money costs of education. Table D.17 Estimated effects of South Australian entitlement policies $ million, South Australian young learners (aged 15 to 24) Outcomes Gross payments to labour less value of non-market activity forgone by working less money cost of education plus adjustment to capture the value of government revenue plus external benefitsb plus other benefitsc Net social benefit
Changea 423 122 -4 -3 0 -142 160
aTotal change in present value terms over 42 years. bZero by assumption. cBenefits of education not explicitly modelled, but estimated through model calibration. Source: Productivity Commission estimates.
Partial fiscal effects The South Australian entitlement project entails additional government expenditure over four years. To the extent that it increases wages and participation it also increases future tax revenue, assuming current income tax settings. The net effect is estimated to increase net government revenue by $13 million. This depends on changes in tax receipts and additional education expenditure associated with the initiative. The model accounts for the additional revenue associated with increased collection of personal income tax over time and for the initial expenditures associated with the initiative. It does not account for the effects of increased wages bills on payroll tax revenue or additional GST revenue that would follow from increased spending.8 8 See section D.1 for a more detailed explanation. 100
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Sensitivity analysis Numerous modelling assumptions have been made regarding parameter values in estimating the effects of the South Australian entitlement. Altering these assumptions alters the estimated effects. The sensitivity analysis below illustrates how the estimated effects change in response to changing many of the sets of assumptions that underlie the model (for example, assumptions about the employment and productivity premiums, the value of government revenue or the tax rate). The lower and upper bounds of the ranges are estimated as 95 per cent confidence intervals (table D.18). Since some assumptions were not able to be tested through sensitivity analysis, the confidence intervals are likely to underestimate the true uncertainty. The change in gross payments to labour is estimated to lie between $0.2 billion and $1.2 billion. The sensitivity analysis shows that estimated net social benefit is likely to be positive. The confidence interval for net social benefit was $56 million to $580 million over the 42 years that the policy is assumed to affect the cohort. Table D.18 Sensitivity analysis of estimated effects of the South Australian entitlementa South Australian young learners (aged 15 to 24)
Productivity (%) Participation (%) Gross payments to labour ($m) Net social benefit ($m)
Lowerb
Upperb
-0.02 0.03 232 56
0.26 0.18 1 186 580
aTotal change in present value terms over 42 years. b ‘Lower’ and ‘upper’ define the bounds of the 95 per cent confidence interval. Source: Productivity Commission estimates.
D.4
Realised effects of the NPAPPP
Effects on education and occupational outcomes The mechanisms at work in the NPAPPP are the same as those described in the cases above, with one main difference — the places associated with the NPAPPP are restricted to areas of ‘need’. The ELMO model abstracts from this complication, spreading additional places across the VET system according to their unconstrained returns. The main results presented in this section relate to the main version of the model and are likely to overestimate the effects of the NPAPPP. YOUNG LEARNERS
101
The increase in qualification completions due to the NPAPPP reflects the associated reduction in the costs of VET courses to students relative to alternatives (table D.19).9 It is estimated that the percentage of the young learner cohort with Year 11 and below as their highest level of education would fall by 1.5 per cent, and with Year 12, by 16 per cent. Table D.19 Realised change in the distribution of highest educational attainment due to the NPAPPP Per cent, Australian young learners (aged 15 to 24), excluding Victoria
Year 11 and below Year 12 Cert. III/IV Diploma Degree Total
Base
Policy
Change
20.31 11.15 25.37 10.80 32.37 100.00
20.01 9.37 27.01 11.24 32.37 100.00
-1.49 -15.99 6.48 4.09 ..
.. less than 0.005 Source: Productivity Commission estimates.
Prospective changes in educational attainment are estimated to increase employment in occupations more commonly entered by people with certificate and diploma qualifications, such as community and personal services (2.7 per cent) and technicians and trades (0.5 per cent) (table D.20). Employment in occupations more commonly entered by people without VET qualifications is estimated to decrease — for example, for machinery operators and drivers, by nearly 1 per cent. Table D.20 Realised occupational outcomes from the NPAPPP Australian young learners (aged 15 to 24), excluding Victoria
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers
Base
Policy
Change
no.
no.
%
243 684 439 676 197 078 137 395 258 413 104 332 71 753 93 953
244 721 440 572 198 143 141 059 256 617 103 847 71 076 92 924
0.43 0.20 0.54 2.67 -0.69 -0.47 -0.94 -1.10
Source: Productivity Commission estimates.
9 Reflecting the experience in New South Wales (reported in Allen Consulting 2010b), a completion rate of 55 per cent was adopted in deriving these estimates. 102
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Effects on participation and productivity Workforce participation by the Australian young learners cohort (outside Victoria) is projected to rise by 0.17 per cent, and productivity to rise by 0.04 per cent, as a consequence of the realised changes (table D.21). Workforce participation increases because there is a positive relationship between education and participation, and the entitlement is estimated to increase general education levels. The contribution to aggregate productivity of individual productivity is positive and participation is negative (box D.2). As in the Victorian case, the increase in participation brings to the labour market some individuals with lower abilities than the incumbents. This is offset by higher individual productivity. Table D.21 Realised participation and productivity effects due to the NPAPPP Per cent, Australian young learners (aged 15 to 24), excluding Victoria Change Participation Productivity Wage component Participation component
0.173 0.037 0.063 -0.026
Source: Productivity Commission estimates.
Effects on net social benefit Net social benefit is estimated to increase by around $3.3 billion in net present value terms as a consequence of the potential changes in young learners’ educational attainment (table D.22). Net social benefit is calculated from changes in labour income before taxes. Overall, gross labour income is estimated to increase by around $8.9 billion as a result of the entitlement. An estimate of the costs of non-market activity foregone (for example, leisure and looking after family) due to the increase in workforce participation ($2.8 billion), is subtracted from this figure, as is the change in the money costs of education.
YOUNG LEARNERS
103
Table D.22 Realised estimated effects of the NPAPPP $ million, Australian young learners (aged 15 to 24), excluding Victoria Outcomes
Changea
Gross payments to labour less value of non-market activity forgone by working less money cost of education plus adjustment to capture the value of government revenue plus external benefitsb plus other benefitsc Net social benefit
8 936 2 808 -81 -91 0 -2 797 3 321
aTotal change in present value terms over 42 years. bZero by assumption. cBenefits of education not explicitly modelled, but estimated through model calibration. Source: Productivity Commission estimates.
As mentioned above, the reported results could be overestimates. The ELMO model was used to assess the likely extent of this (box D.4). Box D.4
Effect of restricting VET places to areas of ‘need’
The ELMO model was modified to estimate the effects of restricting places to areas of ‘need’ for sensitivity analysis. These areas were assumed to constitute about 70 per cent of the VET system. The estimated net social benefit is $3.1 billion. This is about $0.2 billion less than applying the shock to the entire VET system (as reported in table D.22). This is because some of the new subsidised places are taken up by people already in the VET system, so there is less movement into the VET system (and fewer distortions are alleviated). The partial subsidy also distorts education investment decisions within the VET system.
Partial fiscal effects The realised NPAPPP project entails additional government expenditure over four years. To the extent that it increases wages and participation it also increases future tax revenue, assuming current income tax settings. The net effect is estimated to result in net government revenue of around $381 million. This results depends on changes in taxes and education expenditure. The model accounts for the additional revenue associated with increased collection of personal income tax over time and for the initial expenditures associated with the initiative. It does not account for the effects of increased wages bills on payroll tax revenue or of additional GST revenue from increased spending.10 10 See section D.1 for a more detailed explanation. 104
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Sensitivity analysis Numerous modelling assumptions have been made regarding parameter values in estimating the effects of the NPAPPP. Altering these assumptions alters the estimated effects. The sensitivity analysis below illustrates how the estimated effects change in response to changing many of the sets of assumptions that underlie the model (for example, assumptions about the employment and productivity premiums, the value of government revenue or the tax rate). The lower and upper bounds of the ranges are estimated as 95 per cent confidence intervals (table D.23). Since some assumptions were not able to be tested through sensitivity analysis, the confidence intervals are likely to underestimate the true uncertainty. The change in gross payments to labour is estimated to lie between $1 billion and $24.7 billion. The sensitivity analysis shows that estimated net social benefit could be positive or negative, depending on the modelling assumptions. The confidence interval for net social benefit was -$0.4 billion to $12.5 billion. Table D.23 Sensitivity analysis of estimated effects of the NPAPPPa Australian young learners (aged 15 to 24), excluding Victoria
Productivity (%) Participation (%) Gross payments to labour ($m) Net social benefit ($m)
Lowerb
Upperb
-0.02 0.04 1004 -376
0.18 0.41 24 660 12 514
aTotal change in present value terms over 42 years. b ‘Lower’ and ‘upper’ define the bounds of the 95 per cent confidence interval. Source: Productivity Commission estimates.
D.5
Potential effects of reforms to 2020
The scenario underlying the potential effects of reform is based on the number of qualifications required to achieve the proportion of 20–64 year olds without a Certificate III or above (23.6 per cent) by 2020, once all the realised and prospective effects have been achieved (figure 3.4). For simplicity, the numbers of qualifications that are required to achieve the target are assumed to be produced in the seven years to 2020, Australia-wide (table D.24).
YOUNG LEARNERS
105
Table D.24 Additional qualifications required per annum to meet COAG targets, 2013–2019a Number, Australian young learners (aged 15 to 24)
Young learners
Diplomab
Certs. III/IV
Total
8 800
46 600
55 400
a These estimates are for students increasing their highest qualification. b Includes diplomas and advanced diplomas. Source: Productivity Commission estimates.
Effects on education and occupational outcomes The mechanisms at work in the potential policy scenario are the same as those described in the Victorian realised and prospective and South Australian prospective cases above. Results differ because of differences in the way the policies are implemented. In particular, the increase in the number of subsidised places is much larger, in absolute terms and relative to the size of the affected populations, than in the cases above. The increase in qualification completions due to the potential policy scenario reflects the associated reduction in the costs of VET courses to students relative to alternatives (table D.25). It is estimated that the percentage of the young learner cohort with Year 11 and below as their highest level of education would fall by 10 per cent, and with Year 12, by 88 per cent. Table D.25 Potential change in the distribution of highest educational attainment in the Australian population Per cent, Australian young learners (aged 15 to 24)
Year 11 and below Year 12 Cert. III/IV Diploma Degree Total
Base
Policy
Change
19.73 11.58 25.27 10.59 32.82 100.00
17.67 1.38 35.65 12.48 32.82 100.00
-10.41 -88.09 41.06 17.77 ..
.. less than 0.005 Source: Productivity Commission estimates.
Prospective changes in educational attainment are estimated to increase employment in occupations more commonly entered by people with certificate and diploma qualifications, such as community and personal services (19 per cent) and 106
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
technicians and trades (2 per cent) (table D.26). Employment in occupations more commonly entered by people without VET qualifications is estimated to decrease — for example, for machinery operators and drivers by nearly 6.5 per cent and labourers by 7.7 per cent. Table D.26 Potential occupational outcomes from achievement of COAG’s VET targets Australian young learners (aged 15 to 24)
Managers Professionals Technicians and trades Community and personal services Clerical and administrative Sales Machinery operators and drivers Labourers
Base
Policy
Change
no.
no.
%
339 450 600 914 250 958 186 055 334 930 140 003 91 334 120 938
350 592 606 362 256 632 220 482 318 893 136 095 85 489 111 601
3.28 0.91 2.26 18.50 -4.79 -2.79 -6.40 -7.72
Source: Productivity Commission estimates.
Effects on participation and productivity Workforce participation by the Australian young learners cohort is projected to rise by 1 per cent, and productivity to rise by 0.08 per cent, as a consequence of the prospective changes (table D.27). Workforce participation increases because there is a positive relationship between education and participation, and the policy scenario is estimated to increase general education levels. The contribution to aggregate productivity of individual productivity is positive and participation is negative (see box D.2). As in the Victorian case, the increase in participation brings to the labour market some individuals with lower abilities than the incumbents. This is offset by higher individual productivity. Table D.27 Potential participation and productivity changes from achievement of COAG’s VET targets Per cent, Australian young learners (aged 15 to 24), excluding Victoria Change Participation Productivity Individual productivity component Increased participation component
1.044 0.082 0.270 -0.189
Source: Productivity Commission estimates. YOUNG LEARNERS
107
Effects on net social benefit Net social benefit is estimated to increase by around $18 billion in net present value terms as a consequence of the potential changes in young learners’ educational attainment (table D.28). Net social benefit is calculated from changes in labour income before taxes. Overall, gross labour income is estimated to increase by around $64 billion as a result of the policy scenario. An estimate of the costs of non-market activity foregone (for example, leisure and looking after family) due to the increase in workforce participation ($22.7 billion), is subtracted from this figure, as is the change in the money costs of education. Table D.28 Potential estimated effects of achievement of COAG’s VET targets $ million, Australian young learners (aged 15 to 24) Outcomes Gross payments to labour less value of non-market activity forgone by working less money cost of education plus adjustment to capture the value of government revenue plus external benefitsb plus other benefitsc Net social benefit
Changea 63 987 22 662 -672 -2 592 0 -21 358 18 047
aTotal change in present value terms over 42 years. bZero by assumption. cBenefits of education not explicitly modelled, but estimated through model calibration. Source: Productivity Commission estimates.
Partial fiscal effects The potential increase in attainment associated with achievement of the COAG VET targets entails additional government expenditure. To the extent that it increases wages and participation it also increases future tax revenue, assuming current income tax settings. The net effect is estimated to result in net government revenue of around $10.8 billion. This result depends on changes in taxes and education expenditure. The model accounts for the additional revenue associated with increased collection of personal income tax over time and for the initial expenditures associated with the initiative. It does not account for the effects of increased wages on payroll tax revenue or of additional GST revenue of increased spending.11 11 See section D.1 for a more detailed explanation. 108
IMPACTS OF COAG REFORMS – VET AND TRANSITIONS
Sensitivity analysis Numerous modelling assumptions have been made regarding parameter values in estimating the effects of the potential policy scenario. Altering these assumptions alters the estimated effects. The sensitivity analysis below illustrates how the estimated effects change in response to changing many of the sets of assumptions that underlie the model (for example, assumptions about the employment and productivity premiums, the value of government revenue or the tax rate). The lower and upper bounds of the ranges are estimated as 95 per cent confidence intervals (table D.29). Since some assumptions were not able to be tested through sensitivity analysis, the confidence intervals are likely to underestimate the true uncertainty. The change in gross payments to labour is estimated to lie between $3.5 billion and $156 billion. The sensitivity analysis shows that estimated net social benefit could be positive or negative, depending on the modelling assumptions. The confidence interval for net social benefit was -$3 billion to $68 billion. Table D.29 Sensitivity analysis of estimated effects of achievement of COAG’s VET targetsa Australian young learners (aged 15 to 24)
Productivity (%) Participation (%) Gross payments to labour ($m) Net social benefit ($m)
Lowerb
Upperb
-0.11 0.34 3 444 -2 928
0.79 1.83 155 646 68 414
aTotal change in present value terms over 42 years. b ‘Lower’ and ‘upper’ define the bounds of the 95 per cent confidence interval. Source: Productivity Commission estimates.
YOUNG LEARNERS
109