Understanding the responsiveness of ‘older’ workers — those aged 25-54 years — to government subsidies for up-skilling is important in developing policies that can support people in obtaining skills later in life. The particular focus for this report is the first round of Victorian demand-driven reforms, known as the Victorian Training Guarantee (VTG). The VTG reforms were introduced to create a more responsive training market and were implemented between July 2009 and January 2011. How enrolment and course choice responses differ, especially between employed and unemployed individuals, is a key focus of the report. Reforms introduced in Victoria post-2012 are not part of this analysis.
About the research
In 2008, the Council of Australian Governments (COAG) initiated the National Agreement for Skills and Workforce Development, a component of which focused on reforming the training market to be more learner-driven and responsive to the labour market. This resulted in the implementation of national training entitlements for government-subsidised training places. An underlying premise of entitlement models is that, by giving the student a choice in where they train and the capacity to choose their course, they will more likely choose courses that will benefit them economically and, therefore, the wider labour market.
This research looks at the responsiveness of individuals — aged 25 to 54 — to government subsidies designed to support upskilling through the Victorian Training Guarantee (VTG). The timeframe for analysis was 1 Jan 2011 — 30 June 2012, reflecting the first 18 months that the reforms introduced through the VTG were fully implemented for the target age band. This timeframe is also prior to the reforms introduced in July 2012 that focussed on targeting subsidies to influence course choices, and the introduction of Skills First on 1 January 2017, which superseded the Victorian training and TAFE system. Therefore, while this research does not consider reforms post July 2012, the findings are still relevant in the current environment as the criteria for ‘older’ learners remains unchanged, namely government-subsidised training places available to individuals aged 20 years or older that are looking to upskill.
This research builds on that supported previously by NCVER, which explored the early impacts of the VTG on enrolment numbers and graduate outcomes among 15 to 19-year-olds (Leung et al. 2014).A particular focus here is on the impact of the subsidies on enrolments, as well as the alignment of course choices with labour market needs. The research explored how enrolment and course choice responses varied by age, gender and across disadvantaged groups in the community, including the unemployed, people with little formal education, people from non-English speaking backgrounds, people from low socioeconomic areas, and people with disabilities.
- ‘Older’ learners (25 to 54 years) responded positively to the upskilling entitlement of the Victorian Training Guarantee, with the uptake of vocational education and training (VET) by this group estimated to have increased by 4.2 percentage points by comparison with the rest of Australia between January 2011 and June 2012.
- The evidence indicates a significant improvement in the match between course choices and the officially recognised skills in demand, with enrolments skewed towards courses with relatively high expected wages upon completion.
- Importantly, these outcomes were also applicable to people from more disadvantaged backgrounds, such as the unemployed, people from non-English speaking backgrounds and people with disabilities. Improved alignment of course choice with the skills in demand should impact on the subsequent living standards of people from such backgrounds.
- The results underline the capacity of working-age individuals, including those who are more disadvantaged, to make sound course choices in relation to jobs if given the opportunity and access to relevant information.
Dr Craig Fowler
Managing Director, NCVER
Public subsidies for vocational education and training (VET) in Australia have historically been provided directly to select (mostly public) education institutions (providers) on the basis of government projections of skill demand. This approach defined the number and type of VET courses that benefited from a public subsidy, and these were predominantly allocated on a first-come-first-served basis. In 2008 Australia adopted a national reform agenda in which centralised VET funding was replaced by entitlement schemes.
The fundamental premise underlying the 2008 reform agenda is that students, if given the option, will tend to choose VET courses to suit their own economic interests, to the benefit of the wider labour market. This premise was untested when the reform agenda was introduced, and the sparse international evidence that did exist gave cause for some pessimism (see Levin 1991; Ladd 2002; Hastings & Weinstein 2008). As such, the reform agenda represented a significant gamble. This study presents some of the first statistical evidence on the effects of associated reforms.
The research examines the enrolment responses of working-age people aged 25—54 years to an entitlement introduced in Victoria from 1 January 2011 to 30 June 2012 — the Victorian Training Guarantee (VTG)1. The age band of the analysis complements companion studies by Leung et al. (2014) and McVicar and Polidano (2015), which focused on similar outcomes for individuals aged 15—19 years. Given that those aged between 25 and 54 years are likely to have had greater labour market experience and may enter VET for different reasons, the outcomes for this group may be different.
The focus is on the Victorian reforms because they were more ambitious than reforms introduced in other Australian states and territories. The Victorian reforms to VET funding introduced a genuine entitlement, one which substantially increased rights to and choice of publicly subsidised courses for working-age individuals. The timeframe of the analysis reflects the time period the Victorian reforms were fully implemented for the targeted age band but is prior to subsequent reforms, introduced from 1 July 2012, which sought (through the targeting of subsidies) to reshape course choices.
The study considers the extent to which the new education entitlements were taken up and the alignment between course choice and the needs of the domestic labour market. Specifically, we address the following questions:
- What effect did the introduction of the Victorian Training Guarantee have on VET enrolment numbers among those aged 25—54 years?
- What effect did the Victorian Training Guarantee have on course choice? Did it improve the alignment of enrolments with the measures of skill demand available at the time of enrolment?
- How did enrolments and course choices vary by age and gender and across key disadvantaged groups in the community, including the unemployed, people with little formal education, people from non-English speaking backgrounds, people from low socioeconomic areas, and people with disabilities?
An estimation of the effects of the Victorian Training Guarantee were calculated using a difference-in-differences (DiD) approach. This method is a standard quantitative approach for estimating the effects of real-world policy changes. The fundamental innovation of the approach is to disentangle the effects of policy reform from contemporaneous changes in the wider economic environment (for example, the Global Financial Crisis) by comparing the differences observed for a ‘treatment’ population affected by the reforms with those observed for a ‘control’ population which was not. In the current case, the approach compares the variation observed for the Victorian population before and after the policy change with the variation observed for the population of other Australian states during the same period. The study is based upon administrative data drawn from the National VET Provider Collection, augmented by linking information from alternative data sources, including expected graduate earnings estimates from the NCVER National Student Outcomes Survey, National Skill Shortage information from the federal Department of Employment, and regional socioeconomic and population estimates from the Australian Bureau of Statistics (ABS).
Between January 2011 and June 2012, we estimate that the introduction of the Victorian entitlement increased the rate of VET participation among 25 to 54-year-olds by 4.2 percentage points. This represents a two-thirds increase, relative to the 6.3% participation rate reported in 2008 for this age group. A disaggregated analysis indicates that the increase in participation was distributed widely across the population, including a 36-percentage-point increase among unemployed people, a 12-percentage-point increase among people with disabilities and a 13-percentage-point increase among people with only low-level (certificate I and II) VET qualifications. However, we stress that these effects may overestimate the true enrolment effects because for the period the study was undertaken, the National VET Provider Collection did not account for private fee-for-service enrolments, which are likely to have decreased under the Victorian entitlement.2 From 2014 the National VET Provider Collection commenced collecting fee-for-service activity from private providers, in addition to government-funded training activity previously collected, thus providing Total VET Activity of the Australian training market.
The introduction of the entitlement in Victoria was also found to lead to a significant improvement in the match between course choices and officially recognised skills in demand and it skewed enrolment towards courses with a relatively high expected graduate wage. To the extent that graduate wages reflect the value of graduate skills, the latter result is further evidence of an improvement in the responsiveness of enrolments to skill needs. Overall, the greater freedom of choice afforded by the Victorian entitlements is estimated to change the course choice mix in a way that increases the proportion of enrolments in skill shortage areas by four percentage points (a 50% increase, relative to 2008 figures), and increase expected graduate wages by 2.4%. These estimates are close to similar to those reported by McVicar and Polidano (2015) for 15 to 19-year-olds3, which in itself is a somewhat surprising result, given the different training incentives and life experience of the respective population subgroups. In contrast to the study by McVicar and Polidano, however, the current study finds no evidence of offsetting shifts in the types of courses offered to people aged 25—54 years by private providers. We interpret this result as suggesting that working-age individuals were less swayed than their younger contemporaries by the increased availability of privately run courses with relatively high consumption and low labour market benefits (including health and fitness courses).
Across the groups identified in this study, we find no reason to believe that people from disadvantaged backgrounds respond in a less positive way than the remainder of the population. We find improvements in course choice alignment with skill demand for all groups. Moreover, improvements were greater for the unemployed than the employed, greater for those from a non-English speaking background than those who speak English at home, greater for those living in a low socioeconomic status (SES) area than in more advantaged areas and greater for Aboriginal and Torres Strait Islanders than for non-Indigenous people. For individuals with disabilities, we find improvements that are no smaller than for those without disabilities; and for those without a secondary qualification we find greater improvements in expected wage returns, but lower improvements in the match between enrolments and skill shortage areas. In the last case, our results are primarily attributable to the fact that people without a secondary qualification do not tend to enrol in courses that lead to higher-skilled occupations, but rather in courses that are higher paid, especially in construction, logistics and sales. These results suggest that disadvantaged people aged 25 to 54 years do seek out courses with good expected labour market outcomes, outcomes likely to have a particularly profound bearing on subsequent living standards.
The results from this study underline the capacity of people aged 25 to 54 years, including people from disadvantaged backgrounds, to choose VET courses that are better aligned with the prevailing needs of the domestic labour market than was the practice under previous supply-driven funding models. These results are more impressive when taking into account the large number of available courses and the paucity of course graduate outcome information available to support decision-making.
The current results contrast sharply with the relatively negative outcomes reported for large-scale entitlement schemes for private secondary schooling (Hsieh & Urquiola 2006), but the current study is ill-equipped to consider the important question of why this is the case. In any event, there is a clear gap between the educational decisions made by a parent for their child and those made by an older adolescent or working-age adult for themselves. Moreover, vocational education and training is, by design, a form of education closely aligned with the technical skills requirements of the workforce. The results of the current study highlight the choices made by more mature individuals in exercising an ‘entitlement’ to improve the responsiveness of VET to changing skill demands.
For policy-makers, evidence that disadvantaged people seek out courses with good expected labour market outcomes supports current efforts to improve available course outcome information, including employment information, through the MySkills website. However, an area for further improvement may be in the provision of more disaggregated graduate employment information. Currently, graduate occupation information is limited to ANZSCO4 major group level (for example, Technician and trades workers, Managers, Labourers), which provides no indication to prospective students on the likelihood of their finding work after graduating in the occupation for which the course is designed to prepare them. Such information may be important, given that there is a weak match between the course target occupation and the occupations that course graduates attain (Karmel, Mlotkowski & Awodeyi 2008) and the likely importance of occupation choice in driving course choice.
While the provision of labour market information may be important in helping to support decision-making, governments should also be aware of the risks. In particular, there is a risk that people may systematically over-respond to skill demand signals, leading to graduate oversupply in areas where there are few supply-side barriers to entry. Governments can play a role here by monitoring year-on-year changes in course enrolments and by complementing the provision of graduate outcome information with longer-term skill projections.
1 As at 1 January 2017, the Victorian Training Guarantee was superseded by Skills First http://www.education.vic.gov.au/skillsfirst
2 We also attempted to cross-check these results against data from the Survey of Education and Work (SEW), which has information on participation in all VET courses, to quantify the magnitude of any bias. However, estimates from this survey did not provide any conclusive evidence on the magnitude of any bias because, despite collecting information on participation in all VET courses, the rates of VET participation were well below those estimated in the National VET Provider Collection.
3 McVicar and Polidano (2015) report a 3.9-percentage-point increase in the proportion of enrolments in skill shortages and 2.3% improvement in expected graduate wages for 15 to 19-year-olds.
4 ANZSCO = Australian and New Zealand Standard Classification of Occupations.