Microcredentials should not get FEE-HELP assistance

In the last two years the government – the current and former governments are indistinguishable on this point – has encouraged universities to offer ‘microcredentials’, which certify and sell smaller bodies of knowledge and skills than an AQF qualification.

Government support for microcredentials

Late last month Labor reintroduced a Coalition amendment to the Higher Education Support Act 2003 that would extend FEE-HELP income-contingent loans to microcredentials, although with the potentially limiting caveat of ‘that meet the requirements specified in the FEE‑HELP Guidelines.’

Last week they promulgated a legislative instrument for the Coalition’s ‘microcredential pilot’, which offers subsidies to Table A universities to develop microcredentials. According to the explanatory memorandum ‘the purpose of the program is to examine newer, shorter forms of industry focused learning aimed at supporting people to upskill and reskill in areas of national priority such as health, teaching, IT and engineering.’

The pilot does not seem designed to attract applications – universities would have to give away their IP and accept Job-ready Graduates Commonwealth and student contributions – but the bigger issue is FEE-HELP.

Do microcredentials require government intervention?

Contrary to the impression given by some microcredential discussions, people taking short courses to increase their skills is nothing new. The ABS has asked about structured learning not for a credential many times over decades, and always found it is the most common form of post-school education on a headcount basis. The latest ABS survey is no different. Short courses overtake credentialed education by a person’s late 20s, as the chart below shows.

Microcredentials add certification and perhaps standardisation to short courses, which might increase short course informational value in the labour market. But lack of these things has not stopped this market functioning on a large scale. Proxy indicators of employee suitability such as qualifications are important for young or career shifting job applicants, but for people already established in their careers observation – directly by employers, by reputation or referee report – is usually the main information source.

The market structure for short courses is also quite different to the more regulated credentials market. In the credentials market prospective students navigate a market they know little about. Regulatory institutions have been created to reduce their risk of spending years and large sums of money on a poor-quality degree. In short course markets employers are the main paying clients. Employers can use their detailed understanding of what skills their firms need, past experience of training providers and market power to avoid risks faced by individual students purchasing their own education.

There are, therefore, good reasons why neither students nor employers have asked for regulation of the short course market, and government hasn’t previously seen much need to get involved.

Are people missing out on short work-related courses?

There is some unrealised demand for non-formal (ABS-speak for structured but non-credentialed) education, peaking at just under 10 per cent of the population in the 35 to 44 years age group (chart below). However, the main reason for not pursuing additional education is time rather than money (next chart below). ‘Financial reasons’, which could include costs of time off work, peak at about about a quarter of this sub-sample in the 20-34 age groups. They are about 1 per cent of the whole sample. (Note: This section was updated due to an error in the proportions citing financial reasons for not undertaking formal learning.)

Why are financial reasons rarely an obstacle to short course training?

Several reasons explain why financial obstacles affect few people seeking training. The largest one is that employers as the main client usually pay, and so most people undertaking work-related training report no cost. For self-financed study the costs are usually low. Short course takers are also skewed to the upper income deciles, possibly reflecting occupations with professional development requirements. They can afford to pay.

HELP deals with a specific problem, the need for significant capital outlays on education at a stage of life when a large proportion of potential students lack access to that capital. Neither of these issues are significant in the short course/microcredential market. The outlays usually aren’t large and the students usually aren’t needy.

Government financial assistance double dipping

People undertaking work-related training connected to their current job can, if the government’s bill passes, both take out a FEE-HELP loan and get a tax deduction. For people on the top marginal tax rate a tax deduction nearly halves the cost to them and transfers it to the government via ‘tax expenditure’.

FEE-HELP is not costless for government either, with the risk of bad debt and in more normal times interest subsidies. Microcredential FEE-HELP costs are estimated at $5.6 million in ‘fiscal balance’ terms in the financial years to 2025-26. (We are also unhelpfully told that the ‘underlying cash impact of $0.0 million over the same period’, which is unsurprising as ‘underlying cash balance’ does not include most HELP transactions. For HELP measures ‘headline cash balance’ should be used. It makes me wonder whether policymakers are even told the cash-out-the-door implications of their HELP decisions.)

Paying for a course using FEE-HELP can generate cash flow for the student, because slow processing of HELP debt means that they can get their tax refund before they have to make any HELP repayments.

Distorting the short course market

Historically universities have not targeted the short course market, with continuing education in the $120-140 million annual revenue range in recent years, 0.4 per cent of university income in 2020.

I doubt existing non-higher education short course players will be overly worried by FEE-HELP in the higher education sector, as students paying their own fees are a small percentage of the market and few of them need loans. But from a competition policy perspective it is not appropriate to offer financial assistance to only one segment of the market.

ITECA noted a parallel problem with the government’s Microcredential Marketplace, which will showcase only the small university share of the market.

A Budget ‘saving’ for the government

The government’s microcredential policy never had sufficient analysis of either how big the problem was or the potential side-effects of the solution. It was also ‘siloed’ policymaking, making decisions for higher education providers without considering other individuals and organisations operating in the same space. For a government looking for ‘savings’, here is an easy one.

4 thoughts on “Microcredentials should not get FEE-HELP assistance

  1. Thanx for this.

    I would add that government support should be allocated to programs that have at least some educational and public value. However valuable micro credentials may be to employers and employees, most have minimal educational and public value.

    Furthermore, incorporating micro credentials within public quality assurance, accreditation and financial arrangements necessarily involves careful evaluation and checks on the allocation of public funds. This is not just a matter of government bureaucracy, but a necessary incidence of maintaining standards, probity and accountability (Moodie and Wheelahan, 2022). This seems too cumbersome for what is presented as a responsive and inexpensive credential.

    Moodie, Gavin and Wheelahan, Leesa (2022) Credentialing micro-credentials. Journal of Teaching and Learning for Graduate Employability, 12(1), 58-71, https://ojs.deakin.edu.au/index.php/jtlge/article/view/1564

    Like

  2. Thank you for the analysis and comments. No one, however, has mentioned the stackability of micros towards a full award, such as Graduate Certificate, and the transferability (potential credit recognition) of micros via the new framework between institutions – surely this is the whole point? I thoroughly enjoyed the micro I did this year through an Australian university (the class was over-subscribed) and would consider taking more to build skills and towards a fuller qualification. I was also in a privileged position where I could pay. The learning has directly benefitted me in my full-time job – and no, my employer, despite being a university itself, will not cover the cost.

    Like

    • Some argue that the main purpose of micro credentials is to provide short term credentials for what turn out to be short term jobs: gig credentials for the gig economy (Wheelahan and Moodie, 2021).

      Micro credentials are stackable into substantial credentials only by incorporating strong sequencing material and principles, which compromises their purported focus and flexibility.

      Leesa Wheelahan and I (2022) explained the incompatibility of micro credentials as they are commonly promoted with quality assurance, recognition (and public financing) requirements which seem necessary to maintain confidence in academic standards (and accountability for public finances).

      Moodie, Gavin and Wheelahan, Leesa (2022) Credentialing micro-credentials. Journal of Teaching and Learning for Graduate Employability, 12(1), 58-71,

      https://ojs.deakin.edu.au/index.php/jtlge/article/view/1564

      Wheelahan, Leesa and Moodie, Gavin (2021) Gig qualifications for the gig economy: Micro-credentials and the ‘hungry mile’. Higher Education, 83(6), 1279-1295, https://doi.org/10.1007/s10734-021-00742-3

      Like

Leave a Reply

Fill in your details below or click an icon to log in:

WordPress.com Logo

You are commenting using your WordPress.com account. Log Out /  Change )

Twitter picture

You are commenting using your Twitter account. Log Out /  Change )

Facebook photo

You are commenting using your Facebook account. Log Out /  Change )

Connecting to %s