The Commonwealth Budget has triggered confusion about higher education funding. How much does the government spend? Has there been a cut or not?

The Budget documents understate government higher education expenditure

The only summary statement of higher education expenditure in the Budget documents is in Budget Paper No. 1, which reports spending on the higher education ‘sub-function’ (sub- of education generally).

But what is in the higher education sub-function? I’ve collated as much information as I can from the Budget papers and I think it means grants administered under the Higher Education Support Act 2003. I can’t exactly replicate it but my numbers are very close – slightly less in every year. I lack expenditure on the Indigenous Student Success Program, which HESA 2003 funds but PM&C rather than DESE administers.

The ‘higher education sub-function’ significantly understates Commonwealth assistance for higher education. As the top line in grey in the chart below shows, using numbers from Budget Statement No. 4 on agency resourcing, it doesn’t even cover money flowing under HESA 2003 itself. The difference is money lent through the HELP loan scheme. Although the Budget papers don’t specifically quantify HELP lending this is likely to become the single largest source of funding for higher education, as international student revenues collapse and the Commonwealth Grant Scheme stagnates.

While not itemising HELP lending the DESE Portfolio Budget Statement includes its estimated cost. This is primarily doubtful debt and interest subsidies, although low interest rates are keeping the latter down. The annual cost figures are for money lent in that year over the life of the loan, not the stock of HELP debt. These costs are expected to increase over the coming years, almost doubling to $2.1 billion between 2020-21 and 2024-25.

The assumptions behind these estimates are not reported. They could overstate eventual costs if they assume the COVID-19 recession reduces repayments by more than now seems likely. However growth in student numbers and student contribution increases from Job-ready Graduates would push up HELP costs. A humanities graduate with $45,000 in debt will take a long time to fully repay, if they ever do.

The DESE PBS also includes expenditure not funded through HESA 2003, including programs funded under the annual Appropriation Acts and the Australian Research Council Act. Leaving these out means that the ‘higher education sub-function’ figure misses $1.1 billion in expenditure budgeted for 2021-22 and $1.4 billion in 2024-25. (Note: These figures are amended on first publication of this post, which included HELP subsidies in this total).

What the DESE PBS does not include is higher education related expenditure funded through other departments. In the Higher Education Research Data Collection for 2019 universities recorded more than $700 million in income from the National Health and Medical Research Council, another $360 million in non-ARC competitive research grants, and several hundred million more in government contracts (a proportion of which are double counts with DESE programs). There is probably more revenue from government contracts on projects that don’t meet the HERDC definition of research.

The DESE PBS also does not include, because DESE is not the administering department, money spent on Youth Allowance, Austudy and Abstudy. The Department of Social Services doesn’t break down outlays on these payments by educational sector, but apportioning spending according to student numbers I estimated a higher education total of $1.75 billion for 2018-19. This figure would have spiked significantly during COVID, but will drop back again in future years (although there has been a modest permanent increase in the payment rate). The Student Start-Up Loan, also administered by DSS, is similarly missing from public statements of higher education financial assistance. Early last year DSS told me that it lent about $200 million through the Start-Up Loan program in 2018-19.

Categorising higher education expenditure

How much is spent is reportedly spent on higher education depends on how it is categorised. The Budget papers are conceptually organised around legislation or departmental administrative responsibilities rather than around higher education as an activity.

Within higher education as an activity there are also different ways of conceptualising expenditure. It could be organised around purposes, such as on teaching, research, and student income support. It can be classified around who receives it – students, higher education providers, or other bodies delivering higher-education related services. Or spending could be categorised according to who it is primarily intended to help, such as students (Commonwealth Grant Scheme, HELP, student income support) or researchers (research block grants, ARC, NHMRC).

Has higher expenditure been cut?

In trying to answer the question of whether higher education expenditure is to be reduced over coming years I am afraid I am going to have to organise analysis around the available data. So I am looking at higher education expenditure identifiable in the forward estimates, which is what appears in DESE’s portfolio budget statement. The trends say something meaningful about the relevant programs, but understate total government assistance to higher education as an activity by billions of dollars.

The chart below shows my calculations for total subsidies, including HELP’s forecast cost but not HELP lending, from the DESE PBS. If readers would prefer to organise the numbers differently the itemised data is here. The news stories about cuts took 2020-21 as the base year, but including 2019-20 provides useful context. If it is the base year then subsequent years are all higher in nominal dollar terms, but there is a decline through to 2022-23 before the totals start heading up again.

Depending on the purpose of the analysis (or the political point being made) the underlying trends could be analysed more easily if we took out the always temporary COVID measures – the extra $1 billion for research and $550 million for student places. To clarify it further I have also removed some spending items of less interest to institutions, such as HELP subsidies, non-research student payments for which universities are just an intermediary, and government payments for legacy superannuation schemes. I have also taken out non-COVID but time-limited grants.

On this basis underlying grants are flat between 2020-21 and 2023-24 at around $11 billion before trending slightly up in 2024-25. This is driven by phasing out Job-ready Graduates transition funding, which I had included in ‘teaching’, and on-going weakness in the Commonwealth Grant Scheme line item. Flat nominal grants will clearly translate into declining real funding. The 2024-25 increase in mainly due to the CGS returning to nominal growth and a spike in NCRIS.

The policy intent behind Job-ready Graduates is that student-related revenue will increase, through higher average student contributions and more students. Most students take out a HELP loan. Unfortunately the HELP lending estimates I can derive from the Budget papers stop in 2021-22, and therefore the chart below cannot show how HELP cash flows will over time offset stagnation in teaching-related subsidies. One of the many strange aspects of Job-ready Graduates is that even though HELP will become relatively more important this change gets off to a very slow start.

On current estimates HECS-HELP lending will decline between 2020 and 2021, only the second time that it has fallen since HECS began more than 30 years ago. The reason for this is that the government decided that pre-2020 students taking subjects in fields with lower student contributions under JRG would get the discount from 2021, but in fields with higher student contributions only commencing students in 2021 and subsequent years would pay the higher rates. But over time the transition to higher student contributions in arts, business and law, and an increase in the total number of students, will push up HECS-HELP lending.

Conclusion

The way the Budget papers are organised makes analysis difficult. Information is scattered across multiple documents, and much of the money that will end up supporting higher education as an activity is not itemised in ways that show this.

While it does not have to be part of the formal Budget papers – it could be something like the long-running Science, Research and Innovation Budget Tables – a single document that summarises all the relevant information would be very helpful.

And, as I have said many times before, HELP reporting must be improved. We have already had one student lending policy fiasco in VET FEE-HELP. Sending $7 billion a year or so out the door with only rudimentary reporting and no proper analysis is poor financial management. It also contributes to a perception that government provides less financial support for higher education than is really the case.