Towards a third way in higher education reform in Australia: A proposed two-part package

This is the summary paper By Peter Dawkins and Janine Dixon.

Download a copy of the paper (PDF, 1.47 MB)

We are grateful to the Australian Government Department of Education and Training (now called Department of Education) for providing data on which this paper is based, and to comments and suggestions provided by Peter Noonan and Alan Farley.

We have also benefitted from discussions with Bruce Chapman and the opportunity to present an overview of the ideas in this paper to the Senate Committee on Education and Employment in their March 2015 inquiry into the higher education and research reform bill. The paper has also benefitted from discussions with a number of people at the 2015 UA Conference, especially with Glyn Davis.

The Dawkins-Dixon package is based on two key components:

1. A taper model whereby tuition subsidies are reduced according to a progressive taper rate schedule as fees rise above threshold levels; and
2. An enhanced Higher Education Participation Program (HEPP), incorporating scholarships for all low socioeconomic status background students across the system, and additional support for interventions to reducing attrition rates for at-risk students, in the expanded system

The following details of the proposal are, of course, very amenable to variation, and are put up as a starting point for a discussion about what the desirable properties of such a plan should be.

THE TAPER SCHEME

The current version of the Dawkins-Dixon plan reduces the cut in the government subsidy rate from 20 per cent to 5 per cent. It is our view that to undertake a cut of 20 per cent alongside a very significant reform program would place undue strain on the system. It would of course, be open to government not to reduce the subsidy at all.

The chosen threshold fee, at which the first taper is introduced, is set at a level where total funding per student, including the adjusted subsidy, would be 10 per cent higher than at present. This is based on the fact that two government reviews in recent years, Bradley (2008) and Lomax-Smith (2011), concluded that funding per student needed to rise by 10 per cent to adequately provide for the level of quality of teaching and learning on which we should not be willing to economise. 

At the first threshold, the taper rate is 20 per cent.  This is progressively increased in steps of 10 percentage points up to a maximum of 90 per cent at approximately twice the current prices.

Figure 1 below illustrates how the taper scheme might work, using Humanities as an example. 

Under the present system, Humanities students pay $6,152 in fees, with the government contributing a further $5,447, for total resourcing of $11,599.  Humanities does relatively well in the government’s proposed re-estimation of subsidies. Even after the government’s proposed 20 per cent cut to aggregate subsidies, the subsidy to Humanities is barely touched, coming in at $5,332 per full-time equivalent student.  With the lesser cut averaging just 5 per cent, government support for Humanities places actually increases.

The full set of subsidy changes is given in the appendix. This, of course, is another design feature that can be the subject to further discussion, if the basic principles of this policy design are accepted.

In Figure 1 we illustrate how the subsidy would be tapered downwards if the student contribution is increased in the area of Humanities.  Under the taper scheme, as the student contribution increases, the subsidy is gradually tapered away.  At a sufficiently high price, growth in total funding per student eventually grinds to a halt.  This is to curb any tendency in the universities towards excessive price increases.

The green line in Figure 1 illustrates the government’s proposal for a 20 per cent cut to aggregate subsidies and deregulated fees.  With prices close to their present level, the taper scheme returns more revenue to the universities because it embodies only a 5 per cent cut to aggregate subsidies.  However, as prices climb in the fully deregulated version, universities are not penalised.  We are left without an explicit safeguard against excessive price increases.

The full set of taper rates and thresholds is shown in the appendix.

Group of eight universities

Our analysis leads us to the indicative conclusion that the Group of Eight (Go8) universities would be likely to decide to charge their various courses at the tops of either the 50 per cent, 60 per cent, 70 per cent or, somewhat less likely the 80 per cent taper thresholds, and that they would not need to experience a drop in enrolments as a result.  Under these circumstances, the average increase in fees might be approximated by inspecting the student contribution that applies at the top of the 60 per cent taper. This would represent an increase in the face value of a Go8 degree of about $5,600 per annum, substantially less than would expected under fully deregulated fees. It must be remembered, of course, that in practice the student does not actually pay that face value up front. It is paid on their behalf by the government.  With students accessing income contingent loans on average they will typically not start paying any of the increase until around 11 or 12 years’ time, without a real interest rate on the loans.   This would imply that the net present value of such an increase in the student contribution would be of the order of $3000.[2] This is substantially lower than the kind of rises that have been mooted under the fully deregulated model.

As an illustration, if all of the Go8 universities charged at the top of the 60 per cent taper, the government contribution to the Go8 universities falls to $1.7 billion in 2016.[3]  Compared to the business-as-usual case – the present system applied to 2016, with some natural growth in student numbers and inflation – this is a saving of $400 million.  With the increase in fees, the Go8 universities’ revenue increases to $4.3 billion in aggregate in 2016, or 18 per cent compared to business as usual.

---

[2] For the many students who take longer than this to discharge their HECS debts, the net present value of the same fee increase could be much lower. Furthermore as the HECS system is designed to insure students against the risk of default or repayment hardship, this estimate is likely to be an overstatement of the value of the fee increase for most students.

[3] We simplify the analysis by assuming that all students in 2016 will be subject to the new subsidy rates.  In practice, those who commenced prior to the 2014 budget will still fall under the old subsidy rates.

---

Other universities

It is quite possible that some or all of the 31 non-Go8 universities will experience a drop in enrolments if they engage in large fee increases.  Our analysis leads us to the indicative conclusion   that these universities will charge less than the Go8, possibly pricing at the tops of either the 20 per cent, 30 per cent, 40 per cent or 50 per cent taper thresholds.

As an illustration, If all of the non-Go8 universities charged at the top of the 30 per cent taper, the face value of the increase in the student contribution would be about $3,400, and following the same logic as in our discussion of the Go8 situation above,  the net present value of that increase would be of the order of $2,000.  We estimate that there could be a fall in enrolments in non-Go8 universities of around 2.5 per cent.  These students might instead go to non-university higher education providers, or a vocational education course.

We estimate that the government’s contribution to non-Go8 universities would fall to about $3.8 billion, a saving of $700 million compared to business as usual.  The revenue of the universities would grow to $8.2 billion, an increase of 12 per cent compared to business-as-usual in line with the Bradley/Lomax Smith recommendations.

The total government contribution to the universities is therefore $5.5 billion, a total saving of $0.8 billion compared to business as usual. 

The enhanced HEPP program

We propose that at least a proportion of the savings to government should be used to finance a stronger equity package to include a system-wide scholarship fund for all students from low socio-economic backgrounds. Some of the funds could also be used for interventions to reducing attrition rates for at-risk students, in the expanded system.

A possible allocation for the saving in our example above would be:

• $300 million to be allocated to HEPP scholarships for students from low socio economic status backgrounds.  There are presently approximately 150,000 low-SES students, so this fund would be sufficient to pay each individual a scholarship of $2000.
• The remaining funds (approximately $219 million) to be used to support intervention programs through the HEPP program for low SES students and others identified to be at risk of discontinuing their studies
• $315 million, equivalent to the saving from the 5 per cent subsidy reduction, to could be used for other purposes or general budget savings

Other savings

It can be seen that there would be further saving generated by this model which could be used for other purposes. In part this might help take the pressure way from the possibility of cuts to research budgets. It might also help with the implementation of a structural adjustment package for those universities that are most challenged by this kind of reform agenda. 

The results here are hypothetical, but go a long way to describing how a taper system might work.  There are a number of key design issues in implementing the proposal. These include, for example, the following issues.

• What should be the starting point for the level of subsidies before tapering?
• Should the tapering away of the government subsidy operate course by course (thus ceasing when a course’s subsidy is eradicated) or could the tapering away continue beyond this? The latter would have the effect of limiting price rises more than the former and saving more funds from the subsidies to universities that could be used to support a participation and  equity package, such as a university system wide equity scholarship program for students from low socio-economic backgrounds. If the former were adopted, it may need to be supplemented by fee caps or loan caps to avoid excessive prices.
• What rates should the tapers be set at, over what fee ranges, and how progressive should they be?
• How many taper schedules should there be? In our example, there was one for each band. 

More details on the tapers thresholds etc. in the Dawkins-Dixon preferred model.

The taper system

• Universities may increase fees to increase revenue.  Fee increases will result in a reduction of subsides at an increasing taper rate (see Table 1) to act as deterrent to very high increases with the maximum taper at 90 per cent. Relative to Chapman (2015), the number of bands of taper rates is increased, with the aim of producing a greater degree of price competition.
• Assuming that government still requires some savings from the policy, the starting point is that the cut in subsidy rates is 5 per cent which is equivalent to a saving in 2016 of just over $300 million compared to if the 2015 subsidy rates are carried forward to 2016 (assuming growth in student numbers of 5 per cent).
• Universities can increase fees in order to increase net revenue by 10 per cent without incurring any penalty.

Table 1: Taper thresholds & rates

*current student contribution in parentheses (2015 prices)

The structure of subsidies

The subsidies are calculated according to the Department of Education and Training’s (now called Department of Education) cost of delivery model, which is designed to reflect the cost of delivery for each of five tiers.  Each tier is allocated a cost coefficient.  The subsidies are then calculated in order to reflect the relative cost coefficients, and calibrated to achieve the required budget saving.

For this modelling exercise, we adopted the same approach, with two variations.  Firstly, we adjusted the cost coefficient for Tier 1 (equivalent to Cluster 1) from 0.3 to 0.6, effectively doubling the subsidy for this tier.  Even with this adjustment, Tier 1 still receives the lowest subsidy.  Secondly, we calibrated the subsidies to achieve a real cut of 5 per cent to subsidy payments per EFT student. 

Table 1 below show the subsidies calculated for 2016 compared to the existing 2015 subsidies.  In some cases, the 2015 clusters span more than one of the proposed five tiers, so the subsidy rate shown is an average across the relevant tiers, weighted by EFT student numbers.  The subsidy rate for 2016 assumes indexation to inflation of 2.5 per cent.

Table 2

References

• Bradley, D., Noonan, P., Nugent, H., & Scales, B. (2008). Review of Australian Higher Education. Commonwealth of Australia.
• Chapman, B. (2015). A Submission to Senate Enquiry on Higher Education Reform.
• Dawkins, P. J.  (2014). Reconceptualising Tertiary Education and the case for re-crafting aspects of the Abbott Government's Proposed Higher Education Reforms. Mitchell Institute Policy Lecture, 22 May.
• Dawkins, P. J. and Dixon J.M. (2015) Alternative Approaches to Fee Flexibility in Australia: Towards a Third Way in Higher Education Reform in Australia. Centre of Policy Studies Working Paper No. G-252 March 2016.
• Lomax-Smith, J., Watson, L., & Webster, B. (2011). Higher Education Base Funding Review: Final Report. Commonwealth of Australia.