This guest blog has been kindly written for us by Professor Alec Cameron, Vice-Chancellor and Chief Executive of Aston University.
In the context of the current review of post-18 education, one of the ideas considered to have some political support was that of ‘differential fees’, or the setting of different fee levels on the basis of discipline, or institution, or some other basis. The idea appears to have little support within the sector, either with universities collectively (via Universities UK), or with surveys of students. These surveys have shown majority support for maintaining the current system of the setting of a (common) maximum fee level per year of study for home students in undergraduate degrees at English universities.
In the event that external stakeholders are not convinced by the preferences of universities and students, it is worth seeking to identify and analyse the perceived benefits and issues with moving towards a ‘differential fee’ system.
The argument that is generally proposed in favour of differential fees, is that they will promote student choice. Students will have the option to choose from ‘less expensive’ courses, to graduate with lower student loans, and to obtain greater ‘value for money’ via a less expensive degree. Beyond student choice, government could incentivize students towards certain degrees, to produce increased numbers of graduates in fields that are aligned with economic or societal needs.
This argument seems logical. However, it ignores all empirical evidence that has been found to apply under current funding arrangements; namely that, where the upfront cost is met by a government-provided income-contingent loan, the market has proven completely price inelastic. Student choice, it appears, is completely insensitive to price changes when the payment of the fee is moderated by an income-contingent loan.
Given that the current funding arrangement in England does not set the fee level that universities must charge, but rather sets the maximum fee that universities must observe, some universities have experimented with setting fees below the maximum level to attract increased enrolments. To their disappointment, the consequence of such price reductions (with the current funding arrangements) has been a reduction in demand, as students have equated lower price with lower ‘quality’. In Australia (which has differential fees, and more on this later), attempts by the Government to encourage enrolments in certain disciplines (such as teaching and nursing) by reducing the maximum fee for these disciplines, led to no increase in enrolments.
So, the argument, that ‘differential fees’ will make the “higher education market work more efficiently for students”, ignores the evidence of how the ‘market’ responds to fee setting under the current English policy settings.
Different on what basis?
To consider ‘differential fees’ in more depth, we need to address the fundamental question: ‘on what basis could fee differentials be determined?’. There have been several different proposals, generally from politicians, in this regard. The only serious consideration is on the basis of discrimination by discipline, either on the basis of ‘inputs’, typically teaching costs, or ‘outputs’, namely graduate outcomes. (I won’t address the concept of discriminating on the basis of institutions rather than disciplines, as a moral approach would provide the highest level of funding to the institutions teaching the least advantaged students, however, given that most funding originates from student fees, it will not be acceptable to charge these students higher fees). The problem with setting differential fee levels by discipline, either on the basis of ‘inputs’ or ‘outputs’, is that both lead to perverse outcomes.
Let’s address the concept of a fee regime where the funding level is set based on the currently-perceived cost of educating a home undergraduate student in that discipline. This is the type of differential funding that is best-suited to universities, as it aligns their income and expenditure. It is, also, the model that wins most support from student surveys, if they had to choose between differential fee models, as being ‘fairest’.
It is generally accepted that certain STEM (Science, Technology, Engineering and Mathematics) disciplines, particularly those with significant laboratory components in their curricula, are most expensive to deliver, due to greater infrastructure costs, and higher teaching costs, due to greater contact hours. If fees were set to reflect ‘costs’, we would expect higher fees in STEM disciplines. If we believed (although we have previously dismissed) that fee setting influences student behaviour, we would expect a decline in demand for these disciplines, at odds with the objectives of the Government’s industrial strategy. While we contend that price setting will not affect demand, do we want a system where science students are graduating with a higher student loan, particularly given that their future earnings are not generally at a higher level (compared to some professions) to pay off their loans subsequently?
The alternative model is to set fees based on (historical) earnings outcomes that are achieved by graduates in different disciplines. This model works best for the Student Loans Company as more money is loaned to the graduates who will (on average) earn the most, and hence, a higher proportion of the loan values will be repaid. It also relies on the concept of ‘return on investment’; the higher cost charged for a degree with a higher earnings expectation is justified based on the higher expected return.
So how would this work in practice? We can utilise the Longitudinal Educational Outcomes (LEO) data to determine the expected return by discipline. We will find that high returns are achieved in areas such as Medicine, Law, Finance, and Engineering, for example. Some of these courses are expensive to teach, notably Medicine and Engineering. Some (Finance and Law), generally much less so. So high fees for Medicine and Engineering, say, will match the costs that universities will incur to deliver these programmes. On the other hand, universities will have a strong incentive to teach more Law and Finance, as the fees are likely to well exceed the costs. At the other end of the spectrum, some of the creative disciplines (such as Music), which generate (on average) low earnings (hence low fees), have high costs to teach, creating a strong incentive to reduce such activities. While cross-subsidies between disciplines are a reality of current university budgets, an outcome-based ‘differential fee’ model would greatly exacerbate this situation.
Learning from Australia
I mentioned previously that the Australian funding system has a differential fee model. So, what are the differences between the UK and Australian models which enable differential funding in the Australian system?
While Australia has student fees funded upfront by government-funded income-contingent loans repaid by graduates through the tax system (similarly to England), the major difference is that the Australian Government also makes a significant upfront contribution on a per-student basis. (Recent changes to Australian arrangements have led to some change in this system, which I won’t go into here). For each home undergraduate student enrolling at an Australian university, the university receives a student contribution (the student fee, funded upfront by the government via an income-contingent loan), and a government contribution (or grant). The ratio of the student contribution to government contribution is, on average, 40% to 60%, but varies greatly between disciplines. By contrast, in England, the upfront funding is (almost) all from the student fee, notwithstanding that the subsequent debt write-off puts estimated government support at 55%.
The two sources of funding provided by a student contribution (fee) and government contribution (grant) enables the two funding sources to balance funding objectives. Generally, the government contribution relates to perceived costs of programme delivery (‘input basis’), meaning that courses that are expensive to deliver are adequately funded. At the same time, the student contribution generally relates to future graduate earnings (‘outcome basis’), meaning the government is relying more heavily on student fees where graduates have greater capacity (via future earnings) to meet a higher proportion of their costs via loan repayments.
The use of these two funding streams in combination enable the two criteria of adequate funding for more expensive courses, and aligning students costs with economic benefits (or maximising loan repayments) to be achieved. The challenge in the English system, in the absence of an upfront government contribution, is that there is only one lever to pull in setting funding (namely student fees), which would need to be pulled in opposite directions to meet competing ‘cost’ and ‘benefit’ criteria.
In conclusion, advocates for differential fees should be clear about how pricing works in the presence of income-contingent loans; in general, the ‘market’ has been shown not to behave as logic may be perceived to dictate. The concept of a market, in the form of price-based competition, has not eventuated in the past, and is unlikely to eventuate in the future.
Notwithstanding the above, the current single funding source in England (namely student fees) creates perverse consequences in attempting price setting based on either ‘cost’ or ‘benefit’ measures. These cannot be overcome with a single funding stream; alternatively, a model whereby (differential) Government grants could complement (differential) student fees, could address these issues.