How much does it cost to teach an undergraduate student in England for a year, full time?
A new publication written by KPMG on behalf of the Department for Education – Understanding the cost of undergraduate provision and qualifications in the higher education sector in England – provides the latest answer to this perennial question, and the answer varies by subject area and provider.
Medicine, dentistry and veterinary sciences is the subject group with the highest cost – a cool £19,986 for a year of tuition. It is followed by physical sciences (£16,466), and an unlikely melange of earth sciences (geography, earth and environmental sciences, agriculture, food studies at £14,083).
Based on this data the least expensive subject areas are law (£9,475), combined and general studies (£10,276) and other social sciences (£10,340). Those familiar with TRAC data will note that all of these costs are higher than the maximum fee limit: the overall weighted average unit cost across all subjects is £12,317.
There is also a difference in costs between types of provider (and other factors that we will get to), this chart shows average costs for a year of undergraduate study split by subject and TRAC peer group (for the uninitiated, A and B are Russell Group and people who are nearly Russell group, D and E are generally smaller teaching focused universities, and C is the (squeezed) middle.
A bit of context
The sticker price to the UK domiciled student (well, graduate technically) is £9,790: the same across all subjects and all Approved (Fee Cap) registered providers. How much the state actually pays out is modulated by the Office for Students’ use of the strategic priorities grant (SPG) – in particular a student premium that recognises the additional support required for non-traditional students, and a high cost subject supplement.
The cost of the latter was historically determined by an indicator known as Subject FACTs (the “Full Average subject-related Cost of Teaching and OfS-fundable FTE student in a cost centre” if you need that acronym for some reason). This used data collected within the TRAC(T) process to determine an average cost for each major subject area – more expensive subjects received a subsidy.
TRAC(T) data hasn’t been collected since 2019-20 – so the subject price groups OfS uses to allocate high cost supplements is getting on for seven years old. Even before then, it was seen as being past its best (there was a DfE review of TRAC(T) published alongside the Augar review back in May 2019)
What we have in this new report isn’t quite the subject FACTs approach – not least because it looks across just 29 universities and 9 FE colleges rather than the whole sector – but it represents valuable information for anyone trying to figure out what a high cost subject might be in 2026. In essence it looks a little bit like a smaller scale version of the KPMG work that underpinned the mysterious and arbitrary decision by Philip Augar and his panel to recommend that tuition fees should be £7,500.
A bit more on the methods
The figures I’ve cited above used 2023–24 TRAC data as a starting point (the college financial forecast return for the FE colleges), and this has been refined by multiple interactions and data collections with the providers involved. As such, it relies at least in part on providers’ own definitions of what is and isn’t a teaching cost (though this is mediated by TRACT definitions) and it doesn’t quite have the same status as the audited data we are used to in stuff like HESA Finance.
One advantage here is that when we talk about subject costs we are using HeCOS (as with teaching data) rather than cost centres (as with staff and finance data). This makes it easier to check up on student numbers in subjects, which makes the costs per student a little more robust.
It’s only a small sample but it is reasonably robust in terms of subject coverage and provider types for universities, though there is a slight bias towards the south of England and away from smaller providers. We’re only looking at six per cent of FE colleges that deliver HE – it is handy to have this information, but we need to be very careful in relying on it to make comparisons between FE and HE costs.
Expensive specialised equipment
We’re in the world of indicative data even before we started to think about coverage – the CAH1 subject groupings we are using contain a wide range of subjects and courses, all of which will likely have different costs. While the report can safely speculate that:
Specialist and technical courses often necessitate teaching staff with highly specific expertise… the delivery of these courses may demand bespoke physical infrastructure… technical and practical disciplines rely heavily on expensive, specialised equipment
and note that all of these aspects are going to add cost, it is important to note that we do mean “course” here and not “subject.” Film studies, for instance, is not going to have the same costs as film production.
Even within these broad subject areas we see institutional variation within our sample, which reflects – for many disciplines – huge variations in curriculum design and delivery. These differences are more visible in higher cost subjects than lower cost activity, although some of these differences (that we could ascribe to market-driven differentiation) are evened out by threshold professional body requirements.
What is interesting is that, while costs are higher than found in the 2019 study, they are not as high as the 2019 cost plus inflation (we are using RPIX here, which is what we persist in using in setting student loan interest rates). We are seeing very visible evidence of the cost savings that have seen staff cuts, low pay, and course closures in recent years – staff costs have risen by just 14 per cent compared to a 41.2 per cent cumulative RPIX since 2019, something that I am sure UCU is already printing on placards.
Driving differences
Teaching at research-intensive providers is more expensive than teaching elsewhere; likewise delivery using a lower student/staff ratio costs more. As above, specialist courses have their own expensive requirements. And it’s 12 per cent more expensive to do pretty much anything in London.
All of these factors play a part in our understanding of cost variations across subject areas too: a specialist subject (for example medicine) delivered primarily by research-intensive providers in small classes will look more expensive even before we start thinking about what makes medicine different from nursing.
A few other factors are slightly more surprising – foundation degrees, your archetypal sub-degree qualification, looks more expensive per year than traditional undergraduate degrees. As – sorry DfE – do foundation years. We are dealing here with very low numbers of students within the sample, to be clear.
And – expect a lot of hay to be made about this – the average cost of a year of degree tuition in an FE college is just £6,287. As well as the other factors noted above (subject mix, non-specialist provision, variations in staffing type and level) we need to be aware that this represents a tiny non-representative sample of FE colleges.
One thing that is in here but will not be driving differences is the margin for sustainability and investment (MSI – which is set for 2023-24 at a blanket 9 per cent). For some reason the idea that universities may have historic losses to recoup, and solid risk mitigation policies that require some funding to be held to cover known future costs, was very controversial around the time of Augar. Calculated based on audited financial statements and projections, this calculation was last reviewed in 2025, and the updated figures are represented in this years’ TRAC.
Policy implications
This exercise sets out clearly that, in the overwhelming majority of cases, the cost of teaching in a university is not covered by the tuition fee. It also demonstrates that the savings that providers have made in order to address this gap have hit academic staff particularly hard – but mean that, in real terms, university teaching is cheaper now than it was in 2019.
For anyone who works in a university neither of these things will be unexpected. Neither will the fact that, due to recruitment pressures, some universities are currently carrying higher costs per FTE student than were planned or budgeted for.
The addition of a very limited analysis of HE teaching within FE colleges poses more questions than it answers – it is not a safe conclusion to decide that FE colleges are doing a comparable activity cheaper. If this is an argument DfE wishes to make, it needs to collect a lot more data. Likewise, we have very limited data on costs within specialist providers, or on the costs of part time or online (or even blended approaches).
It’s an open secret that OfS and DfE are both thinking very carefully about the future allocation of the strategic priorities grant – should it be used to prop up high-cost courses where student demand exists that would otherwise be unviable, or should it be used to shape the sector via strategic allocations linked to government priorities?
This is not a straightforward question for either body to answer. But the sector is shrinking (unless we see a far more radical policy and funding settlement than pretty much anyone is expecting in the coming budget) – and a managed decline is preferable to a series of disorderly market exits. The decision ministers need to make is about what to save and what to let go. And I don’t envy them that.
