This article is more than 5 years old

TRACing the costs of teaching in higher education

Andrew Bush introduces the groundbreaking KPMG research on the cost of teaching which underpins many Augar recommendations.
This article is more than 5 years old

Andrew Bush is Director of Higher Education Internal Audit and Advisory lead at KPMG.

Cost and price are two different things, but often get conflated.

It’s important to understand the cost to then determine the price, and in a way that is what led the Department for Education and the Augar review panel to commission a study to understand the cost of undergraduate provision across English HEIs.

For almost a year we have worked with institutions to develop, refine and deliver a study to identify the cost of undergraduate provision. We also identified what types of costs are incurred, what differences exist between different institutions and what might cause these differences.

So how did we do it?

We started with the data produced by the Transparent Approach to Costing (TRAC). Some people groan at this point, but there is no need. TRAC is simply the costs reported in the audited financial statements with a margin added to represent a sustainable cost (the full economic cost). A useful guide was produced by the Financial Sustainability Strategy Group (FSSG) – “Mind the Gap”, to explain the importance of the sustainable cost over simply the costs actually incurred. I’ve not gone into detail on that aspect of TRAC here, but take a look at the guide – it’s an easy read.

The cost headings in the financial statements don’t separate out the costs of teaching, research and other commercial and non-commercial activities, so TRAC uses a number of cost drivers to reanalyse the costs into these activities. There are detailed requirements (the TRAC Guidance) that institutions have to follow and a committee of the Governing Body approve that the process followed in producing the TRAC return is in line with the TRAC requirements. The Head of the Institution then approves the TRAC return, in line with other statutory returns that are made to the Office for Students (OfS). This isn’t therefore some fictitious guesstimate of the costs.

Other important attributes of TRAC are that it has been in place for many years (since 1998), the results have been reasonably consistent and it costs all activities. There is a danger in costing a single activity (teaching in this case) in isolation from scratch as the ability for costs to be over- or under-estimated are increased. However, as TRAC includes ‘all’ institutional costs and is a data set that has already been reported, it is far less likely that costs can be ‘engineered’. Some say that Research costs in TRAC may be over-estimated as back in 1998 a core reason for TRAC being introduced was to identify the full cost of research. Hopefully 20 years on, with TRAC being used to inform teaching funding as well as research (TRAC for Teaching was introduced in 2006) in addition to the quality of governance and sign-off of the return, this is less of an issue now.

TRAC(T) or not TRAC(T)… that is the question

So – we wanted to understand the cost of teaching in the DfE study – “that’s easy as there is TRAC(T)!” Well not quite… TRAC for Teaching (TRAC(T)) was introduced in 2006 to provide the funding councils (HEFCE and SfC, Welsh HEIs do not currently do TRAC(T)) with data to inform their teaching funding levels. This is the most important thing to understand about TRAC(T). It was designed to inform what the funding council’s should reasonably fund. It therefore does not include all of the costs of teaching, but does include all levels of publicly funded teaching (e.g. postgraduate taught provision). For instance, the costs related to overseas students, and provision funded by other public sources (e.g. medical courses funded by the NHS) are excluded in TRAC(T). TRAC(T) also focuses on “subject related” costs, therefore “non-subject related” costs are removed. These include the cost of bursaries, scholarships and hardship payments – as well as costs associated with activities for which separate grants are received (e.g. widening access, accelerated and intensive provision).

TRAC(T) is therefore not what it may initially seem. It is not to be dismissed though. All institutions should follow the same requirements, so the relativity of the results is useful. Institutions are able to access benchmarking data, which enables them to compare their TRAC(T) costs relative to the TRAC Peer Group averages, which can be a useful way of identifying where provision may look to be higher or lower cost than it perhaps should be.

The DfE costing study

The study undertaken for the DfE needed to reflect that there is now a different basis of funding HE in England to that in 2006 and that this may change again in the future. Therefore we needed to understand all of the costs involved in delivering undergraduate provision. We did not use TRAC(T) data for the reasons outlined above.

The methodology used the “Teaching” costs (publicly and non-publicly funded) from TRAC, adjusted to remove the costs of teaching that were not on undergraduate or foundation degrees (e.g. postgraduate taught provision) and deducted discrete costs related to international students. Costs were collected at course level, but then aggregated to HESA cost centres, which were further aggregated to eight subject groups.

Through the pilot phase it was agreed that overseas students typically study in the same groups as Home and EU students and there are not material differences in the costs incurred in teaching international students. It would have been arbitrary to adjust the costs to only relate to Home/EU students.

To analyse what makes up the costs, institutions reported costs under a number of headings that were grouped into the following categories:

  • Course delivery staff costs;
  • Non-staff costs
  • [Academic] Departmental running costs
  • Student related central services costs
  • Corporate services costs
  • Estates costs
  • Sustainability adjustment

What we found and what it might mean

In some respects the study identified what many of us knew: that medical, veterinary, engineering, and art and design are the areas with the higher costs; that institutions in London typically incur higher costs and that costs are sensitive to the ratio of staff to students.

We found variation in the subject group unit costs reported by different institutions for the same subject group. But the reasons for this are difficult to pin down. There is typically no such thing as a standard curriculum for each subject, so institutions have discretion, often within the parameters of professional oversight bodies, to determine the content of degrees and the methods of delivery. Institutions vary in size, mission and location, all of which can affect the costs incurred, and institutions won’t all deliver provision equally across the HESA cost centres that are allocated to the different subject groups. However eight subject groups are more granular than the five price bands currently used in the OfS funding methodology.

The institutions participating in the study provided a good cross representation of the sector, in terms of geography, size and mission. But the study was based on one year’s data – 2016-17, so we reflected that in the future there will be different levels of costs incurred that are not included in 2016-17 (e.g. increasing pension costs).

My thanks and appreciation goes to the teams within the institutions that participated in the study. A large amount of work was undertaken in a short space of time. Many of the institutions participating told us that even where they felt they had good information already, they had gained valuable insights from participating in the study.

The study was only focussed on identifying the costs. It did not include any advice or recommendations around what the levels of funding should be in the future. That is something for HM Treasury, the DfE and the OfS to determine. What the study does deliver though is the evidence of what the costs are, which should hopefully inform the price that is ultimately set. Time will tell on that one.

2 responses to “TRACing the costs of teaching in higher education

  1. You are I fear somewhat optimistic about the veracity of Trac figures. In what ways were you able to audit these?

    There appears to be an implicit assumption that teaching costs can be standardised. In reality, how much is spent on teaching is determined significantly by the amount of money available – hence producing extremely heterogeneous quality.

  2. The cost of teaching does not adequately reflect what University’s actually do to encourage students to achieve their potential. All universities offer extra curricular opportunities and support, bespoke bursaries and facilitate peer-learning opportunities which I doubt are in the calculations. Sure, and off-the-shelf analysis can be assumed through TRAC (T), but it won’t contextualise what student’s actually get/could have as part of a wider University experience.

Leave a Reply