Some of the UK’s most prestigious university departments are losing seven-figure sums under reforms tied to the Research Excellence Framework (REF) despite excelling in the latest assessment.
At the University of Oxford, the annual allocation for mathematical sciences – for which it ranked number one nationally in the 2021 REF – fell by ?734,000 to ?7?million this academic year, representing a?cut of almost 10?per cent, while its computer science and informatics grant dropped by 12?per cent to ?3.2?million – a??425,000 cut – despite rising from seventh in 2014 to second on its grade-point average (GPA) ranking.
For economics, Oxford’s annual REF allocation fell by ?1.2?million to ?1.3?million – a 44?per cent reduction, despite improving on its 2014 GPA score. At the London School of Economics (LSE) – ranked top for economics in 2021 – its quality-related allocation for that discipline fell by a third to ?1.1?million this year, following a cut of ?560,000.
Other leading departments are also facing cuts in their annual quality-related (QR) funding despite strong performances in the 2021 REF, the results of which were published last May. It follows several changes from the 2014 exercise, including the requirement for all research-active staff to enter at least one research output, which led to a 46 per cent rise in the number of submitted academics.
With a higher proportion of these outputs also being judged 3* or 4*, the amount received by “REF-scoring” academics (outputs rated 1* or 2* do not attract funding) has fallen substantially. According to an internal report prepared by the LSE and seen by Times 中国A片, the average value of a REF-scoring output fell from ?4,336 in 2021-22 to ?3,026 in 2022-23, despite the overall QR budget increasing by ?205 million to about ?1.26 billion.
The effect of more outputs and higher scores was, however, more pronounced in certain REF panels. While the volume of REF-scoring outputs rose by 72 per cent overall, the increase for Panel C (social sciences) was 92 per cent. Given that the LSE had already received 100 per cent 4* submissions in several units of assessment (UoA) in the 2014 REF, the lower returns per academic seen across Panel C affected QR income significantly.
Overall, its QR income fell in eight UoAs, with law hardest hit by a 43 per cent drop in income, equivalent to ?734,000 a year. Cuts were also felt by business and management (?531,000) and politics and international studies (?443,000). At an institutional level, LSE’s REF income will fall by ?1.9 million a year to ?12.5 million – the largest drop of the 17 institutions whose funding will decrease from 2021-22.
With the share of REF funding fixed at main panel level for the first time – in effect, preventing the transfer of money to disciplines where excellence rose substantially – historically high-achieving universities such as LSE would inevitably lose out, said Hamish Low, head of Oxford’s department of economics.
Economics, whose overall REF funding has fallen by 10 per cent to ?12 million, is hardest hit but would have also received an additional ?6 million if the 2014 rules on inter-panel transfers of funding still applied.
“It is certainly true that this allocation moves funding away from institutions that are more pro-social sciences,” said Professor Low, who said most academics were unaware of the rule change announced by Research England in October.
“If this [shift in funding] had been part of a political discussion, then that would be fine, but I don’t believe there was that open discussion. Instead, the allocation across broad areas was fixed in a simplistic and coarse way that resulted in arbitrary swings.”
Research England said it had sought to “maintain stability and continuity in our investment in research” and noted that, as QR is a block grant, there was “no requirement for [universities] to mirror Research England’s calculations” on subject-level allocations.
jack.grove@timeshighereducation.com
How did REF 2021 alter funding allocations by discipline?
| Unit of assessment | QR 2020-21 (?m) | QR 2022-23 (?m) | Change (?m) | Change (%) |
| Business and management studies | 44 | 66 | 22 | 49% |
| Art and design: history, practice and theory | 27 | 39 | 13 | 47% |
| Public health, health services and primary care | 38 | 55 | 17 | 45% |
| Sport and exercise sciences, leisure and tourism | 13 | 19 | 6 | 42% |
| Allied health professions, dentistry, nursing and pharmacy | 61 | 80 | 19 | 30% |
| Computer science and informatics | 47 | 61 | 14 | 29% |
| English language and literature | 30 | 38 | 8 | 26% |
| Sociology | 9 | 12 | 2 | 26% |
| Architecture, built environment and planning | 17 | 21 | 4 | 24% |
| Engineering | 129 | 159 | 30 | 24% |
| Communication, cultural and media studies, library and information management | 13 | 16 | 3 | 21% |
| Earth systems and environmental sciences | 31 | 38 | 6 | 21% |
| Psychology, psychiatry and neuroscience | 64 | 77 | 13 | 20% |
| Politics and international studies | 17 | 21 | 3 | 19% |
| Area studies | 7 | 9 | 1 | 19% |
| Law | 21 | 25 | 4 | 19% |
| Music, drama, dance, performing arts, film and screen studies | 21 | 24 | 3 | 15% |
| Clinical medicine | 96 | 109 | 14 | 15% |
| Agriculture, food and veterinary sciences | 17 | 20 | 2 | 14% |
| Physics | 47 | 53 | 6 | 13% |
| History | 26 | 30 | 3 | 12% |
| Theology and religious studies | 5 | 6 | 1 | 12% |
| Mathematical sciences | 55 | 61 | 6 | 10% |
| Social work and social policy | 17 | 18 | 2 | 10% |
| Chemistry | 34 | 36 | 2 | 6% |
| Anthropology and development studies | 8 | 8 | 0 | 6% |
| Education | 21 | 21 | 1 | 4% |
| Modern languages and linguistics | 19 | 19 | 1 | 3% |
| Biological sciences | 62 | 64 | 2 | 3% |
| Philosophy | 9 | 9 | 0 | 1% |
| Geography, environmental studies and archaeology | 35 | 34 | 0 | -1% |
| Classics | 6 | 6 | 0 | -4% |
| Economics and econometrics | 13 | 12 | -1 | -9% |
| Total | 1,061 | 1,266 | 205 | 19% |
Source: LSE internal report. Mainstream QR funds; figures rounded to nearest million pounds.
