Abolishing the Higher Education Research Excellence Framework (the REF)

Abstract

New research from the IEA outlines the serious shortfalls of the REF scheme as a way to fund higher education. The Research Excellence Framework (REF) assesses the research generated by UK universities. Most recently conducted in 2014, it is used by the national higher education funding councils (HEFCE in England) to help in allocating Quality Research (QR) money. It is likely that the REF directly influences the distribution of less than £1.5 billion, representing between one fifth and one quarter of total government research funding (most of the rest being allocated by the Research Councils). Resources distributed according to REF criteria account for only around 4 per cent of the total turnover of the higher education sector and less than 0.2 per cent of government spending. The REF shows that most high-quality research is concentrated in a limited number of leading universities, which receive the bulk of government funding. Even amongst this group, the impact of the REF outcome is relatively minor. For example, at University College London, the funding allocated as a direct result of the REF accounts for significantly less than 6 per cent of university income. The REF uses significant resources and distorts resource allocation within the higher education sector away from teaching and knowledge dissemination. It involves the appraisal of nearly 200,000 pieces of research and significant bureaucratic costs within universities. The REF distorts the allocation of resources towards the particular type of research encouraged by the REF. The REF also raises academic salaries as universities seek to improve their REF ranking by hiring key academics who score highly on REF criteria and depriving competing universities of those key academics. Despite the relatively low levels of funding distributed as a result of the REF, universities allow their behaviour to be distorted by the process because, as a government-backed kitemark, a REF rating provides signals that are useful to universities from a marketing perspective. However, there is a wide range of other, much more effective, indicators of quality of higher education institutions. Furthermore, no other country conducts an exercise quite like the REF. In relation to research funding, the same – or better – result in terms of the allocation of funds could be achieved by different means. The REF no longer serves a useful purpose and should not be repeated. We should gradually move away from institution-based funding towards project-based funding via the Research Councils. Any remaining need to allocate research support to institutions, for instance for capital purposes, should rely on available metrics rather than the REF. More generally, there is a strong case for reducing the total amount of government subsidy for research and expecting universities to generate their own funds for research and scholarship or support it by reducing overhead costs.