Do Research Institutes Make Sense?

building castles on a hill

The Francis Crick Institute is undoubtedly impressive. Europe's largest biomedical laboratory occupies an extraordinary building in London, opposite Kings Cross station and just behind the British Library. Its research performance matches its physical presence. It has attracted talented researchers and created a partnership between the founding charities and the universities that supported it. It has achieved an outstanding reputation in the relatively short period since its establishment. It has enjoyed high profile scientific leadership.

Institutes of this kind are surely the model for how research should be undertaken: talent centred, investigator-driven and strategically oriented. So much better than universities with their messy social missions, awkward governance, intrusive teaching obligations, disciplinary heterogeneity, sub-scale teams, competitive business models, and duplicative infrastructure.

This argument has persuaded quite a few research and science policy wonks, many of them my friends. I believe it to be largely (though perhaps not entirely) misleading. Let me explain why.

The UK has achieved its strong position in research on the back of its university system, for all its admitted flaws. Not simply as a result of a few outstanding universities, but as the result of a wider university system that preserves areas of excellence, feeds those universities, employs their talented graduates and drives research dissemination. And along the way this wider university system delivers world-class research at a matching quality, if not volume and intensity. There are plenty of research systems across the world where research institutes are prevalent, Germany and France are examples. They struggle to compete with the UK. Whilst the institutes are excellent in their own terms, as national research systems they do not match up.

The argument in favour of universities over research institutes in the UK has always been predicated on funding. UK universities are an incredibly cheap place to fund research. The exact cost recovery varies but overall universities only secure about 60% of the cost of research (excluding unfunded scholarship) from grants. The shortfall is paid by fees from teaching international students - an export educational business (undertaken, just so we do not forget, by not-for-profit charities). Now, as the funding picture has changed, the surplus garnered from international students is required to subsidise UK students. This might tip the argument in favour of research institutes, where the funder has, essentially, to pick up the full cost. Despite the financial challenges however, universities are still struggling to preserve research, and it seems like a reckless moment to pull funds away from a delicate system navigating a narrow edge, in pursuit of an unproven alternative.

It may appear that research institutes stand alone. In reality, the excellence of research institutes, such as the Crick, rests on the university system. The talent is for the most part derived from it, and the skills and knowledge it deploys have their basis in it. Much of the early career talent will build their longer-term careers within universities. There is only so much room for 'Principal Investigators'. Universities educate the research consumers. Research institutes thrive when embedded within a dynamic university ecosystem rather than as castles on a hill.

Pulling research away from universities has a larger system effect. Having the UK's students taught by researchers who are active at the frontier of knowledge is important - recall that the UK operates an intensive and generally specialised education system. Teaching is good for researchers too, who must organise their material for communication, and the opportunity to integrate advances in research within the body of knowledge for their subject, and - ideally - persuade their students that what they are doing is interesting and important. They will, in a university, need to engage with growth and prosperity, regional and social agendas. They will necessarily connect with the arts and humanities, social sciences and technology, even if it is only to share the same coffee shop. Universities are, at their best, vibrant, diverse places where research can thrive.

One of the major risks in research and science policy is an implicit assumption that all disciplines are alike. It may be that the biomedical and some physical sciences benefit from a concentrated laboratory based model and a more hierarchical, narrowly competitive 'up-or-out' approach, suited to an established institute, but it is far from clear that the same applies to software technologies, behavioural sciences, engineering and so on. Sustained funding, access to state-of-the art facilities, a larger more ambitious research agenda, active programme management, and plenty of doctoral students go with research institutes, but it is perfectly possible to provide these outside institutes.

My conclusion is, I am afraid, somewhat anodyne. We need diversity of delivery models, of funding modalities, of pathways, and of organisation. We need discovery, challenge, innovation, and application support - joined-up. We need industry labs, public-sector labs, research institutes, research and technology organisations, virtual and physical collaboratives, research 'hotels', and - yes universities. We need, in other words a research - system - and to make this effective requires orchestration rather than siloed and competing models.