Jeremy Klein calls for a deeper investigation into wealth creation by universities in the UK
In 1998 the government introduced wealth creation as a third mission for universities in addition to teaching and research, and backed it with £50 million a year. Questions are now being asked about the return on investment from this policy. Trying to find answers raises interesting questions in turn.
On the face of it, the results of third-mission expenditure are impressive: 199 spin-off companies were created in the 1999-2000 academic year, compared with 338 in the previous five years, and at low cost. US universities require six times more research funding to produce each spin-off. But many UK spin-offs are tiny, with questionable prospects, so measuring wealth creation is not just a matter of enumeration.
Despite the enthusiasm for the third mission in the upper echelons of the system, the research assessment exercise gives incentives to academics to optimise their research scores rather than to create wealth. This leads, on occasion, to people rushing to publish their work in academic journals where a more commercially astute approach would be to put in patent applications first. A third-mission metric is needed to counterbalance the RAE. In a study for the Higher Education Funding Council for England, Newcastle University's Centre for Urban and Regional Development Studies (Curds) has started the process by testing "selected indicators that might later inform decisions on further targeted funding for knowledge-transfer activities". A wealth-creation metric is unlikely to be straightforward to devise. Research is predicated on the single imperative of publish, publish, publish, so basing the RAE on publications is reasonable. By analogy, wealth should be measured in pounds sterling. But the revenues accrue to the economy as a whole, not just to the university that incurs the costs. Accounting across so many stakeholders is far from trivial and the benefits usually lag behind the costs by a considerable period. The mean period from invention to early commercial adoption is said to be seven years, and it typically takes 30 years until widespread diffusion. So the contribution of a particular university to national wealth would be of little immediate use.
Policy-makers will need to look at proxy measures - indicators of the potential for wealth creation. The diversity of ways that universities can be involved in wealth creation - industrial contracts, consultancy, patenting, licensing and spin-offs - would make a single proxy measure impossible. But a composite measure might be feasible. There are only four basic classes. Input measures such as "aggregate spend on technology transfer" would be good at measuring the emphasis a university gives to this part of its remit. But the link with wealth creation is weak; a university could spend a lot, but not very effectively.
Process measures could track the extent to which universities are gearing up for technology transfer. Despite the search for best practice there is no single model for wealth creation, and different universities have very different ways of doing things. If wrongly designed, process measures could end up evaluating how well a university's approach matches a non-universal ideal.
Output measures such as the number of patents or spin-offs are easy to quantify. But one strong patent might be worth 100 poor ones. So counting them is only a partial indication of wealth creation. Output measures can also be distortive. Patent applications are relatively cheap, so it would be easy for researchers to inflate the number of patent applications beyond that justified by the quality of their work.
Income from the sale of shares in spin-offs was suggested by Curds. But this measures wealth captured by the university, not the economy as a whole, and suffers from the time lag between early stage activities such as patenting and the realisation of income. So, while outcome measures would give a long-term perspective on how different institutions are performing with respect to wealth creation, they would be a slow indicator of progress, and would not reflect recent efforts to improve technology transfer.
None of these measures is sufficient alone. At this stage, given the need to introduce structural change, there is a strong argument for concentrating on input and process measures. But the government has not always been sophisticated in its use of performance measures - as the introduction of performance-related pay in schools and targets in the National Health Service show all too clearly. Rather than jump into a possibly damaging system of targets or league tables, the work that has been started by the Curds study needs to be extended by a deeper investigation of how wealth creation by universities really happens and how it could be improved. A concerted effort to get to grips with the realities of the third mission is now in order.
Jeremy Klein works for The Generics Group specialising in technology transfer.