A capital way to stretch research and resources

University endowments could be the key to promoting better basic research, writes Julian Le Grand .

The Conservative proposal at the time of the last general election to replace the present system of recurrent grants to universities by one-off endowments deserved a little more attention than it received at the time.

It rapidly became clear that endowing each university with a capital sum that it could invest to fund its activities involved such large sums that it was a non-starter. But there is a grain of merit in the essential idea if it were confined to the funding of basic research.

A university's ability to do basic research should be independent of government and particularly of the annual round of public expenditure decisions. Most basic research is by its very nature long-term, whereas an annual grant is vulnerable to the short-term fluctuations of the political and economic cycle. More importantly, so long as research is dependent on an annual dose of public largesse, there is always the danger that the government will try to direct researchers to serve its own ends.

Similar dangers arise with alternative sources of funding such as the private sector or charitable foundations. Private companies are primarily concerned with profits, while foundations are increasingly directing research, often at the whim of their trustees or their chief executive. He who pays the piper calls the tune may be a cliche but it is not an empty one.

The four UK higher education funding councils provide about £1 billion through the research elements of their grants. In 2001-02, the Higher Education Funding Council for England will provide £888 million. A formula determines the amount received by each university depending on the subject areas the university supports, the number of research-active staff in each area and the grade that departments in the area received in the latest research assessment exercise. Any department that received a grade below 3b gets nothing.

Suppose this were replaced by an endowment scheme. Eligible universities would be allocated a capital sum that they would be required to invest in a trust fund. They would be free to manage the fund and to contribute to it themselves. To encourage contributions, the government could offer to match such money up to a predetermined limit. In the early years, universities would be barred from withdrawing capital from the fund. The income from the fund could be used only to fund basic research.

If this all sounds a little like the higher education equivalent of the Labour government's proposals - also made just before the election - for a capital grant to children (the "baby bond" or children's trust fund) and a special savings account for the poor fed by family savings matched by the government (the "savings gateway"), that is no coincidence. Analysts and policy-makers are increasingly aware of the advantages of independence and autonomy that owning assets confers - advantages that apply as much to institutions as to individuals.

To generate the same overall income in England as universities receive for research from Hefce would require about £18 billion (at a real return of 5 per cent). Hefce currently spends £4.8 billion. The £18 billion allocation could be made more affordable if it were spread over, say, a ten-year period. The extra cost would be only £1.8 billion in the first year, decreasing in each subsequent year. For every year after, universities would receive income from their invested endowments. Hence each year the grant could be reduced by an amount equal to 5 per cent of the endowment already received.

For English universities, this reduction would be £88 million in the second year, £176 million in the third year, £264 million in the fourth year and so on. By the tenth year, the grant would have dropped to zero, universities would have their full endowment and annual UK government spending could be reduced by £1 billion.

Funding could be distributed in direct proportion to the existing method. Although this might lock in the present pattern of research activity, universities that were up-and-coming in research terms would be invited to bid for endowment status at a later stage. Funds would have to be set aside to support this and a procedure set up to assess the bids, perhaps through some RAE-type assessment. But it should be noted that no other university would have to go through the RAE ever again.

University endowments would mean more freedom and less regulation for universities and, eventually, less public spending for governments. But, more importantly, through promoting researchers' independence, they would mean more and better basic research - and we would all benefit from that.

Julian Le Grand
Richard Titmuss professor of social policy
London School of Economics.