Alan Ryan

When you run a great research university, you also run risks - as Harvard University has found to its cost.

Elite but cash-strapped universities such as Oxford, Cambridge, University College London and Imperial College London often gaze longingly at Harvard.

According to almost every measure - other than the quality of undergraduate education - Harvard is top of the heap; it is likely to stay there, sitting on a $20 billion (£10.9 billion) endowment, seducing the best faculty in the world from wherever they might be employed, deploying vast resources on cutting-edge research and contributing the faculty's skill and imagination in the cause of economic development.

But when things go wrong, they go wrong on a grand scale, too.

Anyone given to Schadenfreude - in the literal sense of enjoying the suffering of someone else - will have taken an improper pleasure in the sad tale of Harvard's Institute for International Development, or, more exactly, the tale of one of its component programmes - the Institute for a Law Based Economy.

This saga has been going on, in and out of the courts, since 1997; and the parent Institute for International Development itself was dissolved in 2000. But now that a Boston judge has cleared Harvard of deliberate fraud but found two employees guilty of "conspiring to defraud" the Federal Government, the final curtain is descending.

How much everyone owes the Government will be decided on July 19 - Harvard is relieved that it won't be $102 million; but it might be $34 million, and its employees could be liable for more. For non-mean-spirited observers, it is a cautionary tale about the difficulty of managing a large academic enterprise and the dangers of the Harvard doctrine of "every tub on its own bottom" - the idea that every academic unit tries to raise its own resources and is given wide discretion in what it does with them.

Generally, it's a productive way to run a university. More interestingly, its virtues are its vices. Any energetic, imaginative and ambitious person in command of a centre or institute or whatever will try to expand it; more money means more projects, more research students, a higher reputation and the ability to raise more money. It works best on a light rein, so that everyone feels part of an ambitious enterprise. But the natural recruits for such enterprises are risk-takers, and the line between risk-takers and corner-cutters is notoriously a fine one.

The Institute for International Development was created in 1974 and did good and useful work on the practical implementation of development programmes throughout the world, especially in Asia and Africa, but also in Central and Eastern Europe. When the Soviet Union imploded, the institute exploded. By the time Harvard decided that the institute had become too large, too unwieldy and impossible to police, it was operating in some 40 countries, had 188 staff and a turnover of about £20 million.

The evidence that the institute was impossible to police came from one of its Russian projects. In 1992, the US Agency for International Development coughed up £30 million for a project to encourage Russia to run the economy in a law-abiding and squeaky-clean fashion. In 1997, two of the people running the project were dismissed after being accused of spending their time - and the Government's cash - making their own investments in Russia. Harvard was sued for £60 million in the autumn of 2000 for allowing them to do it. Needless to say, everyone denied wrongdoing and still denies it.

In the meantime, Harvard had bitten the bullet, decided the institute was too much to manage and divided its projects among other departments, schools and programmes. Harvard's account of the matter was that the problems of the Russian project were not decisive - but not irrelevant. So, why does the story induce mixed feelings? Any enterprise the size of Harvard is vulnerable to mistakes or, in extreme cases, malfeasance by employees halfway down the hierarchy; the British experience with Individual Learning Accounts was a model of how not to spend large amounts of money on good causes. But if you spend your time inventing ever more meticulous ways to check up on what everyone is doing - down to requiring them to empty their purses as they leave the campus - you get a limited meticulousness, but not much ambition and enthusiasm, let alone loyalty.

I am psychologically challenged by managing anything larger than a whelk stall, so I wince at the thought of being vulnerable to an enraged government looking for £60 million and a Boston prosecutor looking for scalps. But like ambitious (and non-wincing) vice-chancellors all over the world, I don't see how you can run a great research university without also running the risk that from time to time that is where you will find yourself. Perhaps they do earn their salaries.

Alan Ryan is warden of New College, Oxford.