Inspirational reality check

Howard Davies is impressed with the prizes in a Stiglitz lucky dip

Festschrifts pose a particular challenge to the reviewer. Those who edit them like to ensure that the selection is as broad as possible, to show off the range and variety of work stimulated by our hero. As a result, it is hard for any one individual to claim expertise in every area covered, and difficult to decide how to give fair weight to the different contributors.

Economics for an Imperfect World presents an extreme version of this dilemma. The four editors have assembled 32 pieces of astonishing variety, ranging from an analysis of the cultural origins of different national attitudes to punctuality to a substantial study of regulating non-linear environmental systems under Knightian uncertainty. The diversity is, in itself, a tribute to the remarkable fertility and stimulation of Joseph Stiglitz's own work. Indeed, he may be the only man in the world properly qualified to pen a knowledgeable review.

However, even in the bracing academic climate of US economics departments it is hardly acceptable to evaluate essays proffered as a tribute to one's own work, so others must try their best. They would be assisted a little in their task if the publishers had followed convention and listed the contributors, with their principal publications and affiliations - a curious omission in an otherwise well-produced volume. Some, such as George Akerlof or Kenneth Arrow, need no introduction; others are less familiar.

But running through this cornucopia there is a unifying theme, at which the title hints. The thesis is spelt out by Richard Arnott et al in a short introduction. They argue, following Stiglitz himself, that "in the policy world it still appears that those who analyse and advise draw their basic instincts from the perfect information, perfect competition world of Economics 101, rather than the living, breathing world that we inhabit". By contrast, Stiglitzians (?Stiglitzers) recognise that information is never complete.

The prime target, the policy world he dislikes, is the International Monetary Fund. That is par for the course in books by or about Stiglitz. It is made abundantly clear in the essay that closes the collection - a version of the lecture he gave in Stockholm in December 2001 on receipt of his Nobel prize. The material is familiar: the phrasing as trenchant as ever. He denounces the counterproductive nature of the policies pursued by the IMF before and after the Asian crisis, and praises the "more comprehensive approach to development" pursued by the World Bank while he was there. Even in an acceptance speech, which one might think is a moment for reconciliation, he accuses the IMF of promoting policies that "predictably led to riots" in Indonesia and elsewhere.

That, as I say, is well-trodden territory. Few admirers, or critics, of Stiglitz will be unaware of his critique of Stan Fischer and his team. So it is more fruitful to dip into the other 32 chapters to see what we find.

Is it true, as the editors maintain, that Stiglitz's prizewinning work on information asymmetry has indeed spawned a literature full of insights into macro-economics, public economics and development?

To answer that question let us use a technique unjustly neglected in the academic literature: what we might call "dipstick analysis". The first dip produces paper four in the collection, by Dwight Jaffee and Thomas Russell, called "Markets under stress: the case of extreme event insurance". They ask why insurance markets for extreme events often close up in the period immediately following the occurrence of such an event. The weeks after September 11 2001 are the obvious illustration, when catastrophe or terrorism risks became uninsurable, but there have been other temporary shutdowns, too, after great storms or hurricanes.

This commonly observed market reaction does not seem logical. One might expect premiums to rise, but cover should be available at a new price level. Jaffee and Russell draw on Stiglitzian analysis to develop theories of what they call "ambiguity aversion" and "irrational abhorrence", which have strong analytical power. They explore ways in which these shutdowns might be avoided. They are too critical of the British government's approach to terrorism insurance (Pool Re), but even the short summary presented here of the argument for and against government intervention will be useful to the policy-makers they seek to influence.

The same is true of Aaron Edlin's essay on per-mile premiums for car insurance. Edlin begins with the startling facts that Americans drive 2,360 billion miles a year and that, even a decade ago, accidents cost more than $400 billion, greater than the total cost of petrol. The average insured cost of accidents is roughly 4 cents a mile, but the marginal cost - the cost if an extra mile is driven - is roughly 7.5 cents, because of accident externalities. In high-density states in the US, the marginal accidental cost of an extra mile may be as high as 15 cents.

Yet per-mile charging for insurance is, he says, almost unknown, in spite of strong support from influential pressure groups such as the National Organisation of Women. (Edlin appears unaware of the typical British classic car low-mileage policy. If I drive my elderly Triumph Stag fewer than 2,500 miles a year, my premium is remarkably low. But this is, of course, a small part of the market.) The lack of per-mile insurance is partly attributable to the cost of mileage verification. Edlin argues that these costs are being reduced through technology, so that distance-related premia may soon become a practical proposition. Indeed, insurance could even be billed at the time petrol is purchased: a profitable new business line for BP, perhaps. The environmental benefits could be considerable, if the marginal cost of driving is roughly doubled.

Franklin Allen and Douglas Gale desert the forecourt for the rarefied atmosphere of the Basel Committee's deliberations on the future capital adequacy regime for banks. They do not think much of my former regulatory colleagues at the Financial Services Authority and the Federal Reserve, who, they say, are operating in a theory-free zone. They cannot understand the intellectual rationale for much of what we now call prudential regulation. Indeed, they see no case for government-imposed controls on bank capital, even in the presence of a deposit insurance regime. Banks, they think, "will choose the socially optimal capital structure themselves, without government coercion". The balance of the article is structured as an extended proof of this proposition.

It is true, in the UK at least, that almost all banks hold appreciably more capital than regulators require. In that sense, banks, stimulated by rating agencies and investors, determine their own capital adequacy. On the other hand, a number of unhappy cases have shown that the management of banks faces an asymmetrical choice in determining the institution's risk appetite. In the short run, a more risky portfolio is likely to generate higher returns, bringing immediate rewards to management, sometimes very large ones. Yet losses on the portfolio, if and when they emerge, will be borne by shareholders, or other banks if the deposit protection fund comes into play. So regulators can play a useful role, at least if they are able and willing to ratchet capital requirements upwards at the right time.

Allen and Gale also cheerfully ignore the political dimension. How realistic would it be to close down all banking supervision agencies and to let the market find its level? Would countries then be prepared to allow overseas banks to accept deposits in their jurisdiction? I doubt it.

Nonetheless, Allen and Gale's work does pose a provocative challenge to the Basel Committee, to which they should respond. There are, in spite of the assertions here, plenty of pointy-headed theoreticians lurking in central banks and FSAs around the globe.

My last lucky dip comes up with a provocative piece by Luisa Lambertini and Costas Azariadis, who are painting on an even broader canvas. They seek to explain why government expenditure in most developed countries grew so rapidly, and to such an extent, in the 20th century. They theorise that this may be the result of the formation of a liberal coalition that favours such transfers. The coalition they identify is not one of political parties themselves. It is an alliance between young unskilled workers, increasingly affected by growing income inequality, and pensioners, whose numbers are growing rapidly in North America and Europe. Both groups have a strong interest in inter and intra-generational redistribution.

Retirees, furthermore, will rationally favour spending on more generous pension benefits, funded by budget deficits that involve a transfer of resources from the future to the present. If Lambertini and Azariadis are correct, then the prospect for government deficits in countries such as Germany and Italy, with their high and rising dependency ratios, is especially alarming, and even governments in the US and the UK will find it remarkably difficult to roll back the frontiers of the state. Oliver Letwin would do well to read this chapter.

These are especially interesting contributions. But at least a dozen other contributors would have realistic grounds for action against a reviewer who ignored their offerings. If I had but world enough and time, they would be mentioned in dispatches. But even the relatively generous dimensions of a Times Higher review do not permit it.

Does the collection, finally, prove the editors' point? Has Stiglitz generated a rich literature? Undoubtedly. And, at the same time, they have produced ideal beach reading for regulators, Treasury officials and insurance salesmen. Ibiza will be thick with copies this summer.

Howard Davies is director, London School of Economics and Political Science.