Sleaze damages the economy as well as politics

Most people who have read Joseph Heller's novel Catch 22 remember the catch: if you are mad, then that is grounds for being declared unfit for flying dangerous bombing raids; but claiming insanity to avoid flying is the act of a sane man.

But there is a short scene in the novel which seems to me to contain an idea which is of far greater importance than the eponymous "catch". A central character has been discovered committing a selfish act which could endanger others. He is confronted with this and asked the question much loved by headmasters when faced with naughty schoolboys: "What if everyone did that?" After some thought the answer comes: "Then I'd be a fool not to."

Economists will instantly recognise a Nash equilibrium here (after the mathematical economist John Nash) - a situation where, given everyone else's behaviour, each person is acting in a way which is individually rational. What the Catch 22 example suggests is that such equilibria may not be very pleasant places to get trapped.

Nasty equilibria are often ones where cheating or breaking conventions or stepping outside the law becomes individually advantageous, though collectively very costly. Laws and social conventions can be crucial in preventing societies being trapped in nasty equilibria where standards of living can be low. It is useful that there is a convention, increasingly backed up by sanctions from stewards, that people do not stand on their seats at football matches to get a better view. Although doing so would certainly be rational if other people around you did, it is collectively self-defeating.

The football example is a bit trivial, but there are good reasons to believe that the strength of social institutions (laws, conventions, the ways in which rules are enforced and changed) in preventing inefficient equilibria may be a crucial factor in explaining the massive differences in wealth and income across countries.

Consider two hypotheses to explain income differences across countries. First, there is the common sense idea that differences in resources (land, raw materials) and in accumulated investment (capital and facilities for training) account for most of the differences in wealth across countries. The second hypothesis is that in fact resource differences account for a relatively small part of the inequality in income across countries and that it is a failure of social institutions, including the institutions of state, to prevent economically harmful, though individually rational, activity that explains much of the most dire poverty in the world.

In a fascinating recent article*, Mancur Olsen, a deep thinker about the structure of economies whose work is unsensational and profound, tries to discriminate between these two theories. In his rich analysis he makes some telling points against the theory that differences in resources can explain why the group of poorest countries have income levels less than one-twentieth of those in the industrialised countries. Availability of land certainly does not account for differences: there is actually a negative correlation between real per capita income across countries and the amount of space per person. (The table gives a good idea why this is so.) And in a world where technological knowledge has many of the characteristics of a public good - available fairly widely with a short lag - the idea that lack of access to productive know-how has large sustained effects is not very plausible.

Of course, there are large differences in the level of capital available to workers in rich and poor countries. But this is not an exogenous factor - it reflects sources of economic success rather than being an underlying cause. But anyway, the differences in output per worker and in the productivity of capital across countries cannot be convincingly explained in terms of differences in capital. As Nobel prizewinner Robert Lucas has pointed out, countries with lots of labour relative to capital (the poorer countries) should - if they just differ in the relative amounts of people to machines - have much higher capital productivities than richer countries; but they do not.

Professor Olsen also provides some interesting evidence on the wages of immigrants to developed countries. According to US data, the earnings of immigrants to the US from Haiti (one of the poorest countries in the world) were between a half and two-thirds as high as earnings of immigrants from West Germany. But domestic Haitians earn a very much smaller fraction of domestic West German average earnings. It seems that being in the US dramatically increases the relative productivity of Haitians; this is somewhat hard to square with the idea that it is lack of skills that accounts for Haiti's poverty. But it is consistent with the idea that some other features of Haitian society explain its low incomes.

A final, rather compelling fact, which is consistent with the importance of legal, political and social institutions in accounting for wealth and poverty, is that national borders often sharply divide areas of very different per capita income.

None of the observations above constitutes a knock-down argument that institutions - widely defined - are crucial for prosperity. But the idea that social and legal institutions are of prime importance is persuasive even in the absence of macroeconomic evidence. A society in which fraud and corruption is widespread and adherence for laws (either out of respect or fear) is minimal is one where many types of economic contract cannot be sustained. If cheating in examinations is widespread why study when there are quicker routes to getting technical qualifications? Why try hard at your job if promotion depends on who you bribe rather than on performance? Why compete for a contract in terms of the quality of the product and its price when contracts will be decided on other grounds?

If everyone else breaks the rules it may be rational to do so oneself - as Mr Heller's hero saw all too clearly. Uncoordinated market forces do not prevent inefficient equilibria precisely because it may be in no one's interest to stop "cheating"; nor is it easy to see how institutions which prevent inefficient Nash equilibria can emerge from the actions of individuals.

One implication of all this is that it may be hard for countries to emerge from low level equilibria - an idea that is consistent with persistence in income inequality across countries. But it may also be possible for countries where laws and conventions prevent collectively harmful behaviour to slip into less efficient equilibria; good social institutions are hugely valuable but may be fragile.

A belief that "everyone is at it", that politicians are routinely taking brown envelopes stuffed with fivers, is damaging in itself, even if wildly inaccurate. At the same time, a society where social security rules may give large numbers of people incentives to misrepresent their position - a nice way of describing fraud - is generating problems which go far beyond the pure money values involved. It is ultimately in no one's interest to live in a society where "cheating" is accepted as unexceptional, and where only a fool would not bend the rules.

David Miles is Professor of Economics at Imperial College, University of London and an economic adviser to Merrill Lynch.

*"Big Bills Left on the Sidewalk: Why Some Nations are Rich and Others Poor". The Journal of Economic Perspectives, Spring 1996.

Persons per square kilometre Argentina 11 Zaire 13 Brazil 16 Kenya 25 India 233 West Germany 246 Belgium 322 Japan 325 Holland 357 Singapore 4,185 Hong Kong 5,632