It seems to be quite fashionable for economists to beinterested in institutions nowadays. At least, there is a growinginterest among economists in the economic effects of institutions(reflected most obviously in the award of the 1991 Nobel Prize to RonaldCoase). And quite a few books on the economics of institutions have beencoming out recently.! Sociologists are not impressed, of course.Institutions are and always have been central to much sociologicaltheory. But for economists, an interest in institutions has in the pastbeen off the mainstream. One reason may be that it is difficult to reachagreement on what institutions really represent, because there are somany ways and levels at which one can consider them. One definition isthat “institutions are the humanly devised constraints that structurepolitical, economic and social interaction” [North (1991)]. “Humanlydevised” might seem too purposive for some tastes. Others refer to”rules of a society or of organisations that facilitate co-ordinationamong people by helping them form expectations which each person canreasonably hold in dealing with others” (Ruttan and Hayami); or”complexes of norms of behaviour that persist over time, by servingcollectively valued purposes” (Uphoff) [both cited in Nabli and Nugent(l989a), p. 7].