"[Neoinstitutional Economics...] theory has made an indispensable contribution in recent times to advances of understanding in this area. But it seems to me that in the economics of institutions theory is now outstripping empirical research to an excessive extent. No doubt the same could be said of other fields in economics, but there is a particular point about this one. Theoretical modelling may or may not be more difficult in this field than in others, but empirical work is confronted by a special difficulty. Because economic institutions are complex, they do not lend themselves easily to quantitative measurement. Even in the respects in which they do, the data very often are not routinely collected by national statistical offices. As a result, the statistical approach which has become the bread and butter of applied economics is not straightforwardly applicable. Examples of it do exist, the literature on the economics of slavery being perhaps the most fully developed - not surprisingly because slavery is an institution that is sharply defined. But to a large extent the empirical literature has consisted of case-studies which are interesting but not necessarily representative, together with a certain amount on legal court cases, which are almost certainly not representative. Is this the best we can do? There is a challenge here on the empirical side to economists to see what is the best way forward."