It is some time since I have read a survey as thought-provoking and interesting as this one. The basic issue that Hellwig addresses is the role of financial institutions in the allocation of capital for investment. Historically, there are two types of institution that have been important; the first is banks and the second is stock markets.

These institutions have played varying roles in different countries at different times. For example, in Germany and Japan banks have been much more important than in the UK and US where stock markets have played a more significant role. One of the things that I particularly liked about the paper was the use of historical and contemporary evidence for assessing the various theories that have been suggested. Moreover, the theories that are considered are not restricted to those suggested by economists; the hypotheses developed by economic historians are also considered.

Why is it that different countries have such different financial institutions? As Hellwig points out, standard economic theory has little to say on this issue; the Walrasian model simply assumes a set of frictionless institutions. Until recently, the most popular explanation for the role of intermediation was based on transaction costs. Gurley and Shaw (1960) argued that financial intermediaries such as banks transformed the earnings streams generated by firms into a form that investors found desirable. For this argument to hold, the technology for issuing securities must be such that it is less costly for intermediaries to do this repackaging than for investors to hold securities directly. In this view, banks improve the efficiency of the economy relative to stock markets alone since they allow a more efficient allocation of resources.