The existence of a competitive, efficient and virtually global freight market in the late nineteenth century is indisputable. At the same time, because fleets like the British and American employed many foreign seamen, it seems possible that there even was a parallel market for maritime labour. Recently, however, Charles P. Kindleberger expressed strong doubts about the existence of efficient maritime labour markets. His arguments were primarily provoked by some sweeping statements in a collection edited by David W. Galenson; perhaps for this reason his book is more a commentary than a monograph. Yet his main points merit consideration. He argues that labour markets were not efficient because the violence and force exerted to enlist and supervise sailors meant that they were not voluntary participants. Moreover, he stresses that because wages differed among various fleets there was little market integration, a feature repeatedly argued by Lewis R. Fischer.

Still, I cannot escape the feeling that the historical development of labour markets, whether national or international, is difficult to comprehend in purely neo-classical terms. Microeconomic theory views labour markets as similar to commodity markets: labour is supposedly an uniform “commodity” which can be bought and sold, with its price depending on supply and demand. But this often does not conform with the real world. In the modern world, for example, trade unions and other organizations affect the market decisively. Indeed, even during the nineteenth century - when unions were rare and political economy was often based on the principle of laissez-faire — labour was subjected to important spatial and social constraints. The most fundamental difference from commodity markets is the fact that labour markets involve human beings who could not be transported like logs and had social ties and preferences of their own. While the nineteenth-century European grain market could be affected by knowledge of the last rice harvest in Burma — since rice could be transported from the other side of the globe in a matter of months - the existence of a large unemployed pool of labour in eastern Europe had no direct bearing on North American wages because the Polish or Russian rural proletariat could not be employed on the other side of the Atlantic. Of course, seasonal labour migrations over relatively long distances have occurred for many years but, in addition to being costly, have been limited by various societal and governmental barriers.