Re-thinking the Comparative Study of Commercial Institutions in Pre-modern Europe

The importance of merchant guilds for the commercial development of Europe is beyond doubt. However, there is still little agreement as to why they emerged, persisted and ultimately declined between the 11 and 18 centuries. Historical studies have focused on individual cases and idiosyncratic circumstances that restrict severely comparability, while economic approaches based on game or contract theory often impose narrow assumptions on their models which find it hard to deal with two key features of these institutions: in very imperfect markets merchants used more than one institution to solve a given problem while a given institution often addressed more than one problem. In this paper we suggest a new methodological approach that allows us to pursue a comparative analysis without loosing rigour. We assess a new dataset of 185 observations of merchant organisations from four towns 1300-1800 at 50 year intervals. Our model is based on only one assumption: merchants will only delegate control over their dealings if they can expect a positive return from the loss of control. On this basis, we classify our dataset into five ordinal categories of degrees of control delegation. Using maximum likelihood estimation we investigate the probability that merchants, under a given set of market and political circumstances, delegate control. The authors would like to thank Maristella Botticini, Ann Carlos, the late Larry Epstein, Joost Jonker, and participants of the EHES conference Istanbul 2005 as well as the Antwerp workshop on mercantile institutions 2005, the LSE Global Economic History seminar, the Utrecht GEHN conference 2006, the ASSA 2007, the Cliometric Conference 2007, the economic history seminar at the University of British Columbia, the Institutions in Early Modern Europe workshop at CalTech 2008, and the ICS sociology seminar at Utrecht University, for extensive comments and criticism. All the usual disclaimers apply.