Voting for a Founding: Testing the Effect of Economic Interests at the Federal Convention of 1787

Abstract

Previous work measuring the voting patterns of the delegates to the Constitutional Convention largely focused on either individual delegate positions for a handful of key votes or on state delegation positions for a far broader set of votes. We remedy this limitation by modeling the key first two months of the Convention including both some individual-level and all delegation-level voting, while simultaneously estimating the effect of various economic interests on that voting, controlling for various cultural and ideological factors. The findings suggest that economic factors mattered a great deal at the Convention. The effect of such interests vary however by the dimension of debate-representation, national institutional design, or federalism. We conclude that economic interests exerted a powerful influence on the deep structure of voting at Convention, though not consistently by issue or dimension. Specific interests only mattered on specific dimensions. U nderstanding the voting patterns of the Federal Convention of 1787-more popularly known as the Constitutional Convention-has been a preoccupation of historians, political scientists, sociologists, and economists for well over one hundred years. 1 There is inherent interest in the views of such famous statesmen as James Madison, Benjamin Franklin, and George Washington, among others, but social scientists want to go beyond simply learning their respective views or even measuring their issue positions to a clearer explanation of what interests or ideas could have motivated them. To tackle this complex problem, we must realize that the Convention proceeded in stages. The early period-May through July-focused on key issues for creating a new government: the nature of representation, the internal design of a new national government, and how power would be distributed between the national and the state governments. These decisions had to be resolved before the Committee of Detail could pull together a draft of the compromises. Only then could the delegates proceed to other matters like trade, war powers, and other issues. Without disputing the key role of ideas and and philosophies of government (e.g., Lockean liberalism or republican virtue), we focus on interests that may have shaped debate. Charles Beard (1913) first offered an "economic," or interestbased, theory of the Convention, arguing it was a conflict between two classes: the "personalty" and the "realty." Though Beard's specific thesis has fallen out of favor, the spirit of explaining the Constitution by examining the relationship between key groups and the pattern of votes remains strong whether via a historical narrative or a social science model. In the empirical literature, these efforts typically focus either on modeling coalitions (Jillson 2002; Slez

The Effects of the Great Compromise on the Constitutional Convention of 1787

“…Empirical studies of the convention typically focus on more narrow themes, such as testing different versions of the Beard thesis (Heckelman & Dougherty, 2007, 2010McGuire, 2003;McGuire & Ohsfeldt, 1984, 1986, evaluating coalition formation (Jillson, 1981(Jillson, , 2008Londregan, 1999;Pope & Treier, 2019), and creating spatial maps of delegate preferences based on the statements they made in debate (Dougherty, 2020;Heckelman & Dougherty, 2013;Pope & Treier, 2015, 2019.…”

Father Founders: Did Child Gender Affect Voting at the Constitutional Convention?

“…In doing so we follow in the footsteps of studies that have demonstrated the importance of the gender of children to the attitudes and actions of parents in a modern political context, but in the setting of the eighteenth century. Similarly, we add an entirely novel dimension to the literature on how characteristics of the convention delegates such as wealth (McGuire andOhsfeldt 1984 and1986), slave-holding Dougherty 2013 andPope andTreier 2015), and cosmopolitanism (McGuire 2003), among other things, are clear predictors of delegate voting behavior. Our research confirms that family relationships-specifically the gender of a delegate's children-clearly belong on the list of relevant predictors.…”