PROTECTION WITH MANY SELLERS: AN APPLICATION TO LEGISLATURES WITH MALAPPORTIONMENT WILLIAM R. HAUK JR* What effect, if any, does legislative malapportionment have on international trade protection? This paper argues that in malapportioned legislatures, such as the U.S. Senate, industries become over-represented in a legislature if they are dispro- portionately located in small constituencies. As a result, industries that are dispro- portionately located in smaller constituencies are likely to receive greater protection from international trade. To argue this point theoretically, this paper develops a new model, combining legislative bargaining and a model of lobbying to study trade protection while allowing for a legislature with multiple legislators and differently sized constituencies. We then test the predictions of this new model using tariff votes from the U.S. Senate in the late nineteenth and early twentieth centuries and a panel of tariffs and non-tariff barriers to trade in the U.S. in the 1990s. Considerable support is found for the model’s predictions. Industries concentrated in states where the population is low receive greater protection from imports. 1. INTRODUCTION What effect, if any, does legislative malapportionment have on a country’s inter- national trade policy? This paper argues that industries in small states are better able to lobby their legislator for protection because these industries will be, in some sense, over-represented in a malapportioned legislature. To argue this point theoretically, this paper combines a model of legislative bargaining and a model of lobbying to study trade protection that allows for a legislature with multiple legislators and differently sized constituencies. We then test the predictions of this model using two datasets. The first uses tariff votes from the U.S. Senate in the late nineteenth and early twentieth centuries to show that Senators from small states were generally more pro- tectionist than their large-state colleagues. The second dataset consists of a panel of tariffs and non-tariff barriers to trade in the United States in the 1990s. Our findings indicate that industries that are disproportionately located in smaller states tend to receive more trade protection. The empirical results from both datasets provide con- siderable support for the model, particularly in its predictions on malapportionment. There is a rich and fruitful literature looking at the determinants of trade protection across industries in the United States and other advanced industrial economies. 1 Many of these models are variations on the Ricardo–Viner “specific factors” model, formal- ized by Jones (1971) and Mussa (1974). The major prediction of these models for trade policy is that certain factors of production are not mobile between economic sectors. Consequently, these factors engage in rent-seeking behavior by lobbying for protection from trade for the economic sector in which they are located. Factors in different industries may be more or less successful for different reasons, creating variation in the level of trade protection across economic sectors within a country. *Corresponding author: William R. Hauk Jr, Department of Economics, Moore School of Business, 1705 College St., Columbia, SC 29208, USA. E-mail: hauk@moore.sc.edu 1 For a good summary of early work in this area, see Rodrik (1995). © 2011 Blackwell Publishing Ltd 313 ECONOMICS & POLITICS DOI: 10.1111/j.1468-0343.2011.00387.x Volume 23 November 2011 No. 3