Working Papers

Local Credit and International Trade.” Forum for Research on Empirical International Trade WP No. 1146  Latest draft: April 2017.

Does local access to credit affect large-scale firm outcomes like exporting? I answer this question by modeling the relationship between finance-constrained exporters and bank entry decisions. Heterogeneous firms must finance fixed export costs via local banks that charge interest rates that are decreasing in bank branch presence. This generates bilateral trade equations where local access to banking Brazil_Labelled_Map_banks_exports_spincreases the intensive and extensive margin of exporting. I estimate this model with a panel of Brazilian municipal-level trade and banking data and show that commercial bank presence per person increases bilateral exports. Literature in the financial development field has struggled to deal with the endogenous relationship between finance and economic outcomes. To untangle this reverse causality, I instrument bank presence by using geographic characteristics particular to the bank branching decision in the spirit of Frankel and Romer (1999). I build predictors of city branch levels with bank company characteristics including the geography of bank headquarter locations. I test the robustness of this instrument with measures of geographic financial remoteness. My results show that local bank access matters: a one standard deviation increase in bank branches per person raises city-level bilateral exports by at least 8.1%. The effect is even stronger for industries where the credit constraint binds: bilateral industry exports increase by a much as 46.0% in sectors that use less internal funds and have more difficulty producing collateral.

Works in Progress

“Port Regions, Exports, and Subnational Trade.”

This paper investigates economies of scale in shipping and patterns of subnational trade. I present evidence that port regions – the final location a good travels to before it is exported – have cheaper access to goods produced for domestic consumption. This has important implications for the effect of market access on income: not only do regions with better access to foreign markets benefit by increased export sales, those regions also experience cheaper access to domestically produced final and intermediate goods. I motivate this theoretically with a multi-sector, multi-region trade model that includes domestic trade costs that decrease in foreign exports. I use 2012 U.S. Commodity Flow Survey Data on domestic trade patterns to identify this effect empirically. Controlling for sector, transport mode, and region characteristics, I show that intranational trade between regions in goods destined for consumption in the U.S. is increasing in shipments destined for foreign markets. Ignoring this relationship means means that traditional trade flow estimates will overstate the effect of distance on trade. Including the foreign shipments term reduces the estimated effect of routed distance on domestic trade by 10.9%.

“Distance, Geography, and Branch Banking.” 

In this paper, I analyze the geography of bank company branching decisions and their affect on access to finance. I focus on three essential aspects of this unit of analysis: financial products are largely homogeneous, yet banks still enjoy significant market power, many companies build multiple branches in a single market, and those markets are often geographically distant from the bank company’s headquarters. I capture these effects by building a model of heterogeneously productive banks that compete for the share of monopolistic profits by building bank branches. I embed this in a general equilibrium economic geography model of production. Producers have a simple cash-in-advance constraint: all variable labor costs must be financed with external capital. The model shows that bank branching behavior and market share are driven by company size and geographic characteristics of the bank’s headquarter region. I test the predictions using Brazilian bank branch balance sheet data and show that the bilateral relationship between a company’s headquarter city and the markets they branch into are significant determinants of company behavior. This company-level effect aggregates into explaining city-level financial outcomes: bank access is decreasing in remoteness from headquarter regions. I show that this result means that financial depth is also decreasing in bank access and remoteness. In sum, this paper shows that within-country geography can help explain endogenous financial development due to its effects on bank company behavior.

“A Little Push: Regional Redistribution and Export Expansion.”

Domestic market size can be an important determinant of firm expansion in the presence of trade costs, an effect that is magnified if firms face liquidity constraints. In this paper, I investigate whether regional redistribution policies can relax the liquidity constraint and stimulate firms into becoming exporters. The mechanism is as follows: to expand outside of their local region, firms must finance fixed costs with profits from local sales. Trade costs and non-homothetic preferences mean an increase in average income induces consumers to buy more local manufactured goods, which allows firms to expand outside of their region. Empirically, I look at Municipal Participation Fund (FPM) transfers in Brazil, a redistributive tax policy that accounts for approximately 40% of city government budgets. Preliminary results show that conditional on city size and income, FPM transfers increase city-level exports. To tease out causality, I use an FPM formula change in 2004 as a natural experiment, as the change is an exogenous transfer shock.