Inquiries into the behavior of emerging market firms

Abstract

This dissertation addresses two aspects of firm behavior in emerging markets. Chapters 1 and 3 consider the decision of firms to borrow in domestic or foreign currency and the implications of this decision on fixed capital and inventory investment. Chapter 2 explores the effects of institutions, via transaction costs, on the variety of intermediate goods used by firms in different economies, and the effects of this choice on productivity levels. Much has been written recently about the problems for emerging markets that might result from a mismatch between foreign currency denominated liabilities and assets (or income flows) denominated in local currency. In particular, several models, developed in the aftermath of financial crises of the late 1990's, suggest that the expansion in the "peso" value of "dollar" liabilities resulting from a devaluation either aggravated or triggered many of the recent financial crises. However, little evidence has been presented either on the effects of foreign currency debt on investment or the variables that affect firm level debt choice in the first place. Two of the papers in this dissertation (Chapters 1 and 3) attempt to fill this gap using a new database with accounting information (including the currency composition of liabilities) for close to 400 non-financial firms in five Latin American countries. In Chapter 1 (coauthored with Hoyt Bleakley) we estimate, at the firm level, the reduced-form effect on investment of holding foreign currency denominated debt during an exchange rate realignment. We consistently find that this effect is positive, contrary to the predicted sign of the net-worth effect.

(cont.) Additionally, we show that the estimated coefficient can be decomposed into competitiveness and net-worth effects, and provide direct evidence that the competitiveness effect dominates the net-worth effect. We argue that the positive response of investment is the result of firms debt choice decisions. If those firms holding dollar debt are also those firms whose current and future incomes benefit most from a devaluation then it is not surprising that the contractionary balance sheet effects of devaluations are more than offset by expansionary competitiveness effects. In Chapter 3 I explore the issue of debt choice in more detail with the help of a simple model of debt composition. In the model, firms hedge against exchange rate shocks by altering the currency mix of their liabilities. Within a sample of non-financial Latin American firms I find that firms holding dollar debt are those whose income we expect a priori to be more highly correlated with the exchange rate. I also find evidence that firms holding dollar debt are less credit constrained. Finally, I explore variations across countries and over time in the determinants and levels of dollarization. Within the sample of Latin American corporations I find that exchange rate stability has a significant positive effect on the level of dollarization. All three empirical findings are consistent with a model of debt composition in which credit constraints lead to risk averse behavior by firms Although there is growing evidence of the role of institutions in explaining cross country differences in income per capita, there is still little evidence as to the precise mechanisms by which institutions affect economic outcomes ...