Essays in banking and risk management

Abstract

(cont.) Risk Management have begun implementing strategies to provide commodity price and weather insurance in the developing world. In Chapter 3 (joint with Professor Rob Townsend from the University of Chicago), we examine how shocks to the price of rubber, an important but volatile Thai export commodity, affect the income, consumption and intra-household remittances of rural Thai households. In contrast to related work on rainfall shocks, we find rubber price innovations are not well insured or smoothed--remittances, borrowing and saving play only small roles in ameliorating the effect of these shocks on the consumption of affected households. We argue that differences in the relative persistence of the two types of shocks provide a plausible reason for these divergent findings, drawing on the literature on buffer stock models of consumption behavior and risk sharing with limited commitment.

This thesis consists of three essays at the intersection of banking, corporate finance and macroeconomics. Unifying the essays are two themes: firstly a focus on how firms (Chapter 1 and Chapter 2) and individuals (Chapter 3) insure against, and react to, sources of macroeconomic risk; secondly the role of financial institutions in the transmission of macroeconomic shocks. Turning to specifics, Chapter 1 is a theoretical and empirical examination of risk management behavior amongst small and medium sized firms, in particular firms' choices between fixed and adjustable rate loan contracts. (Although theory suggests small, privately held firms should have strong incentives to engage in risk management, such firms are rarely studied in empirical work.) I develop a simple agency model of risk management behavior, and then present several pieces of empirical evidence that suggest small US firms do use the banking system to help manage interest rate risk, based on microeconomic data on bank dependent US firms. Chapter 2 presents evidence that banking relationships are most valuable to firms during periods of tight credit, in the extreme during a 'credit crunch'. Relationships alleviate delegated monitoring costs; when banks are credit constrained, these costs are extreme, so the informational advantage of relationships is magnified. I develop these intuitions using a simple agency model. Empirical evidence, based on data from a survey of manufacturing firms during the Asian financial crisis, supports the thesis. Several pieces of evidence also suggest my empirical results are not driven by the endogenous nature of bank relationship formation. Financial institutions in co-operation with the World Bank and the International Taskforce on Commodity