Essays on Empirical Macroeconomics and Finance

Abstract

The first chapter investigates the risk-shifting agency problem for banks when faced with high competition. Protected by limited liabilities and insured deposits, banks faced with high competition may have the incentive to overly invest in risky assets. This chapter tests the risk-shifting hypothesis of competition by exploiting an exogenous variation in house price volatility. Using loan-application level data for the U.S. mortgage market from 2000 to 2005, I find that banks in high competitive county-level mortgage markets lowered lending standards (e.g., loan-to-income ratio, denial rate) by twice as much as those in less competitive markets in response to volatility in the local house price. Such competition-risk taking pattern was associated with worse economic outcomes after the crisis, including higher foreclosure rate, higher bank failure rate and higher unemployment rate in the non-financial sector.

The second chapter studies whether the increase in inter-state capital flows in the U.S. has substituted labor mobility in explaining the puzzling secular decline in inter-state labor mobility over the past three decades. The identification strategy is to use inter-state bank branching deregulation since the late 1970s and early 1980s as the exogenous shocks to inter-state capital mobility. Furthermore, using staggered bilateral bank branching deregulation between U.S. states, the exact substitution effect can be tested by looking at how labor between states moved before and after. Empirical results found in this chapter suggest that financial market integration and greater capital mobility can explain a significant fraction of the decline in inter-state labor mobility.

The third chapter examines the fire-sales activities in the U.S. corporate bond market by insurance companies after a severe climate shock, Hurricane Katrina. The chapter tests whether property and casualty insurance companies unexposed to the shock also participated in the fire sales due to mark-to-market accounting rules. Empirical results are found to support this hypothesis and suggest that there was a negative price impact associated.