Essays in international finance

Thesis (Ph. D.)--Massachusetts Institute of Technology, Dept. of Economics, 2003.Includes bibliographical references.This thesis is a collection of three empirical essays in international finance. The first chapter studies the transmission of monetary policy through the lending channel in a partially dollarized banking system. Taking advantage of the cross-sectional and time-series variation in the individual Mexican bank balance sheets, I find that the deposits and loans of banks with a larger share of foreign deposits are less sensitive to domestic monetary shocks, particularly for small banks. This result is reinforced when foreign monetary shocks and country risk shocks are controlled for. The results also suggest that banks with a larger foreign deposit share are more sensitive to foreign (U.S.) monetary shocks. Finally, these banks are more sensitive to country risk. That is, they are more prone to lose deposits when Brady bond spreads increase, although the size of their loan portfolio is not reduced. The second chapter examines whether bank credit fuels asset prices, using evidence from the Japanese real estate boom during the 1980's. The decline in banks' loans to keiretsu firms is used as the shock to bank real estate credit. The evidence supports using keiretsu loans as an instrument. Financial deregulation allowed large firms to replace bank finance with financing from public markets. The main part determines that those prefectures that experienced a larger loss in their banks' proportion of keiretsu loans experienced a positive increase in real estate lending which fuelled land inflation. An increase of 0.01 in a prefecture's instrumented share of real estate loans for 3 years implies a 28 % higher land inflation rate. The third chapter evaluates the behavior of sovereign credit ratings. This chapter questions the view that credit rating agencies aggravated the Asian crisis by excessively downgrading those countries. I find that ratings are, if anything, sticky rather than excessively procyclical. Assigned ratings exceeded predicted ratings prior to the crisis, mostly matched predicted ratings during the crisis period, and did not increase as much as predictions in the recent period following the crisis. Ratings are also found to react to nonmacroeconomic factors, lagged spreads and default history.by Nada Mora.Ph.D