In this thesis we investigate the effects of real exchange rate (RER) uncertainty on output in the context of Southern Cone economies. The first chapter provides a framework to analyze the output effects of RER uncertainty when firms contract dollar-debt and no hedging instruments are available, by focusing on the channel uncertainty-output operating through the firms financial strategy. An increase in uncertainty increases the probability of bankruptcy, raising expected marginal bankruptcy costs, and reducing optimal output of a risk-neutral firm. We find the output response to uncertainty shocks to depend on firms' liquidity balances, trade orientation and perceptions about government assistance if large exchange rate movements occur. The second chapter examines empirically RER uncertainty effects on sectoral output for 28 manufacturing sectors in the Southern Cone over 1970-2002. We use alternative uncertainty measures allowing different degrees of sophistication in agents' expectation mechanisms to estimate a supply function. We use instrumental variable techniques to address potential simultaneity problems. Results suggest a negative non-negligible effect of uncertainty on output, threshold effects, and sectoral heterogeneity, explained by trade orientation, the intensity with which sectors trade within Mercosur and by sectoral productivity. The fourth chapter investigates the importance of past exchange rate behaviour when forming expectations and tests for the uncovered interest parity (UIP) hypothesis in Uruguay. Using interest rate differentials over 1980-2010 we identify a strong extrapolative component in expectations, following an inverted-U pattern over time. Agents internalise announcements and shocks that may affect fundamentals. Deviations from UIP are low for high-inflation periods, and highfor low-inflation periods and freely floating regimes. As long as what it takes to predict well is simple (look backwards, follow announcements), interest rate differentials perform well. Once exchange rate determination becomes intricate, agents fail at predicting. This finding remains unchanged when survey data are used for the period 2005-2010.
| Identifer | oai:union.ndltd.org:bl.uk/oai:ethos.bl.uk:554562 |
| Date | January 2011 |
| Creators | Varela, Gonzalo |
| Publisher | University of Sussex |
| Source Sets | Ethos UK |
| Detected Language | English |
| Type | Electronic Thesis or Dissertation |
| Source | http://sro.sussex.ac.uk/id/eprint/7345/ |
Page generated in 0.0021 seconds