Abstract:

This dissertation contains three papers about exchange rates. In Chapter 1, I develop a model of trade with monopolistic competition, heterogeneous firms, and endogenous markups, to study the implications of firm reallocations and average productivity changes on exchange rate pass-through to import prices and trade flows. I obtain that pass-through elasticities are strictly decreasing with firm productivity. For an individual exporting firm, the exchange rate pass-through to the import price (in the importer's currency) reflects two opposite forces: an industry-wide competition effect, driven by the response of the productivity cutoff rule; and a firm-specific effect, driven by the relative position of a firm with respect to the cutoff point. Both effects produce asymmetric responses of import prices (at the firm level) to appreciations and depreciations. With respect to aggregate prices, I find that due to firm reallocations, average import prices and trade flows are disconnected. That is, average import prices, which are directly affected by changes in the extensive margin of trade, are poorly suited to derive conclusions about the expenditure-switching effects of exchange rate changes. I then extend the model to a general equilibrium New Open Economy Macroeconomics setup with sticky wages. The general equilibrium model replicates the results of the partial equilibrium model for an endogenous exchange rate change coming from an unexpected monetary shock. However, the GE model is solved using a first-order linearization, so it loses the second order effects (reflected in asymmetric responses) of the partial equilibrium model.

In Chapter 2, I look further into the issue of asymmetric exchange rate pass-through for appreciations and depreciations. I review previously identified sources of asymmetric pass-through (market share considerations and capacity constraints) and identify three more sources (importer's buyer power and the two obtained in Chapter 1). Then, using disaggregated data from a retailer in the Chicago area--wholesale prices for 63 brands of imported beer from 7 countries--I find statistically significant asymmetric responses of wholesale prices to exchange rate changes. In the short-run, wholesale prices for US beer imports respond more to appreciations than for depreciations of the US dollar. The use of brand level data prevents aggregation biases coming from heterogeneous sectors (as in Imbs, Mumtaz, Ravn, and Rey (2005)) or from inter-firm reallocations due to firm heterogeneity (as in Chapter 1).

Finally, in Chapter 3, co-authored with Carlos Felipe Lopez Suarez, we study whether the nonlinear behavior of the real exchange rate can help us account for the persistent lack of predictability of the nominal exchange rate. We construct a smooth nonlinear error-correction model that allows us to test the hypotheses of nonlinear predictability on the nominal exchange rate and nonlinear behavior on the real exchange rate in the context of a fully specified cointegrated system. Using a panel of 19 countries, we show that standard t -statistics reveal strong evidence of nonlinear predictability of the nominal exchange rate while out-of-sample U -statistics show a higher forecast precision of the nonlinear model than the one obtained with a naive random walk specification. Our results are both encouraging and limited. They are encouraging since we find a simple parsimonious model based in economic fundamentals that can persistently beat the random walk across countries, numeraires and horizons. They are also limited in the sense that the improvements in forecast precision are modest and therefore generally not significant at 5 or 10 percent levels.