Essays on the dynamics of the U.S. real exchange rate

The existing new open-economy macroeconomic literature is almost entirely developed based on the sticky price model. However, the sticky price framework has received both theoretical and empirical criticism over the years. Recently, the sticky information model is introduced as an alternative by addressing some of the concerns with the sticky price model. While developed within a closed economy framework, such model has not yet been extended to the research in international macroeconomics. Therefore, this thesis primarily aims to explore the power of the sticky information model in replicating and explaining some stylized facts of the U.S. real exchange rate dynamics. A structural vector autoregressive (VAR) analysis in Chapter 2 detects (1) declining and insignificant impact of monetary policy on the U.S. real exchange rate and other variables and (2) disappearance of hump-shape real exchange rate response to monetary policy shock during the great moderation (starting from the mid-1980s). Given this background, a two-country sticky information dynamic stochastic general equilibrium (DSGE) model with public misperception on the true nature of monetary policy shock is constructed to explore the possible sources of such structural change. Sensitivity tests show that less persistent monetary policy shock and smaller price stickiness are two key factors contributing to the weaker real exchange rate response, while the hump-shape response (i.e. delayed overshooting) is a phenomenon associated with public misperception. Historically, the fluctuations in the real exchange rate are highly persistent, commonly known as the purchasing power parity (PPP) puzzle. This puzzle is investigated in Chapter 3 using a two-country sticky information DSGE model with tradable and non-tradable goods. Highly persistent PPP deviations are reproduced in the benchmark model with persistent productivity shocks, nonpersistent monetary policy shock and fairly flexible prices. Sensitivity tests show that persistent monetary policy shock can also generate such persistent deviations with conventionally used price stickiness in existing literature, which is fairly high. It is shown analytically that the sticky information model can generate high persistence in the variables without much difficulty as all agents are allowed to make adjustments in each period and thus can fully respond to shocks in their information sets. Therefore, given a persistent shock, the pass-through of the shock and its persistence to various macroeconomic variables can be fairly high in the sticky information model. In Chapter 4, structural vector error-correction model (SVECM) is used to identify the sources of the U.S. real effective exchange rate fluctuations during the great moderation. Time series of other variables are constructed against an aggregate of industrialized countries. Three long-run relationships are found in the data, one of which states that the U.S. real exchange rate appreciates with higher relative output. Subsequent structural form analysis shows that only relative productivity shock has a significant long-run impact on the U.S. real exchange rate. Forecast error variance decomposition also shows that while transitory shocks and relative fiscal policy shock account for more than two thirds of short-run movements in the U.S. real exchange rate, relative productivity shock dominates its long-run fluctuations. Again, the effect of relative monetary policy shock on the U.S. real exchange rate is found to be insignificant in this period.DOCTOR OF PHILOSOPHY (HSS