This paper studies the sectoral decomposition of the volatility of real exchange rates. The real exchange rate between two countries is the relative price of a representative goods basket. The sectoral decomposition of real exchange rate fluctuations is important, because it has important implications for the dynamic adjustment of an open economy to exogenous shocks. For some countries, the movements in relative price levels come from the relative prices of internationally nontraded goods such as housing or construction, while for others from those of traded goods such as manufactures. The empirical part in the next section finds that, while in general the relative prices of traded goods are the most important in driving real exchange rate movements, the relative prices of nontraded goods are relatively more important for the country pairs that maintain stable nominal exchange rates.
To explain the empirical evidence, I construct a general equilibrium monetary model with heterogeneous productivity and endogenous tradability. The model shows how real exchange rate dynamics are connected to shifts in the tradability of goods through firms’ price setting behavior. In some cases, the shifts and the simultaneous movements in nominal exchange rates lead to strong substitution effects effects among traded goods. In these cases, the relative prices of traded goods dictate the movements in real exchange rates. Therefore, limiting flexibility in nominal exchange rates can delay the adjustment in the relative prices of traded goods and raise the contribution to the volatility of the relative prices of nontraded goods, as observed in the data. My model offers a set of reasons why the relative price of nontraded to traded goods may be the major source of fluctuations in real exchange rates. The model can be extended to perform welfare analysis under various trade structures. Clearly, it also has strong implications for a design of exchange rate policy. These are the contributions of my paper.
In principle, we can decompose the fluctuations in real exchange rates into their traded and their nontraded goods components. The traded component is deviations from the law of one price for traded goods, while the nontraded component is fluctuations in the relative prices of nontraded to traded goods across countries. The traded component is by far dominant for the real exchange rates among the OECD countries that allow their currency to move freely (Engel 1999; Obstfeld 2001). To the contrary, the nontraded component is significantly more important in international data (Betts and Kehoe 2001a). The relative volatility of the nontraded real exchange rates and the overall volatility is 30 percent on average.
The number becomes much higher as the volatility of real and nominal exchange rates falls. Such a pattern is consistent with what was found in the study on Mexico’s real exchange rates by Mendoza (2000). As I present in the next section, the volatility of the nontraded real exchange rates even exceeds its traded counterpart and the overall volatility in some cases. In particular, the nontraded component of the real exchange rates between several member countries of the European Monetary System has displayed substantial volatility. In light of these empirical features, there seem to be linkages between the exchange rate regime and the sectoral decomposition of real exchange rate fluctuations.
Download
Real Exchange Rate Fluctuations and Endogenous Tradability
