رژیم نرخ ارز و چرخه های کسب و کار بین المللی

This paper investigates the impact of exchange rate regimes on international business cycles and focuses on the consequences of membership to the European Monetary System. The volatility puzzle uncovered by Baxter and Stockman [1989, Journal of Monetary Economics 23, 377–401] after assessing the consequences of the Bretton Woods system turns out to be a robust stylized fact: real and nominal exchange rates display a higher volatility under floating rates while the variability of macroeconomic quantities remains unchanged across exchange rate regimes. Besides, there is evidence that fixed rates are associated with enhanced comovement in output, consumption and investment. We find that a two-country model, featuring monopolistic competition, pricing-to-market, and price stickiness, captures all of the empirical features of the data but one; namely the stronger output comovement following the transition to fixed rates.

There are two well-established empirical regularities that describe the relationship between exchange rate regimes, the volatilities of exchange rates and macroeconomic quantities, and the cross-country correlations of macroeconomic quantities. First, Mussa (1986) reports that the volatility of real and nominal exchange rates is substantially lower under fixed than under flexible exchange rates. In contrast, the variability of macroeconomic quantities is unchanged across regimes (Baxter and Stockman, 1989; Flood and Rose, 1995). Second, there is no evidence that cross-country correlations ofmacroeconomic variables rise under flexible exchange rates, relative to fixed exchange rate periods (Baxter and Stockman, 1989). These facts are puzzling once we compare them to the predictions of most modern theoretical models of exchange rates. In Obstfeld and Rogoff’s (1995a) seminal paper, the volatility of macroeconomic quantities is linked to the volatility of real and nominal exchange rates. Indeed, a home monetary shock implies a depreciation of the home currency. If we consider the response of the domestic economy to this depreciation within Obstfeld and Rogoff’s (1995a) framework, where all goods satisfy the law of one price and the Purchasing Power Parity holds, then the depreciation makes home goods (foreign goods) less expensive (more expensive) for foreign households (for domestic households). As a result, the demand for home goods, and hence home aggregate output, increases. The exchange pass-through to consumer prices introduces a close relationship between fluctuations in the nominal exchange rate and changes in quantities. Cross-exchange rate regime evidence refutes this idea. Finally, many traditional models of international exchange predict that the degree of insulation of an economy from shocks in trading partners’ economies should rise under flexible exchange rates, as a direct consequence of nominal exchange rate and current account adjustments. The data suggest, however, that cross country correlations of macroeconomic aggregates are no lower—and may be actually higher—under flexible than under fixed exchange rates. The contribution of this paper to the literature on the relative performance of fixed versus floating exchange rate regimes is twofold.

This paper has examined international evidence on the question of whether business cycle properties of some key macroeconomic variables have changed across exchange rate regimes. While previous studies have focused on measuring the consequences of the fall of the Bretton Woods System, this paper provides empirical evidence on the EMS case. First, we found that salient features of the data reported by Mussa (1986) and Baxter and Stockman (1989) also hold for pre- and post-EMS periods. The volatility of real and nominal exchange rates is significantly lower under fixed than under flexible exchange rates with no corresponding changes in the variability of macroeconomic quantities. Secondly, in contrast to former investigations, our empirical research suggests that membership to the EMS is associated with higher cross-country correlations of output, consumption and investment. In the second part of the paper, we gauge the ability of the sticky price PTM model to account for these stylized facts. The model accounts for all of the empirical regularities but one; namely the stronger output co-movement, which follows the transition to fixed rates. Bayoumi and Taylor (1995) may indicate the route to the resolution to this remaining puzzle. After using VAR à la Blanchard and Quah (1989) to identify supply and demand shocks in EMS and non-EMS countries, they report that the intra-EMS output response function correlations rose significantly between the pre- and post-EMS periods. Furthermore, they uncover a marked reduction in the speed of output response to shocks in EMS countries during the post-1979 era. This last result is consistent with the idea that, in fixed exchange rate regimes, as prices and wages are not flexible enough to compensate for the loss of exchange rates, and given the limited labor mobility in Europe, the burden of adjustment mainly falls on real variables such as output and employment. As the hiring and firing process is costly and time-consuming, fixed exchange rates might be associated with more elongated employment, thus output, responses to shocks. Bayoumi and Taylor’s (1995) findings suggest that taking into account the rigidity of the European labor market might improve our understanding of output dynamics across exchange rate regimes.