The question of the impact of EMU on the stability of the transatlantic exchange rate raises the more general question of whether the exchange rate is a useful adjustment instrument or an additional source of shocks. The end of the Bretton Woods system was motivated by the hope that flexible exchange rates would isolate the economies from shocks coming from their partners', and help them to face domestic shocks. This hope was largely shattered by the experience of the post-Bretton Woods system. In particular flexible exchange rates did not translate into a reduced instability of other macroeconomic variables (Flood and Rose, 1995). Hence, fixing the intra-European exchange rates may not lead to more instability elsewhere and specifically on the transatlantic exchange rate.

In this paper, we compare the role of the intra-European exchange rate as an instrument for economic stabilisation to its role as a source of economic instability, in order to infer the potential impact of EMU on the transatlantic exchange-rate variability. To do so, we estimate a simple, three-country model for the United States, Germany and France, over the 1972-1995 period. The structure of each economy is assumed to be independent of the exchange rate regime: a general floating regime, EMU and the ERM. Regimes are defined along the way European central banks settle the nominal interest rate. In the general floating regime, each follows strictly domestic objectives. In the ERM, France follows the German interest rate. In EMU, the single European interest rate in settled with respect to the averages of French and German objectives. Stochastic simulations are performed in order to compare the variability of various macroeconomic variables, including the transatlantic exchange rate, in the three regimes, and to highlight the role of the intra-European exchange rate as a source of shocks or as an adjustment variable.

Our simulations show that EMU could reduce the variability of the transatlantic exchange rate compared both to the ERM and to a floating regime. Eliminating the shocks to the intra-European exchange rate is crucial for the stabilisation of the European economies, as suggested by Minford et al. (1992). However, EMU stabilises the transatlantic exchange rate even if the removal of shocks to the intra-European risk premium is not attributed to the regime shift. By contrast, the ERM is the regime producing the most unstable transatlantic exchange rate because it is an asymmetric regime.

Due to structural and stochastic asymmetries, the benefits of EMU are smaller for France than for Germany, in terms of the variability of inflation and of the real effective exchange rate. Finally, EMU is the regime producing the largest instability in the US economy, because it eliminates the stabilising fluctuations of the transatlantic exchange rate.

More generally, our simulations show that the transfer of volatility is not systematic and can be indirect. Here, fixing the intra-European exchange rate does not increase the variability of the extra-European exchange rate, but it destabilises a third economy.