ABSTRACT
Does the currency regime matter? The main industrial countries have experienced a wide range of exchange-rate arrangements in the last two centuries, ranging between the two polar systems of fixed and flexible exchange rates. Before the establishment in 1871 of the international gold standard, effectively a system of fixed exchange rates, the prevailing arrangement was the bimetallic system, based on the relative price of gold and silver. After the demise of the gold standard and the associated turmoil in the international financial system between the two World Wars, the Bretton Woods system was chartered in 1944. A system of fixed but adjustable exchange rates, it was abandoned in the early seventies in favor of a system of generalized floating exchange rates. In the late seventies, several European countries started the process that culminated with the launch of the euro in January 1999. The same variety of exchange-rate regimes that we observe through time for
an individual country also exists today across countries. In 1998, according to the IMF taxonomy (IMF, 1998), 66 countries unilaterally pegged either to a single or to a composite currency, 17 adhered to an exchange-rate regime with partial flexibility, and the remaining 101 followed more flexible arrangements, such as managed or independent floating. Even among the more homogeneous group of 33 OECD and newly industrialized countries, 11 were on the verge of sharing a single currency, the euro; four were pegged; and the remaining ones followed arrangements of independent and managed floating. Against this backdrop, it is not surprising that since the early important contri-
butions of Friedman (1953) and Mundell (1961, 1963), the above questions have ranked high in the international finance research agenda. Cooper (1999) goes as far as arguing that for many countries “the choice of exchange rate policy is probably their single most important macroeconomic policy decision.” This chapter studies the welfare effects of fixing the exchange rate from a quan-
titative viewpoint. We analyze a two-country equilibrium business cycle model, featuring nominal rigidities and deviations from the law of one price, due to firms pricing-to-market. In this now rather standard class of economies, real effects stem from both systematic and nonsystematic components of monetary policy. This is consistent with the consensus view that, speculative attacks aside, the
exchange-rate arrangement affects real economic variables only if the latter are influenced by a systematic component of monetary policy.1 Dedola and Leduc (2001) showed that a realistically calibrated model with such building blocks can also quantitatively account for some key stylized facts regarding the real and allocative consequences of the choice of the currency regime. In particular, the model could reproduce the findings of Baxter and Stockman (1989) and Flood and Rose (1995) that, among the statistical properties of most macroeconomic variables, only the volatility of the real and nominal exchange rates has dramatically changed after the fall of the Bretton Woods system. We explore the welfare costs of an exchange-rate peg relative to a float, when
central banks attempt to stabilize their economy by following forward-looking interest-rate rules. We find that, although the welfare differences across exchangerate regimes are small, a flexible exchange-rate system is still preferred to a currency peg. Our result is driven by the fact that, under the flexible exchangerate system, the central bank, via its interest-rate policy, is able to dampen the movements in output and, therefore, the volatility of employment. This does not occur under a currency peg. In this case, the central bank foregoes its interest-rate rule and sets monetary policy to keep the value of the currency fixed. Output and employment are more volatile under this regime, as a result. Since agents are risk averse, they prefer the relatively more stable employment/leisure path brought about when the currency floats. The last step in our analysis is motivated by the observation that some spec-
ifications of the interest-rate rule may lead to monetary instability. Following a recent strand in the macroeconomic literature (e.g. Clarida et al., 2000), we focus on the ability of the monetary policy rule and the exchange-rate regime to provide a credible nominal anchor and rule out bad outcomes. The idea is that the lack of credibility can result in the adoption of a monetary rule that may enable inflation expectations to become self-fulfilling (because of the indeterminacy of the steady state equilibrium). The economy may therefore fluctuate due to movements in non-fundamental shocks (or sunspots), which may reduce welfare. We show that if inflation expectations are self-fulfilling, a country would be better off by fixing its currency. The choice of exchange-rate regimes has recently been studied by Chinn and
Miller (1998) – in a flexible-price environment – andDevereux and Engel (1998) – who studied the impact of the price setting on the optimal exchange-rate regime with uncertainty arising frompermanentmoney supply shocks. Gali andMonacelli (1999) showed that the optimal monetary policy for a small, open economy entails somemanagement of the exchange rate. Finally, Devereux (1999) andObstfeld and Rogoff (2000a) derive the optimal Ramsey monetary policy in tractable versions of two-country models with predetermined prices. The remainder of the chapter is organized as follows. Section 8.2 briefly lays
down the structure of the model, whose properties were fully explored in Dedola and Leduc (2001), and Section 8.3 presents the calibration. In Section 8.4, we first explore the impact of the exchange-rate regime on the volatility of macroeconomic variables. We then compute the welfare consequences of the currency regime and
briefly assess the advantages of tying one’s hands, via an exchange-rate peg, when monetary policy is not credible and let inflation expectations become self-fulfilling. We assess the sensitivity of our results before concluding.
