تداوم نرخ ارز واقعی و قواعد سیاست پولی

The objective of this paper is to analyze the effects of alternative monetary rules on real exchange rate persistence. Using a two-country stochastic dynamic general equilibrium with nominal price stickiness and local currency pricing, we will show how the persistence of purchasing power parity deviations can be related to a monetary theory of these deviations. When monetary policy lean against the wind, there is no relationship of proportionality between the time during which prices remain sticky and the persistence of the response of the real exchange rate: in this case high nominal price rigidity is not sufficient, per se, in generating any persistence following a monetary shock. Moreover, we emphasize the role of interest rates smoothing policies and relative price stickiness within countries in understanding the relationship between the real exchange rate and monetary shocks. With reasonable parameters values, a wide range of monetary policy rules can generate real exchange rate autocorrelations around the ones observed in the data.

Is it possible to generate persistence in the real exchange rate without assuming implausible values for nominal price rigidities? Recently several papers have attempted to address the issue of large and persistent deviations of the real exchange rate from purchasing power parity (PPP). As Rogoff (1996) underlines, short-term exchange rate volatility can be explained in terms of monetary shocks while the rate at which PPP deviations damp out may be rationalized if real shocks are predominant. How is it possible to reconcile in a common framework, the enormous short-term volatility of the real exchange rates with the extremely slow rate at which shocks appear to damp out? This is the purchasing power parity puzzle. Models based on sticky prices are one explanation offered for these real exchange rate movements: monetary shocks induce an immediate change of the nominal exchange rate that translates into a change in the real exchange rate as long as national prices adjust slowly. In this spirit, the papers by Bergin and Feenstra (2001), Chari et al. (2002), Kollmann (2001) and Tille (1998) have shown how implausibly long-lasting contracts are needed in order to generate the observed persistence. The purpose of this paper is to show that the persistence of PPP deviations is not necessarily inconsistent with a monetary theory of these deviations. In particular we explore the role of a systematic monetary policy behavior, modelled through interest-rate feedback rules, in rationalizing the observed dynamics of the real exchange rate. We will demonstrate that, under systematic monetary policy, while price stickiness is a necessary condition for monetary shocks to have real effects, high nominal price rigidity is not sufficient, per se, in generating any persistence following a monetary shock. This result implies that, when monetary policy leans against the wind and as long as monetary shocks are serially uncorrelated, the real exchange rate does not exhibit any persistence no matter how high is the degree of nominal price rigidity.1 How is then possible to generate persistence? We highlight the role of two new factors in explaining the observed persistence in the real exchange rate. When monetary policy is conducted in an “inertial” way, i.e. the adjustment of the instrument toward its target is smoothed over time, then the real exchange rate exhibits persistence because, through the interest rate differential, its adjustment is also smoothed over time. Moreover, we emphasize the role of the relative duration of nominal contracts within countries in understanding the relationship between the real exchange rate and monetary shocks. Inertia in relative prices (e.g. in the ratio of imported versus domestically produced goods) can be generated when the expected duration of the corresponding nominal contracts is different. Relative prices’ inertia induces inertia in the inflation rates and this transfers into a persistent behavior of the real exchange rate. After calibrating the model, we illustrate how various rules are able to replicate the autocorrelation observed in the data. We do not assume high nominal price rigidity: the maximum expected duration of a price is of five quarters. For a wide range of rules, we obtain an autocorrelation coefficient that varies around 0.78. These rules are characterized by a smoothing coefficient on the nominal interest rate of 0.85 (at a quarterly frequency). The focus of this paper on the relationship between alternative monetary rules and real exchange rate dynamics is motivated by several reasons. First, recent empirical evidence (see Clarida 1998 and Clarida 2000) has witnessed the adoption of some kind of systematic monetary policy behavior by the major Central Banks. It is then natural to ask how alternative monetary rules affect the dynamics of the real exchange rate, which plays a central role in the transmission mechanism of the Mundell–Fleming framework. If monetary policy reacts to current economic conditions (i.e. it is “endogenous”), then the way monetary policy is designed is important in understanding the dynamics of economic variables following a shock. In particular, if monetary policy is designed in such a way to smooth the behavior of the real economy, monetary policy itself will tend to generate a gradual return to PPP as opposed to a “jump” back to it, once a deviation from PPP is observed. A high degree of nominal price stickiness2 is not needed in order to generate the persistent deviations of the real exchange rate from the PPP. Endogenous monetary policy and nominal price stickiness can interact in a complementary way to generate real exchange rate persistence.3 In a broader perspective, the important lesson of our analysis is that shocks’ transmission mechanisms depend on monetary policy rules’ design. We show that when monetary policy rules react only to deviations of inflation from its target (i.e. inflation targeting instrument rules), the real exchange rate is isolated from productivity shocks and monetary shocks do not have any persistent effect. The orthogonality of the real exchange rate to supply shocks disappears once we consider the so-called “Taylor rule” where the nominal interest rate reacts to deviations of inflation and output from their targets. Still, under “Taylor rules” monetary shocks do not have any persistent effect. As long as the degrees of rigidity are the same across and within countries, monetary policy inertia is necessary to generate real exchange rate persistence following a white-noise nominal shock. In the next section, we briefly summarize the empirical evidence as well as the related theoretical literature. In 3 and 4, we present the structure and the solution of the model. In Section 5, we explore the positive consequences of different rules for the dynamic behavior of the real exchange rate. These results are summarized and analyzed in Section 6. In Section 7 the model is calibrated and some examples are examined while Section 8 concludes.

In this paper we have analyzed the persistence of the real exchange rate when monetary policy is conducted in an endogenous way. Several interesting results have emerged. In contrast to previous contributions (Bergin and Feenstra, 2001; Chari et al., 2002; Kollmann, 2001) we have shown that there is no relationship of proportionality between the time during which prices remain sticky and the persistence of the response of the real exchange rate. In particular if monetary shocks are serially uncorrelated, high nominal price rigidity is not sufficient, per se, in generating any persistence following a monetary shock. We highlight the role of two new factors in explaining the observed persistence in the real exchange rate. An inertial behavior in the real exchange rate can be generated when the degrees of nominal price stickiness vary within countries and when monetary policy exhibits inertia. The model is suitable for different analysis. A welfare analysis of real exchange rate fluctuations is the next step. In terms of real exchange rate dynamics it can be used to fully examine the volatility of the real exchange rate across different monetary policy rules and to examine the relation between volatility and persistence across different monetary regime and exchange rate regimes (see Baxter and Stockman, 1989; Flood and Rose, 1998). The relative importance of real versus monetary shocks can also be analyzed within this framework. Moreover the prediction of the model can be compared with international real business cycle statistics.