برخی از شواهد تجربی در مورد اثرات شوک سیاست پولی ایالات متحده در نرخ ارز متقاطع

This paper examines the impact of U.S. monetary policy shocks on the cross exchange rates of sterling, yen and mark. The main finding of the paper is a ‘delayed overshooting’ pattern for all currency cross rates examined (sterling/yen, yen/mark and mark/sterling) following an unexpected U.S. monetary policy change, which in turn generates excess returns. We also provide evidence that the ‘delayed overshooting’ pattern in cross exchange rates is accompanied by asymmetric interventions by central banks in the foreign exchange markets under consideration triggered by a U.S. monetary policy shock.

Following the classic Dornbusch (1976) ‘overshooting’ model on the impact of monetary policy on exchange rates and interest rates in the presence of price stickiness, a large empirical literature has focused on the assessment of these effects.1 The standard theoretical framework predicts that a monetary tightening leads to an immediate exchange rate ‘overshooting’. Several empirical papers have focused on the impact of U.S. monetary policy on interest rates and exchange rates vis-à-vis the U.S. dollar, and have established that the peak timing of this response occurs one to three years after rather than immediately (‘delayed overshooting’). This accumulated evidence creates in turn a ‘conditional forward premium puzzle’ with opportunities for excess returns by borrowing abroad and investing in the U.S. Although the standard ‘forward premium puzzle’ is well documented in empirical studies of the foreign exchange market (Froot and Thaler, 1990 and Hodrick, 1987), the puzzle now arises conditional on an exogenous change in U.S. monetary policy. In particular, Eichenbaum and Evans (1995) have shown that, in response to a tighter U.S. monetary policy, the dollar exhibits a ‘delayed overshooting’ behaviour of two to three years vis-à-vis the major currencies, a pattern that is confirmed by Clarida and Gali (1994). ‘Delayed overshooting’ is also assessed by Evans (1994), who uses weekly data and finds that the dollar overshoots with a delay of two to three years vis-à-vis the mark and the yen, and Lewis (1995), who finds that the dollar appreciates against the mark and the yen for the first five months after the monetary policy shock. Eichenbaum and Evans (1995) also offer empirical evidence that after a contractionary U.S. monetary policy shock, domestic interest rates rise relative to foreign interest rates and the dollar appreciates. Kalyvitis and Michaelides (2001) adopt a specification with relative output and prices in order to capture the relative business cycle position of the U.S. and the domestic country; this approach solves the ‘delayed overshooting’ effect, but the authors report persistent deviations from the uncovered interest rate parity hypothesis. The ‘delayed overshooting’ and the ‘conditional forward premium puzzle’ are specific examples of how economies might be influenced by monetary policy abroad. However, the size of these impacts is likely to vary between countries due, for instance, to discrepancies in the structural characteristics of the economy, differences in nominal rigidities and/or asymmetries in the monetary policy functions of foreign countries driven by differential responses to deviations of inflation or output from the respective targets. A challenging empirical issue is therefore the potential asymmetry of ‘delayed overshooting’ between cross exchange rates, i.e. whether the magnitude of, say, the dollar appreciation against major currencies following a U.S. monetary policy tightening differs. This question was originally posed by Frankel (1986), but has only recently received some attention in the study by Lobo, Darrat, and Ramchander (2006) who examine the impact of changes in the Federal Funds Rate (FFR) target on daily dollar exchange rates and find that surprises associated with monetary contraction have a larger effect compared to monetary easing for the sterling, mark and Canadian dollar rates, whereas the opposite holds for the yen. Taking ‘delayed overshooting’ as a stylized fact, in this paper we first attempt to investigate how and to what extent cross exchange rates of major currencies (sterling, Japanese yen and German mark) respond to U.S. monetary policy shocks. In this vein, we define ‘delayed overshooting’ as a change in the three cross exchange rates examined, which is triggered by a contractionary U.S. monetary policy shock, and gradually continues for some months or quarters after the shock. We also assess the response of interest rate differentials between the three countries (UK, Japan and Germany) and the existence of excess returns in these economies. To this end, we use the Structural Vector Autoregression (SVAR) approach that has been routinely used in the recent literature, in which we proxy U.S. monetary policy changes by the Romer and Romer (2004) narrative index (RR). This measure of U.S. monetary policy is better equipped to overcome the endogeneity and anticipatory biases that are inherent in other standard indices, such as the FFR or the non-borrowed reserves ratio. We find a ‘delayed overshooting’ pattern for all the currency cross rates examined (sterling/yen, yen/mark and mark/sterling) when the Federal Reserve tightens U.S. monetary policy. Specifically, the mark appreciates with a delay relative to the yen with a peak at around ten months following the U.S. monetary policy shock, whereas the mark appreciates against the sterling, and the sterling against the yen with a peak at around six months after the shock. We find that the responses of exchange rates and interest rates following the U.S. monetary policy tightening leave room for excess returns in the three foreign exchange markets under consideration. Next, we take our analysis one step further by attempting to explore whether the driving force of ‘delayed overshooting’ in the cross exchange rates has been asymmetric foreign exchange interventions by central banks (that is, interventions following a specific foreign monetary policy shock, which differ in size and/or direction). Our investigation is motivated by several studies that have investigated the repercussions of foreign exchange intervention on the exchange rate following a monetary policy shock. Lewis (1995) has reported that central banks intervene to support a monetary policy that is consistent with their exchange rate targets, as monetary policy changes driven by domestic targets may trigger counter-movements in the exchange rate. Bonser-Neal, Roley, and Sellon (1998) have claimed that interventions may at times react to past monetary policy actions, e.g. if U.S. monetary authorities wished to keep the exchange rate path within a particular zone. They find that not only various policies were evident at different periods, but also that the exchange rate responses to FFR changes are unaffected when one controls for central bank foreign exchange intervention. Kim, 2003 and Kim, 2005 has examined the joint effects of monetary policy and foreign exchange intervention on the U.S.–Canada exchange rate and has found evidence that ‘delayed overshooting’ can be explained by interventions in the foreign exchange market intended to weaken the exchange rate appreciation following a monetary policy tightening. Given these points, we use recently published foreign exchange intervention data to assess the potential impact of central bank intervention in the foreign exchange markets under consideration following a U.S. monetary policy shock. Our evidence shows that the ‘delayed overshooting’ pattern in cross exchange rates is accompanied by asymmetric foreign exchange interventions. The evidence points towards a ‘leaning-with-the-wind’ intervention policy by the Japanese monetary authorities, a finding that is in accordance with an effort to drive the yen to its new equilibrium against the dollar following a U.S. monetary policy tightening. In contrast, we find evidence of a ‘leaning-against-the-wind’ intervention by the German monetary authorities, which is consistent with a strengthening of the mark against the dollar following a U.S. monetary policy shock. The paper makes two contributions in the relevant literature. First, the ‘delayed overshooting’ pattern of cross exchange rates has, to our knowledge, been unnoticed in existing empirical studies on the monetary policy transmission mechanism. Given the robustness of our empirical results, this pattern is likely to have comprised a substantial factor of exchange rate variability in the floating rate era. Second, the paper points out a source of exchange rate variability triggered by asymmetric central bank interventions, which offers a route for further explorations of the role of central banks in excess exchange rate movements, a role that is now conditional on an external monetary policy shock. The rest of the paper is structured as follows. Section 2 outlines the empirical methodology and the data utilized. Section 3 assesses the impact of the monetary policy shocks using a SVAR model, while Section 4 develops robustness analysis. Section 5 explores the role of foreign exchange interventions. Finally, Section 6 concludes the paper.

This paper has investigated the effects of U.S. monetary policy shocks on foreign interest rates and cross exchange rates. Building upon the spirit of Eichenbaum and Evans (1995) and using a SVAR approach, we first attempted to assess the impact of unanticipated U.S. monetary policy on foreign interest rates and cross exchange rates in the UK, Japan and Germany. We provided evidence in support of ‘delayed overshooting’ in all the currency crosses examined (sterling/yen, yen/mark and mark/sterling) accompanied by excess returns in the foreign exchange markets considered, whereas a non-negligible portion of exchange rate variability between the currency crosses can be attributed to the U.S. monetary policy shock. We then investigated the possibility that ‘delayed overshooting’ in cross exchange rates is triggered by central bank interventions in the corresponding foreign exchange market following the U.S. monetary policy shock. Our evidence indicated the presence of asymmetric foreign exchange interventions, which are likely to have generated ‘delayed overshooting’ effects in the cross exchange rates examined. Thus, our paper stresses another aspect of asymmetry in central bank behaviour during the floating exchange rate era, namely the possibility that exchange rate fluctuations were generated by foreign exchange interventions following external monetary policy disturbances. We close the paper by noting that here we focused on the international effects of monetary policy that mainly arise through the exchange rate channel and work through the uncovered interest rate parity condition. These effects may be exacerbated by the presence of asymmetries in the monetary policy functions of foreign countries, like differential responses to deviations of inflation or output from the respective targets. To sharpen our understanding of the implications generated by the conduct of foreign monetary policy, we delegate a more in-depth analysis of the structural characteristics of domestic monetary policy responses following external shocks to future research.