تا به مکانیزم انتقال سیاست پولی اروپا درحالت اجرا به اتحادیه پولی اروپا تغییر کرده است؟
کد مقاله | سال انتشار | تعداد صفحات مقاله انگلیسی |
---|---|---|
25975 | 2006 | 40 صفحه PDF |
Publisher : Elsevier - Science Direct (الزویر - ساینس دایرکت)
Journal : European Economic Review, Volume 50, Issue 3, April 2006, Pages 737–776
چکیده انگلیسی
This paper studies empirically the transmission mechanism of European monetary policy by means of time-varying, heterogeneous coefficient models estimated in a numerical Bayesian fashion. Based on pre-European Monetary Union evidence from Germany, France, Italy, and Spain, we find that (i) the long-run cumulative impact on output of a common, homoskedastic monetary policy shock has decreased in all countries after 1991. These declines are statistically significant and accompanied by some changes in the conduct of monetary policy over the same period. At the same time, we also find that (ii) cross-country differences in the effects of this shock have not decreased over time
مقدمه انگلیسی
Theory (e.g., Lucas, 1976), as well as a growing body of empirical evidence on the U.S. post-war monetary history – e.g., Boivin and Giannoni, 2002 and Boivin and Giannoni, 2003, Cogley and Sargent (2004), and Primiceri (2004), among others – suggest that the transmission mechanism of monetary policy may change in response to expected or actual changes in the monetary policy regime. The European Monetary Union (EMU) was launched in January 1999, and the euro started to circulate in January 2001 (almost a decade after the adoption of the Maastricht treaty), replacing the exchange rate mechanism that had anchored the European Monetary System (EMS) for more than 20 years. It is thus natural to ask what has happened to the transmission mechanism of European monetary policy in the run-up to the EMU. At the same time, a fairly large, albeit sometimes contradicting, body of empirical evidence based on pre-EMU data points to the presence of significant differences across countries (or heterogeneity) in the transmission mechanism of monetary policy in Europe – see Guiso et al. (1999) and Angeloni et al. (2003) for recent surveys of this literature. On the one hand, trade and financial integration may increase, and business cycles may become more synchronized, as a result of the currency union (Frankel and Rose, 1998). Hence, these differences might decrease over time if due to these factors. On the other hand, they could also persist for a long time if due to differences in the financial structures rooted in the legal frameworks of individual countries (Cecchetti, 1999). Thus, it would also be useful to have some idea on the time profile of these differences, but there is no hard evidence on whether they are decreasing or persisting over time. This paper analyses the evolution of the transmission mechanism of European monetary policy by means of time-varying, heterogeneous coefficient models that allow us to test alternative stability and homogeneity assumptions, including particularly the extent to which the impact of monetary policy has changed and cross-country differences have decreased over time. As far as we know, this is the first empirical study of the transmission mechanism of European monetary policy that attempts to do so. The “experiment” we design incorporates suggestions from the most recent contributions to the empirical literature on the transmission mechanism of European monetary policy and also innovates in one other respect by allowing for regional interdependence in the analysis. First, the specification of the econometric model is the same for all countries considered. Second, following Clements et al. (2001) and Sala (2003), we control for different central banks preferences and procedures as well as intra-Europe exchange rate movements. In addition, by allowing for contemporaneous and lagged interdependence among the open and integrated economies we avoid aggregation biases and provide for a more realistic description of the international transmission mechanism of monetary policy. Overall, these features bring our experiment closer to the “ideal” one described by Guiso et al. (1999). Obviously, such a framework cannot be estimated without introducing restrictions on the model because of the very large number of parameters involved. We specify the econometric model in terms of a few hyper-parameters and take a Bayesian approach to estimation. In addition, we take a two-stage approach to the analysis. In the first stage, we measure monetary policy by estimating a system of reaction functions à la Clarida et al., 1998 and Clarida et al., 2000. In the second stage, we assess the impact of monetary policy on economic activity by estimating a system of output equations as done by Dornbusch et al. (1998) and Peersman and Smets (1999). Thus, we do not model nominal exchange rates and inflation rates endogenously. We also focus on a small group of European countries: Germany, France, Italy and Spain. These are the four largest economies currently in the EMU, accounting for about 80 percent of the Euro-area GDP, and there are no strong reasons to expect an homogeneous or stable response to a common monetary policy shock across these economies. France has been closely tied to Germany throughout the period considered and is widely regarded as a “core” European economy, but its legal structure is different. Italy has often been singled out as a “divergent” European partner mainly because of developments in its public finances until the mid-1990s and its dual economic structure. Spain, finally, joined the European Union and the EMS much later than the other three countries and has been catching up with the rest of Europe throughout the period considered. Consistent with the predictions of the Lucas’ critique and the evidence emerging from the empirical literature on the recent U.S. monetary history, we find that the long-run cumulative impact of a common, homoskedastic monetary policy shock, identified with a homoskedastic innovation to the German reaction function, has changed after 1991, decreasing by about 10–20 percent in all countries. Interestingly, these declines are statistically significant and accompanied by some changes in the conduct of monetary policy over the same period. Consistent with Cecchetti (1999), we also find that existing differences in the transmission mechanism of European monetary policy have not decreased over the same period. We interpret these results as evidence that the transmission mechanism of European monetary policy is probably changing over time under the EMU in all countries, but rather slowly and in a synchronized manner. The paper is organized as follows. We present the econometric framework in Section 2. Here we illustrate first the empirical model for the system of reactions functions and then the model for the system of output equations. The rest of the paper reports and discusses the empirical results. In Section 3, we present the estimated monetary policy shocks and reaction function parameters. In Section 4, we present the estimated effects of monetary policy on economic activity as well as their degree of stability over time and cross-country heterogeneity. Section 5 concludes. We report details of the estimation techniques used in appendix.
نتیجه گیری انگلیسی
In this paper, we studied empirically the transmission mechanism of a common, homoskedastic monetary policy shock, identified as an innovation to the reaction function of the Bundesbank, in the four largest European countries currently in the EMU, by using time-varying, heterogenous models estimated in a Bayesian fashion based on per-EMU data. The analysis documented in the paper shares several features of the ‘ideal experiment’ described by Guiso et al. (1999): the model specification is the same for all countries considered; differences in central banks objectives and procedures as well as intra-Europe exchange rate movements, which both have disappeared under the EMU, are controlled for; regional interdependence, through which monetary policy in small open economies in part operates, is also allowed for; and, most importantly, both model parameters and shock variance–covariance matrices are allowed to change over time. We found that the transmission mechanism of such a monetary policy shock has changed in the run-up to the EMU in all countries considered, albeit not dramatically, with a long-run cumulative impact on output decreasing by about 10–20 percent in all countries after 1991. These changes are statistically significant and also associated with some shifts in the behavior of the monetary authorities in all countries considered. They are particularly evident in the case of Germany and less clear cut in the case of Italy. Interestingly, we also found that cross-country differences in the effects of our common, homoskedastic shock have not decreased over the same period. These results are not only consistent with the predictions of the Lucas’ critique, but also with two separate bodies of existing empirical evidence. First, they are consistent with the now large, albeit sometime contradicting, literature on cross-country differences in the transmission mechanism of European monetary policy. Second, our results are also consistent with the more recent literature on the evolution of the U.S. monetary policy and its effects on the economy. We conclude from this evidence that the transmission mechanism of European monetary policy is probably changing over time, albeit slowly and in all countries at the same time.