به سمت توضیح عدم تقارن بین کشوری در انتقال پولی

This paper quantifies the importance of financial structure, labor market rigidities and industry mix for the monetary transmission mechanism. To do so, I determine how closely the impulse responses to a monetary policy shock obtained from country-specific vectorautoregressive (VAR) models and a non-standard panel VAR model match. In the country-specific VAR models, the impulse responses vary across countries in an unrestricted fashion. In the panel VAR model, the impulse responses also vary across countries, but only to the extent that countries differ regarding their financial structure, labor market rigidities and industry mix. For a sample of 20 industrialized countries over the time period from 1995 to 2009, I find that up to 70% (50%) of the cross-country asymmetries in the responses of output (prices) to a monetary policy shock can be replicated by accounting for cross-country differences in financial structure, labor market rigidities and industry mix. Moreover, while in the short run asymmetries in the output responses arise mainly due to cross-country differences in industry mix, in the medium run differences in financial structure and labor market rigidities are more important. Finally, cross-country differences in industry mix appear to be of rather minor importance for cross-country asymmetries in the transmission of monetary policy to prices.

What determines the monetary transmission mechanism? How important are financial structure, labor market rigidities and industry mix? Do these structural characteristics shape the monetary transmission mechanism at different horizons? In this paper, I shed light on these issues by establishing several stylized facts about the quantitative importance of a set of an economies’ structural characteristics for the monetary transmission mechanism. In particular, I find that in a sample of 20 industrialized countries over the time period from 1995 to 2009 up to 70% (50%) of the cross-country asymmetries in the responses of output (prices) to a monetary policy shock can be replicated by accounting for differences in countries’ financial structure, labor market rigidities and industry mix. Moreover, while in the short run asymmetries in the output responses arise mainly due to cross-country differences in industry mix, in the medium run differences in financial structure and labor market rigidities are more important. Finally, cross-country differences in industry mix appear to be of rather minor importance for cross-country asymmetries in the transmission of monetary policy to prices. Numerous papers have attempted to identify the determinants of the monetary transmission mechanism by exploiting asymmetries in the effects of monetary policy on output and prices across countries (or regions and/or industries). The standard approach is to regress a feature of countries’ impulse responses to a monetary policy shock (typically the maximum or the cumulated response) on time averages of countries’ structural characteristics in a cross-section regression (see Carlino and DeFina, 1998, Hayo and Uhlenbrock, 1999, Mihov, 2001, Arnold and Vrugt, 2004, Dedola and Lippi, 2005 and Peersman and Smets, 2005).1 The standard approach is inefficient and provides only limited guidance for policymakers. First, it does not exploit the time-series variation in countries’ structural characteristics to identify the determinants of the monetary transmission mechanism. This is inefficient, as many determinants of the monetary transmission mechanism display variation over time. The panel vectorautoregressive (VAR) model employed in this paper does exploit the time-series variation in countries’ structural characteristics and should, therefore, pin down more precisely the importance of financial structure, labor market rigidities and industry mix for the monetary transmission mechanism. Second, the standard approach focuses only on a few of the features of the monetary transmission mechanism. However, besides the maximum and the cumulated impulse response to a monetary policy shock routinely examined, other features of the monetary transmission mechanism such as the persistence of the response or the time it takes until output and prices reach their trough response are of interest as well. In the panel VAR model employed in this paper, the entire shape of the impulse responses of output and prices to a monetary policy shock is conditioned on countries’ structural characteristics. Third, because the standard approach focuses on identifying the determinants of the monetary transmission mechanism rather than assessing their quantitative importance, it provides no guidance to policy—for example in a currency union—as to how large the returns of different harmonization policies (in terms of reducing asymmetries in monetary transmission) are. In contrast, the purpose of this paper is to quantify the importance of financial structure, labor market rigidities and industry mix for the monetary transmission mechanism, which would allow policy makers in currency unions to compare the cost-benefit analyses of different harmonization policies. The remainder of this paper is organized as follows: Section 2 presents the empirical evidence on cross-country asymmetries in monetary transmission. In Section 3, I show that financial structure, labor market rigidities and industry mix differ across countries, and discuss the mechanisms through which these structural characteristics may affect monetary transmission. In Section 4, I motivate the design of the panel VAR model employed in this paper, lay out how impulse responses can be constructed and describe the empirical model specification. Section 5 presents results and Section 6 robustness checks as well as a discussion of several elements of the empirical approach taken in this paper. Finally, Section 7 concludes.

This paper uses the PCHVAR model to analyze the importance of financial structure, labor market rigidities and industry mix for the monetary transmission mechanism. In the PCHVAR model, cross-country asymmetries in the impulse responses of output and prices to a monetary policy shock are exploited in order to learn about the determinants of the monetary transmission mechanism. In contrast to the existing empirical literature on cross-country asymmetries in monetary transmission based on cross-sectional regressions of impulse response estimates on time-averages of countries’ structural characteristics (the standard approach), the PCHVAR model allows to (i) exploit the time-series variation in structural characteristics, (ii) to take into account the entire shape of the impulse responses rather than only the maximum response, and (iii) to quantify the importance of a set of structural characteristics for cross-country asymmetries in monetary transmission. For these reasons, the analysis based on the PCHVAR should improve our understanding of the monetary transmission mechanism relative to the standard approach. I find that up to 70% (50%) of the cross-country asymmetries in the responses of output (prices) to a monetary policy shock can be replicated by jointly accounting for differences in financial structure, labor market rigidities and industry mix. A tentative decomposition of these figures into the contributions of each structural characteristic shows that while in the short run cross-country asymmetries in the output responses arise mainly due to differences in industry mix, in the medium run differences in financial structure and labor market rigidities gain more importance. Moreover, differences in industry mix appear to be of rather minor importance for cross-country asymmetries in the transmission of monetary policy to prices. A possible explanation for these findings is that the transmission of monetary policy to real activity first works through the interest rate channel and then gets amplified through credit channel effects and other frictions such as labor market rigidities; in contrast, only credit channel effects are sufficiently persistent to induce firms to lower prices in the presence of adjustment costs such as those generated by labor market rigidities. Given the observed trends in financial innovation and integration, labor market deregulation and structural change in industrialized countries, the results of this paper point to potentially large variations in the future monetary transmission mechanism. Moreover, even though the results of this paper are not specific to countries in currency unions, some of them are particularly relevant for policymakers in currency unions: Policies aimed at harmonizing labor markets and fostering financial integration as well as structural change may markedly reduce cross-country or region asymmetries in monetary transmission.