The Regional Effects of Monetary Policy in Europe

Transcription

1 The Regional Effects of Monetary Policy in Europe Ivo J.M. Arnold Universiteit Nyenrode November 1999 Abstract This paper provides an empirical analysis of monetary transmission in 68 European regions. The evidence supports previous findings for the US on the importance of the industry mix channel of monetary transmission.. In Europe, there is a significant relationship between the regional impact of monetary policy and the proportion of the labour force working in industry. In addition, countryspecific dummy variables are used to proxy for the more institutionallydetermined differences in monetary transmission. These are significant for the regions in Spain and in a few smaller EU countries. Finally, an analysis of variance for the German, French, Italian and British regions shows that the between-countries variation in the regional impact of monetary policy is not significantly larger than the within-countries variation. I conclude that, at present, regional differences in the impact of monetary policy are unlikely to seriously hamper ECB policy. Keywords: monetary transmission, regional effects, EMU JEL Classifications: E50 Contact address: Universiteit Nyenrode, Straatweg BG Breukelen, The Netherlands; address: arnold@nyenrode.nl. I am grateful to Fred Lee for research assistance.

2 1. Introduction Before EMU came into being, a major economic debate concerned the costs and benefits of monetary union. Regarding the costs, the absence or presence of asymmetric shocks became a well-researched issue (see OECD (1999)). A broad consensus on whether asymmetric shocks constitute a major impediment to monetary union has, however, failed to emerge. Bayoumi and Eichengreen (1993) exemplify the pessimistic view that the presence of asymmetric shocks will entail severe costs, while Bini, Smaghi and Vori (1993) represent a more optimistic viewpoint. Now that EMU has become reality, attention has shifted away from asymmetric shocks towards the asymmetric transmission of uniform monetary policy shocks originating from the ECB. The concern is that a common monetary policy might have differential effects on EMU member states, caused by differences in the monetary transmission mechanism. When one size doesn t fit all, this may complicate macro-economic management, as the ECB will have to weight the varying consequences of its actions on the EMU countries. More importantly, when ECB policy is seen to be incapable of addressing the economic needs of individual member states, this might erode political and public support for monetary union. For the ECB, it is therefore important to understand how interest rates affect the euro area and what it can do to mitigate any differential effects. A large literature deals with this topic. Notable papers in this line of research are BIS (1995), Barran, Coudert and Mojon (1996), Brittan and Whitley (1997), Dornbusch, Favero and Giavazzi (1998), Ehrmann (1998), Favero and Giavazzi (1999), Gerlach and Smets (1995), Hughes Hallett and Piscitelli (1999), Hughes Hallett and Warmedinger (1999), Ramaswamy and Sloek (1997) and Taylor (1995). However, these studies offer the ECB little practical help. The OECD (1999) survey shows that most studies report differences in the monetary transmission across European countries, but that a consensus on the ordering of countries according to the interest rate responsiveness of GDP is lacking. In addition, Kieler and Saarenheimo (1998) question the statistical significance of the reported cross-country differences. 2

3 Empirical studies documenting differences in monetary transmission across European countries typically use country data. Exceptions are Ganley and Salmon (1997) and Hayo and Uhlenbrock (1999), who explore regional differences in monetary transmission for respectively the UK and Germany. Carlino and DeFina (1999) apply estimates from regional US data to the EMU. But these three studies do not provide a comparison across all EMU regions. In contrast, the literature on asymmetric shocks includes regional evidence, see De Grauwe and Vanhaverbeke (1991) and De Nardis, Goglio and Malgarini (1996). One of their findings is that the variability of output is much greater at the regional than at the national level. In that case, regional diversification within EU countries can mitigate the potential instability arising from asymmetric shocks. Comparing regional with national variation also makes sense in assessing the severity of cross-country differences in monetary transmission. This paper will therefore apply such an approach to monetary transmission. Using data from 68 European regions, I measure the impact of monetary policy shocks on regional GDP. The estimates will be used to answer two questions. First, building on Carlino and DeFina (1998), I look whether the impact of monetary policy shocks is related to the industrial composition of regions. Second, I will test whether the differential effects of monetary policy vary more within countries or between countries. The organization of this paper is as follows. The next section offers a brief review of factors which may cause differential effects of monetary policy. Section 3 is the main part of this paper. It includes a discussion of the data, the methodology and the empirical results. Section 4 concludes with policy implications. 3

4 2. Factors causing differential effects of monetary policy For a comprehensive overview of the literature on the monetary transmission mechanism, see the surveys by Bernanke and Gertler (1995), Cecchetti (1995), Christiano, Eichenbaum and Evans (1998), Gertler (1988), Gertler and Gilchrist (1993) and Mishkin (1995, 1996). Below, I will briefly list the factors which might be held responsible for a differential impact of monetary policy in the EMU. Traditionally, economic theory has relied on the interest rate channel to explain monetary transmission: an increase in interest rates reduces demand for investment goods and (durable) consumer goods. When industries differ in their sensitivity to changes in the interest rate and regions differ in industrial composition, this theory is able to explain how a uniform monetary policy generates regional differences. Carlino and DeFina (1998), Ganley and Simon (1997) and Hayo and Uhlenbrock (1999) use this channel to explain differential effects of monetary policy inside respectively the US, the UK and Germany. See Taylor (1995) for a survey of the empirical work in this area. The interest rate is just one of many asset prices. Monetary policy shocks may also affect other assets, such as the exchange rate and share prices. Through the exchange rate channel, monetary policy influences net exports. Regional effects then arise in the presence of cross-regional variation in openness, see Dornbusch, Favero and Giavazzi (1998). The equity channel of monetary transmission works either through Tobin s q theory of investment demand or through a wealth effect on consumer demand, see Mishkin (1996). Regional differences in either Tobin s q or in the distribution of wealth may cause regional effects. A recent theory of monetary transmission focuses on the role of information problems in credit markets. This so-called credit view identifies two transmission channels: the bank lending and the balance sheet channel. The former channel looks at the ability and willingness of banks to provide loans, see Kashyap and Stein (1997). Monetary policy affects the economy through the supply of bank credit, as some borrowers (such as small firms) lack substitutes for bank loans. Regional differential effects arise when regions differ in the dependence on and availability of bank credit. The balance-sheet channel of monetary transmission 4

5 works through the net worth and cash flows of firms. An expansionary monetary policy will raise both, reducing asymmetrical information problems in credit markets. As a result, lending and investment spending may increase. The balancesheet channel may also explain changes in consumer spending, see Mishkin (1978). In the credit view, differential regional effects of monetary policy may be caused by cross-regional differences in financial structure, measured by e.g. the proportion of small banks and small firms in an economy, the health of the banking sector and the availability of non-bank funding, and the amount of collateral (see respectively Kashyap and Stein (1997) and Dornbusch, Favero and Giavazzi (1998)). Other contributions to the literature on the credit channel in the euro area are Allen and Vega (1998), Borio (1996), De Bondt (1998), Favero, Giavazzi and Flabbi (1999), MacLennan, Muellbauer and M. Stephens (1998) and Mojon (1999a, 1999b). All transmission channels described above relate to the effect of monetary policy on aggregate demand. The final effect on income is the result of the interaction of supply and demand. Differential effects of monetary policy could therefore also be the result of regional differences in the supply curve, caused by e.g. differences in the flexibility and institutional features of labour and product markets, see OECD (1999). Differential effects of monetary policy not only arise because regions or countries vary in the importance of the transmission channels. The speed of interest rate adjustment also matters. Though not a separate channel, the speed of adjustment will determine how fast a change in interest rates will work its way through the transmission channels. For Europe, many studies have identified large crosscountry differences in the adjustability of interest rates, see Borio (1996), Barran, Coudert and Mojon (1997), Kashyap (1997), De Bondt (1998) and Mojon (1999a). Ceteris paribus, the impact of monetary policy shocks will be stronger in countries or regions where the interest rates on debt contracts adjust more rapidly to monetary tightening by the central bank. But in theory, these differences will disappear following the introduction of a common monetary regime, see Smets (1997) and Arnold and De Vries (1999). Missale and Piga (1999) document the quick convergence of the duration of public debt of EMU governments. 5

6 3. Empirical Evidence The empirical analysis uses regional data to estimate the differential effects of monetary policy in Europe. Data limitations make it hard to precisely attribute any differential effects to the factors discussed above. For example, the regional data do not include measures for the bank lending or balance sheet channel. Given these limitations I will proceed as follows. First, regional data on the shares of the labour force in agriculture, services and industry are used to measure the importance of the interest rate channel. Second, all factors which are likely to be the same within countries but different between countries - such as institutional features of labour and product markets, the legal system (Cecchetti (1999)) and presumably also many aspects of financial structure - are captured by country dummy variables. The regional classification is based on the Eurostat NUTS1 classification. Data on regional GDP (in current Ecu s) are from the economics accounts in Eurostat s regional statistics database. They have been converted to real GDP by first converting the data into national currencies and next deflating the resulting series by the national price indices (CPI). The end result is an annual series for real GDP, for most regions covering the period from 1979 to The mean and standard deviation of real regional growth rates are in Table 1. In addition, Table 1 lists population estimates for 1994 and the share of the labour force working in industry (LFI) for The latter data are from Eurostat s community labour force survey. The LFI measure has been calculated as the number of people working in industry as a percentage of the total labour force (working in either agriculture, industry or services). This measure differs from the one in Carlino and DeFina (1998), who use the share of manufacturing industry in regional GDP. 1 The sample is shorter for Spain ( ), Sweden ( ), Austria ( ) and Finland ( ). Nine regions were left out. The sample period was considered too short for the East German states Brandenburg, Mecklenburg-Vorpommern, Sachsen, Sachsen-Anhalt and Thüringen ( ) and for the Portugese regions of Madeira and the Azores ( ). Berlin was left out because of the effect of German unification; the French overseas departments (Départments d Outre-Mer) because of its non-european character. Apart from the German states, these regions are quantitatively unimportant. 6

7 The empirical approach consists of two steps. First, the effects of monetary policy on regional GDP are estimated using a time-series regression of real GDP growth on the interest rate. Second, the regional effects are related to the regional industry mix, measured by LFI, in a cross-sectional regression. Regarding the first step, the dataset is not ideal to do an extensive econometric analysis, comparable to e.g. the (structural) vector autoregression of Carlino and DeFina (1998) and Ganley and Salmon (1997). The brief sample period and the low data frequency limit the degrees of freedom. Further, a requirement is model uniformity across all regions, to avoid that estimates depend on data availability and the amount of research effort allocated to a specific region. I use the following model to measure the impact of monetary policy on the regional economies: (1) y i,j,t = α + β 1, i,j i j,t-1 + β 2, i,j y i,j,t-1 In equation (1), real GDP growth in region i of country j ( y i,j,t ) is modeled as a function of the lagged change in interest rate in country j ( i j,t-1 ) and the lagged growth rate ( y i,j,t-1 ). The short term interest rate (IFS line 60B) is used as the indicator of monetary policy, see Leeper, Sims and Zha (1996). The interest rate has been first differenced to account for the presence of a unit root in the level of the interest rate, which was detected for all countries except Germany using a Augmented Dickey Fuller (ADF) unit root test. The ADF test statistics for the German interest rate is -3.65, which is significant at a 5% level. For the Netherlands and Portugal, the ADF statistics were respectively 2.79 and 2.92, in both cases significant at a 10% level, but not at a 5% level. For all other countries the ADF test statistics were insignificant at a 10% level. The lagged growth rate is used to pick up any auto-correlation in the real growth series. For countries consisting of just one region (Denmark, Ireland, Luxembourg, Portugal, Sweden) estimation is done by ordinary least squares. For all other countries, a pooled estimation is conducted using Seemingly Unrelated Regression. The pooled estimation allows for fixed effects and cross-section specific coefficients on i j,t-1 and y i,j,t-1. However, the results from the pooled estimation do not deviate from the results of OLS estimates for all regions separately. 7

8 In the second step, the β 1,i,j coefficient from equation (1) is used in the following cross-section regression: (2) β 1,i,j = γ + δ 1 LFI i,j + Σ dum j In equation (2), LFI i,j is the regional share of the labour force working in industry in region i of country j. The hypothesis is that industry is of a more cyclical nature than services or agriculture, see Carlino and DeFina (1998). Given negative β 1 s, one would expect δ 1 to have a negative sign. Equation (2) also allows for countryspecific effects through the use of country dummy variables dum j. As discussed above, the country dummies may capture all institutional differences between the European countries affecting monetary transmission. Tables 2 and 3 contain the estimation results for respectively step one and step two. Table 2 shows that for 57 out of 68 regions, β 1 has the theoretical sign, whereby an increase in the interest rate reduces economic growth. In 27 out of 68 cases, β 1 is significantly negative at a 5% significance level. As noted above, the level of the German interest rate doesn t contain a unit root, in contrast to the other interest rates. When in equation (1) the German interest rate is inserted in levels instead of in first differences as would be appropriate in the absence of a unit root the interest rate coefficients become significantly negative at a 5% level for all German regions. These results are also reported in Table 2. This poses a problem. Econometrically, the results using the level of the interest rates are more attractive. But for the cross-section regression, it would be preferable to estimate all β 1 s with the same model specification. Fortunately, the correlation coefficient between the German β 1 s from the levels specification and from the first differences specification is very high (0.82), which makes the choice between the two rather inconsequential. I have chosen to use same specification for all countries. All results have, however, been checked using the German β 1 s from the levels specification. The results corresponded closely. Table 2 also shows that the model fit differs between countries, with Austria and France being the countries where the model fits the data least. The short sample period, the annual data frequency and the desirability of a uniform model across all regions, have limited the model size and the possibilities to fine-tune the model 8

9 to national idiosyncrasies. Without question this limits the accuracy of the estimates. Bearing this in mind, I will now discuss the second step. The results for the cross-section regressions are in Table 3. This table contains estimates for the complete cross-section of 15 EU countries (EU15), for the 11 countries participating in EMU (Euroland) and for the big four (EU4: Germany, France, Italy and the UK). In addition, estimates are reported for each of the four biggest countries separately. Figures 1 to 7 show the corresponding scatter diagrams. The regression results for the country groupings show that the coefficient of LFI is both of the right sign (a higher share of the labour force working in industry implies a higher interest rate sensitivity and thus more negative interest rate coefficient) and significantly different from zero at a 5% level. The dummy variables give the size of the country-specific effects after controlling for LFI. The benchmark country (without dummy variable) is Belgium. The EU15 estimates show the Spanish and Austrian dummies to be highest and most significant. This is not surprising, as the Spanish and Austrian interest rate coefficients in Table 2 deviated the most from the theoretical negative value. But for these two countries, the accuracy of the coefficients is low. At the other side of the spectrum is Greece, whose dummy value is the most negative one. Note that for France, Italy and Germany the dummy variables are very close, suggesting that between these three countries, country-specific differences of monetary transmission are unlikely to be very important. This finding corroborates the evidence in BIS (1995), Brittan and Whitley (1997), Dornbusch, Favero and Giavazzi (1998) and Taylor (1995). The results for the euroland regions are very similar, as the third and fourth columns in Table 3 show. The final country grouping (EU4) includes 40 out of 68 regions. This grouping has the advantage that all interest rate coefficients are estimated over exactly the same sample period. Germany is the benchmark country in this group. The only country for which the dummy remains significant is the UK. After controlling for LFI, the UK interest rate coefficient is significantly more negative than that of the other large countries, which is in line with the existing evidence. The coefficient of LFI stays significant after the elimination of all dummy variables. So its significance does not depend on the inclusion of dummies. The country regressions at the bottom of Table 3 have to be interpreted with caution, as the number of observations is very low. All countries 9

10 except Germany give a negative relationship between the interest rate coefficient and LFI. The relationship is particularly strong in France, as is evident from Figure 5. Finally, Table 4 reports the results of a one-way analysis of variance on β 1 and LFI for Germany, France, Italy and the UK. For both variables, the variance between the four countries is not significantly greater than the variance between the regions within each country. Both F-statistics are insignificant at any reasonable significance level. These results echo earlier findings on output variability at the regional and national level by De Grauwe and Haverbeeke (1991) and De Nardis, Goglio and Malgarini (1996). 10

11 4. Conclusions and policy implications An empirical analysis of monetary transmission in 68 European regions leads to the following conclusions. First, there appears to be a significant relationship between the regional impact of monetary policy measured by the interest rate coefficient and the proportion of the labour force working in industry. This link supports the existing US evidence on the importance of the industry mix channel of monetary transmission. Second, country-specific dummy variables, which proxy for the more institutionally-determined differences in monetary transmission, are important for Spain and a few smaller EU countries. Third, an analysis of variance for Germany, France, Italy and the UK shows that the between-countries variation in the interest rate coefficient and in the proportion of the labour force working in industry is not significantly larger than the withincountries variation. The regional mix between employment in agriculture, services and industry is important for the transmission of monetary policy. However, the evidence seems to indicate that at least the large European countries are regionally welldiversified enough to minimize the risk that ECB policy will produce a markedly different impact across countries. At present, the risk that regional differential effects of monetary transmission give rise to national instability and tensions in the EMU seems small, see also Gros and Thygesen (1998). This may change, as has been pointed out by Krugman (1993). Increased specialization within an integrated Europe could result in a more heterogeneous industrial structure, as producers flock together to reap the benefits of greater geographic concentration. This would increase the differential effects of monetary policy. In contrast, regional differences in transmission which are the result of institutional differences between EU countries - such as cross-country differences in taxation, law, regulation of markets and financial structure - are likely to be further reduced in the process of European integration. Breaking down these institutional barriers will take time, as will the process of industrial specialization. But note that these two developments will have an opposite impact on monetary transmission. Whereas the former will reduce any differential effects of monetary policy, the latter will increase them. 11

12 As we have seen, at present regional variation in industrial composition does not yet produce severe cross-country disparities in the impact of monetary policy. Suppose, for the sake of argument, that in the future these disparities will increase as a result of industrial specialization. What would the policy implications of such a development be? In most Western countries, the industrial composition results from the free choice of private sector agents, not government planners. Regional effects of monetary policy caused by differences in industry mix are therefore difficult to eradicate through direct government intervention. However, governments can try to compensate regions through fiscal policy. The wisdom of such a policy depends on the welfare effects. Economic theory tells us that economic agents who voluntarily take on more risk should be compensated by a higher return. In the context of monetary transmission, one would expect industries which disproportionately suffer from the impact of monetary policy to compensate employers and employees for taking on this risk. For example, job security is higher as a civil servant than as a employee in the car industry, but pay will be less. If there indeed appears to be such a risk-return relationship, the case for fiscal compensation is weak. If differential regional effects of monetary policy are the result of industrial composition, there is little governments can or should do. However, to the extent that a uniform transmission of ECB policy is still hampered by institutional differences between EMU countries, the first-best solution would be to further harmonize the institutional features of the European economies. 12

13 References Allen, C. and J.L. Vega, (1997), The Credit Channel in the Monetary Transmission Process in the EU countries, MESD mimeo, European Monetary Institute Arnold, I.J.M. and C.G. de Vries, (1999), Endogenous Financial Structure and the Transmission of ECB Policy, in: J. von Hagen en C. Waller (eds.) Common Money, Uncommon Regions, Kluwer Academic Publishers, forthcoming. Barran, F., V. Coudert and B. Mojon, (1997), The Transmission of Monetary Policy in European Countries, in: European Monetary Policy, Pinter, London, Bayoumi, T. and B. Eichengreen, (1993), Shocking Aspects of European Monetary Union, in: Torres and Giavazzi (eds), Adjustment and Growth in the European Monetary Union, Cambridge University Press, Cambridge Bernanke, B.S. and M. Gertler, (1995), Inside the Black Box, The Credit Channel of Monetary Policy Transmission, Journal of Economic Perspectives, 9(4), Bini Smaghi, L. and S. Vori, (1993), Rating the EC as an Optimal Currency Area, Temi di Discussione, No. 187, Banca d Italia, Rome Bondt, G.J. de, (1998), Credit and Asymmetric Effects of Monetary Policy in Six EU Countries: an Overview, DNB Staff Reports, No. 23 Borio, C.E.V. (1996), Credit characteristics and the monetary policy transmission mechanism in fourteen industrial countries: Facts, conjectures and some econometric evidence, in: K. Alders, K.Koedijk, C.Kool and C. Winder (eds.), Monetary Policy in a Converging Europe, Kluwer Academic Publishers, Brittan, E. and J. Whitley, (1997), Comparing the Monetary Transmission Mechanism in France, Germany and the United Kingdom: Some Issues and Results, Bank of England Quarterly Bulletin, May, Carlino G. en R. DeFina, (1998), The Differential Regional Effects of Monetary Policy, Review of Economics and Statistics, 80(2), Carlino G. en R. DeFina, (1999), Monetary Policy and The US States and Regions: Some Implications for European Monetary Union, in: J. von Hagen en C. Waller (eds.) Common Money, Uncommon Regions, Kluwer Academic Publishers, forthcoming. Cecchetti, S.G., (1995), Distinguishing Theories of the Monetary Transmission Mechanism, Federal Reserve Bank of St. Louis Review, 77, Cecchetti, S.G., (1999), Legal Structure and the Monetary Policy Transmission, mimeo, Federal Reserve Bank of New York Christiano, L.J., M. Eichenbaum and C.L. Evans, (1998), Monetary Policy Shocks: What Have We Learned and to What End?, NBER Working Paper, No

14 De Grauwe, P. and W. Vanhaverbeke, (1991), Is Europe and Optimal Currency Area? Evidence from Regional Data, CEPR Discussion Paper, No. 555 De Nardis, S., A. Goglio and M. Malgarini, (1996), Regional Specialization and Shocks in Europe: Some Evidence from Regional Data, Weltwirtschaftliches Archiv, 132(2), Dornbusch, R., C. Favero and F. Giavazzi, (1998), Immediate Challenges for the ECB, Economic Policy, April, Ehrmann, M. (1998), Will EMU Generate Asymmetry? Comparing Monetary Policy Transmission Across European Countries, EUI Working Paper ECO No. 98/28, European University Institute Favero, C. and F. Giavazzi, (1999), An Evaluation of Monetary Policy Transmission in the Context of the European Central Bank, A Report to the European Parliament Favero, C., F. Giavazzi and L. Flabbi, (1999), The Transmission Mechanism of Monetary Policy in Europe: Evidence from Banks Balance Sheets, NBER Working Paper, No Ganley, J. and C. Salmon, (1997), The Industrial Impact of Monetary Policy Shocks; Some Stylised Facts, Bank of England Working Paper, No. 68 Gerlach, S. and F. Smets, (1998), The Monetary Transmission: Evidence from G- 7 Countries, CEPR Discussion Paper, No Gertler, M., (1988), Financial Structure and Aggregate Economic Activity: An Overview, Journal of Money, Credit and Banking, 95, Gertler, M. and S. Gilchrist, (1993), The Role of Credit Market Imperfections in the Monetary Transmission Mechanism: Arguments and Evidence, Scandinavian Journal of Economics, 95, Gros, D. and N. Thygesen (1998), European Monetary Integration: From the European Monetary System to European Monetary Union, Addison Wesley Longman, London and New York. Hayo, B., and B. Uhlenbrock (1999), Sectoral Effects of Monetary Policy in Germany, in: J. von Hagen en C. Waller (eds.) Common Money, Uncommon Regions, Kluwer Academic Publishers, forthcoming. Hughes Hallett, A.J. and L. Piscitelli (1999), EMU in Reality: The Effect of a Common Monetary Policy on Economies with Different Transmission Mechanisms, CEPR Discussion Paper, No Hughes Hallett, A.J. and T. Warmedinger (1999), On the Asymmetric Impacts of a Common Monetary Policy, in: J. von Hagen en C. Waller (eds.) Common Money, Uncommon Regions, Kluwer Academic Publishers, forthcoming. Kashyap, A.K., (1997), The Lending Channel and European Monetary Union, in: European Monetary Policy, Pinter, London,

15 Kashyap, A.K. and J.C. Stein, (1997), The Role of Banks in Monetary Policy: a Survey with Implications for European Monetary Union, Economic Perspectives, A Review from the Federal Reserve Bank of Chicago, September/October, 2-18 Kieler, M. and T. Saarenheimo, (1998), Differences in Monetary Policy Transmission? A Case not Closed, draft Krugman, P. (1993), Lessons of Massachusetts for EMU, in: Torres and Giavazzi (eds), Adjustment and Growth in the European Monetary Union, Cambridge University Press, Cambridge Leeper, Sims and Zha (1996), What Does Monetary Policy Do?, Brookings Papers on Economic Activity, No. 2, 1-78 MacLennan, D., J. Muellbauer and M. Stephens, (1998), Asymmetries in Housing and Financial Market Institutions amd EMU, Oxford Review of Economic Policy, 14(3), Mishkin, F., (1978), The Household Balance Sheet and the Great Depression, Journal of Economic History, 38(4), Mishkin, F., (1995), Symposium on the Monetary Transmission Mechanism, Journal of Economic Perspectives, 9(4), 3-10 Mishkin, F., (1996), The Channels of Monetary Transmission: Lessons for Monetary Policy, Banque de France Bulletin: Digest, No. 27, Missale, A. and G. Piga, (1999), Public Debt Management in the Euro Area, paper presented at the CEPR/European Summer Institute 1999 Conference, Lisbon Mojon, B., (1999a), Financial Structure and the Interest Channel of Monetary Policy in the euro area, mimeo, European Central Bank Mojon, B., (1999b), Credit Channel(s)in the Euro Area: What is the Evidence of Distributional Effects?, mimeo, European Central Bank Organisation for Economic Co-operation and Development, (1999), EMU, Facts, Challenges and Policies, OECD, Paris Ramaswamy, R. and T. Sloek, (1997), The Real Effects of Monetary Policy in the European Union: What are the Differences?, IMF Working Paper, No. 97/160 Smets, F., (1997), Les Différences de Structures Financières Importent-elles en Vue de la Future et Unique Politique Monétaire Européenne?, Revue Française D Economie, 12(2), Taylor, J., (1995), The Monetary Transmission Mechanism: An Empirical Framework, Journal of Economic Perspectives, 9(4),

16 Table 1: Regional classification and descriptive statistics Code Region Population GDP Growth Labour force share (x1000, 1994) mean (%) stdev (%) in industry (1997) BE1 Reg. Bruxelles-Cap BE2 Vlaams Gewest BE3 Région Wallonne DK Danmark DE1 Baden-Württemberg DE2 Bayern DE3 Bremen DE4 Hamburg DE5 Hessen DE6 Niedersachsen DE7 Nordrhein-Westfalen DE8 Rheinland-Pfalz DE9 Saarland DE10 Schleswig-Holstein GR1 Voreia Ellada GR2 Kentriki Ellada GR3 Attiki GR4 Nisia Aigaiou, Kriti ES1 Noroeste ES2 Noreste ES3 Madrid ES4 Centro ES5 Este ES6 Sur ES7 Canarias FR1 Ile de France FR2 Bassin Parisien FR3 Nord-Pas-de-Calais FR4 Est FR5 Ouest FR6 Sud-Ouest FR7 Centre-Est FR8 Méditerranée SE Sverige IR Ireland IT1 Nord Ouest IT2 Lombardia IT3 Nord Est IT4 Emilia-Romagna IT5 Centro IT6 Lazio IT7 Abruzzo-Molise IT8 Campania IT9 Sud IT10 Sicilia IT11 Sardegna LU Luxembourg NL1 Noord-Nederland NL2 West-Nederland NL3 Noord-Holland NL4 Zuid-Nederland AT1 Ostösterreich AT2 Südösterreich AT3 Westösterreich PT Portugal (Continente) FI1 Manner-Suomi FI2 Ahvenanmaa/Åland UK1 North UK2 Yorkshire and Humberside UK3 East Midlands UK4 East Anglia UK5 South East UK6 South West UK7 West Midlands UK8 North West UK9 Wales UK10 Scotland UK11 Northern Ireland

17 Table 2: Time-series results Country Code Coefficient T-stat Adj. R 2 DW # obs β 1 Belgium BE BE BE Denmark DK Germany DE (results for levels DE specification in DE italics, see text) DE DE DE DE DE DE DE Greece GR GR GR GR Spain ES ES ES ES ES ES ES France FR FR FR FR FR FR FR FR Sweden SE Ireland IR Italy IT IT IT IT IT IT IT IT IT IT IT Luxembourg LU Netherlands NL NL NL NL Austria AT AT AT Portugal PT Finland FI FI United Kingdom UK UK UK UK UK UK UK UK UK UK UK

18 Table 3: Cross-section results EU15 Euroland EU4 EU4 Coefficient T-stat Coefficient T-stat Coefficient T-stat Coefficient T-stat Constant LFI DUMDE DUMFR DUMIT DUMES DUMIR DUMLU DUMNL DUMAT DUMPT DUMFI DUMUK DUMSE DUMGR DUMDK Adj. R DW F # obs DE DE FR IT UK first dif. i levels i Constant T-stat LFI T-stat Adj. R DW # obs

19 Table 4: One-way Analysis of Variance A: Labour force share in industry (LFI) Groups Number Sum Mean Variance DE FR IT UK Source SS DF MS F p-value Between groups Within groups Total B: Interest rate coefficient Groups Number Sum Mean Variance DE FR IT UK Source SS DF MS F p-value Between groups Within groups Total