Empirical evaluation of nominal convergence in Czech Republic, Poland and Hungary (CPH)

Economic Modelling 26 (2009) 993–999

Contents lists available at ScienceDirect

Economic Modelling j o u r n a l h o m e p a g e : w w w. e l s ev i e r. c o m . l o c a t e / e c m o d

Empirical evaluation of nominal convergence in Czech Republic, Poland and Hungary (CPH) Toure Mamoudou ⁎, Trabelsi Jamel 1, Dufourt Frédéric University of Strasbourg, 61, Avenue de la Forêt Noire F-67085 Strasbourg Cedex, France

a r t i c l e

i n f o

Article history: Accepted 4 March 2009 JEL classification: C51 E31 E52 E58 O11 F41

a b s t r a c t We estimate a four variable structural vector auto regression (SVAR) model of the Czech Republic, Poland and Hungary economies in order to evaluate the links between the instruments of monetary policy and inflation outcomes. We find that the linkages between the interest rates and price levels are weak. However, the exchange rate constitutes the most important channel of monetary policy transmission for Poland and Hungary. For the Czech Republic, the link between interest rate rise and price level is rather indirect. © 2009 Elsevier B.V. All rights reserved.

Keywords: CPH Nominal convergence Monetary policy shock Structural VAR

1. Introduction This paper is concerned with the evaluation of the monetary transmission mechanisms in Central and Eastern European Countries (CEECs). These countries adopted several macroeconomic stabilization programs in order to initiate a monetary convergence towards the eurozone. In the early 1990s, they enacted the pegging exchange rates framework to minimize the difference between the domestic and the euro-area inflation rates. This was accomplished by applying a set of measures including alternative monetary and exchange rate regimes (Jonas and Mishkin, 2003; Orlowski, 2004). However, for Czech Republic, Poland and Hungary, these policies appear to have had little success in promoting sustainable price stability (Orlowski, 2003). In response to these monetary and exchange risks the CPH (Czech Republic, Poland and Hungary) adopted a strict inflation targeting and were directed towards more flexible exchange regimes (Aglietta et al., 2003; Golinelli and Rovelli, 2005). Indeed, with the DIT, the CEECs abandoned exchange rate pegs and adopted an autonomous monetary policy based on predominant inflation targeting, combined with a conditional exchange rate

⁎ Corresponding author. Tel.:+33 3 90 24 21 36. E-mail addresses: [email protected] (T. Mamoudou), [email protected] (T. Jamel), [email protected] (D. Frédéric). 1 $ LEFA University of Sfax, Tunisia. 0264-9993/$ – see front matter © 2009 Elsevier B.V. All rights reserved. doi:10.1016/j.econmod.2009.03.007

stability target. This new strategy has been in fact successful in generating sustainable prices because it is devoted solely to lowering inflation. However, it is important in any inflation targeting strategy to understand the transmission mechanism of the monetary policy shocks to inflation outcomes. Are the links between the interest rate, which is considered as the main monetary instrument, and the price level direct or do they use other channels? Ball (2002) argues that the effects of exchange rates on inflation through import prices are the fastest channel from monetary policy to inflation. A further motivation of this paper is to evaluate the role of the exchange rates to consolidate the effects of interest rates in attempting to reduce the price level. Following Blanchard and Quah (1989), Gali (1992), Cecchetti and Rich (2001), Bruneau and De Bandt (2003), Elbourne and De Haan (2005), and Jarocinski (2006), we propose an econometric analysis based on the structural VAR approach which is quite convenient to provide information on monetary policy by including forward-looking monetary policy rules, which take into account the role of the exchange rate. We try to examine how monetary shocks can induce significant bendings in CPH price level dynamics and through which channels. We find, especially for the Czech Republic, that the transmission of the interest shock on prices level decline is statistically significant and persistent. The structure of the paper is the following: Section 2 presents an overview of the evolution of monetary policies in CPH. Section 3 discusses the methodology we use to model monetary policy and to

994

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

identify the structural shocks in the VAR model. Section 4 reports impulse responses and variance decomposition to show the domestic contractionary monetary shock effects and their contributions to the dynamics of selected variables. Section 5 contains conclusions with some suggestions for possible further research.

2. Overview of monetary policies in CPH The central banks of several European transition economies implemented over-all (??) several basic monetary regimes to ensure their macroeconomic stabilization in order to join the Economic and Monetary Union (EMU). They adopted the currency peg regimes in the early 1990s, followed by implicit targets for exchange rates. These policies represented nominal anchors for macroeconomic stabilization. Thus, the exchange rate pegs induced lowering inflation at a significant level and this exacerbated the real appreciation of domestic currencies, which led to worsening current account deficits and to attracting floating capital. The fix exchange rate was carrying financial instability and accelerating inflation. In response to these monetary and exchange risks the CPH (Czech Republic, Poland and Hungary) abandoned the exchange rate targeting regimes (at different times and in different ways) to finally adopt an inflation targeting framework with more flexible exchange rate regimes (Aglietta et al., 2003; Golinelli and Rovelli, 2005). This program reinforces commitment to contain inflation volatility in spite of Harrod–Balassa–Samuelson effect problems and international capital inflows. The Czech Republic was the first European transition economy to implement the inflation targeting framework after abandoning a fixed exchange rate regime following turbulence in May 1997. In the early 1990s, the fixed exchange rate regime generated low and significant levels of inflation, before producing real appreciations inducing an erosion of competitiveness. The policy mix implemented by the Czech authorities contributed to higher interest rates which attracted more short-term foreign capital, keeping inflation high and widening the current account deficit to be unsustainable. Furthermore, uncertainties in financial markets, triggered by speculative attacks, accelerated the flight of foreign investors, forcing the authorities to stop defending a fixed exchange rate. On May 26, 1997, the Czech National Bank (CNB) decided to allow the domestic currency to float freely. Accordingly, in 1998, the monetary policy changed to a strategy of inflation targeting owing to the instability of money demand.2 However, during the period of inflation targeting, the CNB kept on intervening in the foreign exchange market (Frommel and Schobert, 2006). Holub (2004) argues that these interventions (in early 1998 and in 1999/2000) were not consistent with inflation targeting. Like the Czech Republic, Poland was pegged to a basket of currencies in the early 1990s. However, inflation did not decline and the fixed nominal exchange rate led to rapid real appreciation. Therefore, a crawling peg was introduced in October 1991, although until April 2000, Poland maintained implicit target zones accompanied by crawling devaluation for the exchange rate (Jonas and Mishkin, 2003). Nevertheless capital account liberalization led in 1994 and 1995 to large capital inflows, which constrained the authorities to widen the crawling exchange rate band in May 1996. During the period 1996– 1998, the National Bank of Poland (NBP) experimented with targeting interest rates. But these alternative regimes were not really consistent with low inflation. Due to the increased integration of financial markets, Poland's transition to an inflation targeting regime began during the last quarter of 1998, although it was not announced formally until January 1, 1999. Currently, the exchange rate plays no major role in the monetary policy as the NBP does not intervene on the foreign exchange market (Frommel and Schobert, 2006). 2

Annual reports, CNB Annual Report, 1998, p.48.

Hungary adopted early in transition economic an exchange rate peg against the basket of currencies. However, the peg was adjusted downward quite often to maintain external competitiveness. The fluctuation band was gradually widened to reduce speculative pressures ahead of the predictable adjustments of the parity. This mechanism did not prevent large short-term capital inflows in 1994– 1995. After some devaluation, the regime of ad hoc adjustment was replaced by a crawling band. Indeed, this regime succeeded in bringing inflation down below 10% in 1999 (Jonas and Mishkin, 2003). Unlike the Czech Republic and Poland, the Hungarian authorities continued to view the narrow fluctuation band as a useful nominal anchor. The band helped to reduce inflation and anchor inflation expectations; it also avoided excessive real appreciation. However, Hungary was forced to deal with the problems caused by large capital inflows. In May 2001, the authorities finally decided to widen the fluctuation band around the parity against the Euro. The crawling peg was completely abandoned in October 2001. Before, in August 2001, the National Bank of Hungary (NBH) explained that for the next couple of years, it would be using the inflation targeting system to achieve a gradual reduction of inflation to a level corresponding to price stability. The inflation target is surrounded by a 1% tolerance band, which was applied until 2002. The Hungarian policy mix was successful in providing internal and external stabilities of the economy (Dibooglu and Kutan, 2001; and Barlow, 2005). 3. Theoretical foundations There are several advantages in relying on the SVAR specification for analyzing monetary policies. In particular, it allows modelling non recursive structures of the economy and it facilitates the interpretation of the contemporaneous correlations of disturbances. Indeed, the decomposition of the variance–covariance matrix in the SVAR framework implies a recursive scheme among the variables that has clear economic implications and can be tested as any other relationship. The SVAR methodology suggests imposing various kinds of restrictions on the structural parameters only. The basic approach derives from the studies of Blanchard and Watson (1986), Bernanke (1986), Blanchard and Quah (1989), Shapiro and Watson (1988) and Blanchard (1989) on structural modelling. Unlike many SVAR model identification processes which define either short run (Kim and Roubini, 2000) or long run (Blanchard and Quah, 1989) restrictions, Gali (1992) and Bruneau and De Bandt (2003) favour a combination of both, in order to assess interactions between structural shocks and several economic variables on the short and long run. Gali (1992) was the first to implement a combination of two types of restrictions. He took as starting point Blanchard and Quah (1989) and Shapiro and Watson (1988) to define permanent and transitory dynamic effects of various disturbances on economic variables. However, Gali's study (1992) was carried out in the context of a closed economy. In this paper, we try to extend Gali's approach within a small open economy framework taking the exchange rate into account. Indeed, to analyze monetary transmission mechanisms in European Union accession countries, we propose a SVAR model including the industrial production index as a real output “proxy”, exchange rate3, consumer price index and short-term nominal interest rate for the Czech Republic, Poland and Hungary. Formally, the VAR representation of the reduced-form model can be written as ðDÞyt = et

3

Eðet eV t Þ = X:

The price of one unity of euro in domestic currency.

ð1Þ

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

We assume that y is a covariance stationary vector process. Each element of y can be expressed as a linear combination of current and past structural shocks. This Wold representation will be yt = C ðLÞet :

ð2Þ

Now, we assume that the VAR representation of the structural form can be written as: BðLÞyt = ut

Eðut uV t Þ = I:

The long run constraints R1: no long run effects of aggregates demand shocks on real output. R2: no long run effects of exchange rate shocks on real output. R3: no long run effects of monetary shocks on real activity.

ð4Þ

The short-term constraints R4: no contemporaneous effects of exchange rate shocks on inflation rate. R5: contemporaneous inflation rate does not enter the function of the monetary authority's reaction. R6: no contemporaneous effects of monetary shocks on exchange rate.

where A(L) = B(L)− 1. The structural MA representation in Eq. (4) is also called the final form of an economic model because the endogenous variables yt are expressed as distributed lags of the exogenous variables, given by the elements of ut. However the exogenous structural shocks ut are indirectly observed through their effects on the elements of yt. We can obtain the structural shocks ut by first estimating the reduced-form VAR (1) and transforming its residuals. From Eqs. (2) and (4) we have A(L)ut = C(L)εt, therefore4 Ai = Ci A0 :

ð5Þ

The identification of the structural VAR is achieved by imposing n (n − 1) / 2 restrictions, usually taken from economic theory and intended to represent some meaningful short run or long run relationships between the variables and the structural shocks. Short run restrictions are imposed directly on A0 which describes the contemporaneous reaction of the variables to structural innovations. The long run restrictions on the coefficients of A(1)5 can also be used for identification. From Eq. (5) we obtain C ð1ÞA0 = Að1Þ

ð6Þ

where C(1) and A(1) represent the cumulated effects of innovations. Again, C(1) can be obtained from the estimation of the reduced-form system. Therefore, restrictions on A(1) can be used to identify A0. Shapiro and Watson (1988) and Blanchard and Quah (1989) are early examples in which long run restrictions are used in order to identify structural VAR. The structural shocks are identified from their reduced-form counterparts by imposing restrictions. Each restriction incorporates a different assumption that identifies the causal influence of monetary policy on output, price, exchange rate and interest rates. The last equation in the matrices (A(1) and A0) relies on a version of the Taylor rules for open economies (Taylor, 2001) which takes into account the role of the exchange rate and explicitly includes forward-looking elements (Clarida et al., 1998):
 it = i + α 1 Et Δpt

+ 12



− Δpt

 + 12

+ α 2 Et yt

+ 12

rate. The role of exchange rate changes in monetary policy rules is discussed in the theoretical literature. Ball (1999, 2002) argues that large shifts in the exchange rate produce large fluctuations in output and inflation through import prices. Consequently, direct inflation targeting without explicit attention to the exchange rate is dangerous in an open economy. The salient features of the identification strategy can be summarized by the following table:

ð3Þ

The structural MA(∞) representation can be written as yt = AðLÞut ;

995

+ α 3 ðet − et − 1 Þ

+ ums;t : The Taylor rule is “forward- looking” in the sense that the CPH National banks react to deviations of forecasts of inflation and output from their respective targets and to transitory changes of the exchange

4 Let subscripts indicate the matrix of coefficients at the corresponding lag. As C0 =1and Eq. (5) must hold for all t, we have: A0ut =εt. Squaring both sides and taking expectations yields: A0A′=Ω. Combining Eq. (5) and A0A′=Ω we find: A(L)ut =C(L)A0ut. 0 0 5 Structural vector MA representation.

This set of restrictions imposes the following ordering of variables (the industrial production, yr, the consumer price index p, the nominal exchange rate e, and a short-term domestic interest rate, i) and shocks: Variables: ½ yr ; p; e; i 

Shocks: ½us ; ud ; uers ; ums :

Therefore, we have the following restrictions on the matrices A(1), A0 or B0. 2

* 0 0 0

6 6* * Að1Þ = 6 6 4* * * *

3

2

* * *

*

* * *

*

3

*

6 7 7 *7 6* * 0 07 6 7: 7 A0 = B1 0 = 6 7 7 *5 4* * * 05

*

*

*

4. Empirical results In this section, we report the results of estimating SVAR models for the Czech Republic, Hungary and Poland. In our empirical analysis, monthly data are used. The sample period is from January, 1998, to September, 2007. The choice of 1998 as departure year presents three important advantages. First, it covers the inflation targeting framework implementation periods (long) in the three examined countries. Second, it allows us to avoid the beginning of the transition period, which is characterized by movements and turbulence. And third, it does not call for many statistics sources6, which minimizes disturbances coming from the differences or the error measure. We use four variables for each country, namely industrial production, consumer price index, short-term domestic interest rate and nominal exchange rate with respect to the euro. The data are taken from OCDE and Eurostat databases. Most series for the countries in our sample appear to be integrated of order one, I(1) according to ADF, PP and KPSS tests. 4.1. Expected effects of a monetary contraction What are the expected movements of the macroeconomic variables following a monetary contraction? Some of them, such as the absence of a “price” or “liquidity puzzle7, serve as quite strict criteria for judging the validity of our empirical models since it may be difficult to tell whether our identified monetary policy shocks are close to true monetary policy shocks if such puzzles persist (Kim and 6

Only three in our case. This behavior means that the inflation rate increases following a monetary contraction shock. 7

996

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

Fig. 1. Czech Republic impulse response.

Roubini, 2000). In a monetary contraction, initially interest rates rise and initially monetary aggregates fall. These are general indications of a tighter monetary policy stance. An initial rise in interest rates may be reversed in the very short run due to deflationary pressure from a monetary contraction; however, the initial impact must be a rise of the interest rate and a fall of the money supply. Second, the price level declines and the output level does not increase. We may observe an output increase or a price level increase after a monetary contraction, but if the monetary contraction is really exogenous in the sense that it is not a systematic response to any shock8, the output and price level should decrease. However, for the output response some models predict no real effect of monetary policy; even in the short run, therefore the empirical results provide evidence on whether money is non-neutral in the short run and on the magnitude of such potential real effects. What are the effects of monetary contraction on the exchange rate? The effect of an interest rate increase on the exchange rate depends on the disturbance that leads to the change in interest rates. A tightening monetary policy that increases domestic interest rates for a given expected inflation rate will lead to a nominal appreciation of the exchange rate. However not all increases in interest rates will be

8 For example, oil shocks, inflationary pressure, money demand shocks, or foreign policy shocks.

associated with a currency appreciation: if there is an increase in the expected inflation, the ensuing Fisherian increase in the nominal interest rate would be associated with an impact depreciation of the exchange rate. Therefore the impact response of the exchange rate to an increase in the interest rate will depend on whether it is the nominal or the real interest rate that is increasing. In this study, we will try to illustrate the role of exchange rate in the monetary policy transmission process through the impulse response analysis. 4.2. Impulse responses to exchange rates and monetary shocks In Fig. 1, we display the estimated impulse response functions in the Czech Republic for all four variables together with the respective 95% confidence intervals.9 Indeed, the industrial production declines significantly in response to interest rate hike (monetary tightening). As predicted by theory we find negative and persistent responses for price level. This monetary shock effect on price level is statistically significant at the 95% level. However, the interest rate increases trigger a significant appreciation of the local currency. This effect is not

9 The exchange rate and monetary shocks are expressed through the first and second columns respectively. The rows express responses of industrial production, price level, exchange rate and interest rate respectively to exchange rate and interest rate shocks.

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

997

Fig. 2. Polish impulse response.

persistent for a long time only for eight months. The exchange rate plays an important role as a transmission channel of interest rate effects on prices and output. Indeed, the persistent and significant decline in output is not only due to the rise in short interest rate but also to the exchange rate appreciation. The both long-lived and statistically significant level responses suggest that the interest rate affects the price level through the exchange rate channel. Consequently, we can argue that, in the context of inflation targeting, the interest rate is no more the main policy instrument, it must be consolidated by the exchange rate policy. Following a monetary contraction, the industrial production response dynamic in Poland is quite similar to the Czech Republic (Fig. 2) but, it is not significant. However, the negative response of the price level occurs latter and is not significantly different from zero at the 95% level in contrast with Czech Republic's case. However, the monetary policy tightening led in short term a significant appreciation of the exchange rate over one year. The exchange rate appreciation has worsened the economic recession (industrial production fall) although non significant. These results suggest that interest rate is not an effective monetary transmission channel in Poland. In contrast, the exchange rate channel seems to affect significantly the price level. Therefore, as opposed to the Czech Republic, in the case of Poland the exchange rate transmission channel appeared more important. In Hungary, the industrial production responses are negative and statistically significant before returning to its pre shock level. This return occurs about one year after the monetary contractionary shock. The peak magnitude of − 0.5% is reached after eight months and is statistically significant. However, the price level response shows a

sharp decline, followed by a significant stabilization. As in the case of Poland, the direct linkages between the interest rate and the price level are not explicit in short term because of its non significant effect. Contrary to Polish price level dynamic, in medium and long term the Hungarian one declines significantly. Therefore, the monetary policy contraction induces a fall of price level in Czech Republic and Hungary. Thereby, the role of the exchange rate as a monetary transmission channel on the price level appears to be not particularly strong in Hungary. In order to gauge the direct role played by the exchange rate in the transmission mechanism, we investigate the effects of the nominal exchange rate shock on all variables. Except for the Czech Republic, these effects appear globally significant. Indeed, as shown by Fig. 3 (for Hungary), following a depreciation of one percentage point, the decline of the prices level is quite quick but not persistent. For the Poland, the Fig. 2 displays the same phenomenon that is a significant fall in prices level. However, the responses of industrial production and price level are not significant for Czech Republic. In sum, these findings suggest that, for Hungary and Poland the exchange rate plays a direct and important role in the transmission of the monetary policy and can act on interest rates to offset some of the tightening in the monetary conditions.

4.3. Decomposition of variance Tables 1a,b,c show variance decompositions for industrial production and price level. Initially, except for Hungary, the monetary shocks

998

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

Fig. 3. Hungarian impulse response.

are the most important cause of variations in industrial production. Supply shock becomes more important after two years. The variance decomposition for the Czech Republic price level gives us a hint of the relative significance of the exchange rate as an important channel in monetary transmission, which is opposite to our impulse response findings. The largest proportion of price level variance after two years is attributed to the demand (49%) and

exchange rate (39%) shocks. The proportion of price level variation attributed to monetary shocks is low (0.09%). For Poland, the variability of industrial production is mainly due to monetary shocks. The price level fluctuations come from demand and exchange rate shocks. The contribution in short and long terms of the exchange rate shock is very weak for Hungary situation. These results are less consistent with evidence from the impulse response function showing the important role of the exchange rate in

Table 1a Variance decompositions of Czech variables.

Table 1b Variance decompositions of Polish variables.

Components

Supply

Demand

Horizons

Exchange rate

Monetary

Components

Supply

Demand

Exchange rate

Monetary

Horizons

Output (industrial production) 1 month 0.3173 12 months 0.4848 24 months 0.6216 36 months 0.7175

0.0528 0.0347 0.0305 0.0229

0.1748 0.2098 0.1476 0.1091

0.4550 0.2707 0.2003 0.1504

Output 1 month 12 months 24 months 36 months

0.0513 0.3244 0.6479 0.7770

0.0102 0.0338 0.0180 0.0127

0.0488 0.0879 0.0489 0.0310

0.8897 0.5539 0.2852 0.1793

Price level 1 month 12 months 24 months 36 months

0.9052 0.7257 0.4921 0.5034

0.0000 0.1187 0.3903 0.3538

0.0000 0.1397 0.0944 0.0886

Price level 1 month 12 months 24 months 36 months

0.1635 0.2998 0.3248 0.3423

0.8365 0.3347 0.2662 0.2492

0.0000 0.1336 0.2039 0.2118

0.0000 0.2320 0.2052 0.1967

0.094 0.0159 0.0231 0.0542

T. Mamoudou et al. / Economic Modelling 26 (2009) 993–999

paper. Toure Mamoudou thanks Robert Schuman Foundation for the financial support.

Table 1c Variance decomposition of Hungarian variables. Components

Supply

Demand

Horizons

Exchange rate

Monetary

References

Output (industrial production) 1 month 0.7131 12 months 0.6427 24 months 0.7438 36 months 0.7939

0.1196 0.1525 0.1074 0.0913

0.1649 0.1997 0.1447 0.1117

0.0025 0.0051 0.0041 0.0032

Price level 1 month 12 months 24 months 36 months

0.8410 0.6646 0.4376 0.4202

0.0000 0.1436 0.2428 0.2294

0.0000 0.0070 0.0181 0.0174

0.1590 0.1848 0.3015 0.3330

999

the monetary policy transmission process, especially for Poland and Hungary. 5. Concluding remarks This paper examines the impact of the inflation targeting as the monetary policy implemented in the Czech Republic, Poland and Hungary respectively from 1998, 1999 and 2001 on the price level and industrial production dynamics. An interesting aspect of the analysis is the evidence that the effects of the exchange rate on the price level can be considered as in some cases, the fastest channel from monetary policy to price level in the context of inflation targeting. Our empirical results confirm the important role of exchange rate, especially for Poland and Hungary. These empirical findings indicate that, for the Czech Republic, the link between interest rate and price level is rather indirect. It uses the exchange rate channel to reduce the price level. The main question is how the new inflation targeting framework might become successful for these countries. There are some arguments suggesting more flexibility in the inflation targeting strategy in order to avoid an exchange rate overuse induced by the uncertain policy environment. Indeed, policy-makers should focus their policy instruments on both the interest rate and the exchange rate by adopting a Monetary Condition Index (MCI). They should also choose “long run inflation” as a target variable in order to avoid the transitory effects of exchange rate fluctuations induced by excessive volatility of import prices. Acknowledgements We are grateful to the editor, Stephen George Hall and anonymous referees for the comments and suggestions that have improved the

Aglietta, M., et al., 2003. les PECO devant la tentation de l'euro. Economie internationale, 2003/1 n°93, pp. 11–36. Ball, Laurence, 1999. Policy Rules for Open Economies. In: Taylor, J. (Ed.), Monetary Policy Rules. Chicago, chap. 3, pp. 129–156. University of Chicago Press. Ball, Laurence, 2002. Policy rules and external shocks. In: Loayza, N., Schmidt-Hebbel, K. (Eds.), Monetary Policy: Rules and Transmission Mechanisms, Central Bank of Chile, pp. 47–62. Barlow, D, 2005. The Hungarian exchange rate and inflation over the transition. Economic Systems 29, 87–97. Bernanke, B., 1986. Alternative explanations of the money-income correlation. In: Brunner, K., Metzler, A.H. (Eds.), Real Business Cycles, Real Exchange Rates, and Actual Policies. Carnegie-Rochester Series on Public Policy 25. North-Holland, Amsterdam, pp. 49–99. Blanchard, 1989. A traditional interprétation of macroeconomic fluctuations. American Economic Review 79, 1146–1164. Blanchard, O.J., Watson, M.W., 1986. Are business cycles all alike? In: Gordon, R.J. (Ed.), The American Business Cycle: Continuity and Change, National Bureau of Economic Research Studies in Business Cycles vol. 25. University of Chicago Press, Chicago, pp. 123–182. Blanchard, O., Quah, D., 1989. The dynamic effects of aggregate demand and supply disturbances. Am. Econ. Rev. 79, 655–673. Bruneau et de Bandt, 2003. Monetary and fiscal policy in the transition countries to EMU: what do SVAR models tell us. Economic Modelling 20, 959–985 (2003). Cecchetti, S., Rich, R., 2001. Structural estimates of the US sacrifice ratio. Journal of Business and Economics Statistics 19 (4), 416–427. Clarida et al., 1998. ‘Monetary policy rules in practice Some international evidence European Economic Review, Volume 42, Issue 6, June 1998, Pages 1033–1067. Dibooglu, S., Kutan, A.M., 2001. Sources of real exchange rate fluctuations in transition economies: the case of Hungary and Poland. Journal of Comparative Economics 21 (13), 1–19. Elbourne, A., de Haan, J., 2005. Financial Structure and Monetary Policy Transmission in Transition Countries”, mimeo, University of Gronigen, Department of Economics. Frommel, M., Schobert, F., 2006. Exchange rates regimes in Central and East European countries: deeds and word. Journal of Comparative Economics 34 (3)(September), 467–483. Gali, 1992. How well does the IS-LM model fit postwar US data. Quaterly Journal of Economics 709–738. Golinelli, R., Rovelli, R., 2006. Monetary policy transmission, interest rate rules and inflation targeting in three transition countries. Journal of banking and Finance 29 (1)(January), 183–201. Holub, T., 2004. Foreign Exchange Interventions under Inflation Targeting. Prague: CNB. Internal Research and Policy Note, 1/2004. Jarocinski, 2006. Response to Monetary Policy Shocks in the East and the West of Europe: A Comparison. Jonas, J. et Mishkin, F.S., 2003. ‘Inflation targeting in transition countries: experience and prospects’, NBER working paper no 9667. Kim et Roubini, 2000. Exchange rate anomalies in the industrial countries: a solution with a structural VAR approach. Journal of Monetary Economics 45, 561–586 (2000). Orlowski, L.T., 2003. Monetary convergence and risk premiums in the EU accession countries. Open Economies Review 14 (3), 251–267. Orlowski, L.T., 2004. Money rules for monetary convergence to the euro. Journal of Policy Modeling 26, 817–837 (2004). Shapiro et Watson, 1988. Sources of business cycle fluctuations. In: Fisher, S. (Ed.), NBER Macroeconomic 1988. The MIT Press, Cambridge (Mass.). and London. Taylor, 2001. The role of the exchange rate in monetary policy rules. The American Economic Review 91 (2), 263–267.