stock market synchronization and monetary integration

Journal of International Money and Finance 30 (2011) 96–110

Contents lists available at ScienceDirect

Journal of International Moneyand Finance

journal homepage: www.elsevier .com/locate/ j imf

Stock market synchronization and monetary integration

Sébastien Wälti*

Swiss National Bank, Börsenstrasse 15, 8022 Zurich, Switzerland

JEL classification:E44F15F21F36G15

Keywords:Stock marketsCorrelationComovementMonetary unionExchange rate regime

* Tel.: þ41 44 6313943.E-mail address: [email protected].

0261-5606/$ – see front matter � 2010 Elsevier Ltdoi:10.1016/j.jimonfin.2010.07.004

a b s t r a c t

This paper focuses on the relationship between stock marketcomovements and monetary integration. A panel specificationis used to explain bilateral stock market return correlationsbetween fifteen developed economies over the period 1975–2006.Time fixed effects are included to capture global shocks andwe also examine the role of bilateral trade linkages and interna-tional financial integration. Monetary integration leads to strongerstock market synchronization, both through the elimination ofexchange rate volatility and through the common monetary policyand the convergence of inflation expectations. Trade and financialintegration also contribute to higher stock market returncomovements.

� 2010 Elsevier Ltd. All rights reserved.

1. Introduction

Stock market synchronization refers to the fact that national stock markets display significantcomovements. The degree of comovement or correlation between national stock market returnsremains very important for international portfolio diversification. This correlation is likely to changeover time as a result of rising economic interdependence through international trade and internationalfinancial markets. Stock market synchronization will also be affected by monetary integration, forexample through the removal of exchange rate volatility. Taking into account rising economic inter-dependence, this paper focuses on the impact of monetary integration on cross-country stock marketreturn correlations.

d. All rights reserved.

mailto:[email protected]

www.sciencedirect.com/science/journal/02615606

www.elsevier.com/locate/jimf

http://dx.doi.org/10.1016/j.jimonfin.2010.07.004

http://dx.doi.org/10.1016/j.jimonfin.2010.07.004

S. Wälti / Journal of International Money and Finance 30 (2011) 96–110 97

Most of the available evidence compares mean correlations between national stock market returns,before and after the creation of the euro (Adjaouté and Danthine, 2003, 2004). The hypothesis that theeuro has led to higher stock market comovements should translate into higher correlations betweenEMU participants’ returns and the EMU return after the creation of the euro, and relatively stablecorrelations for European non-EMU countries across both sub-periods. Lane and Wälti (2007) showthat correlation coefficients between the returns of most individual EMUmembers and the EMU returnhave generally increased after the introduction of the euro. Although these results could be interpretedas evidence that the euro has led to greater stock market comovements, they could also arise fromamore general tendency towards trade and financial integration at the global level. In fact, correlationsbetween the respective returns of the United Kingdom, Sweden and Switzerland, and the EMU returnhave also increased after the introduction of the euro.

This informal evidence raises several issues of interest. First, any analysis of the relationshipbetween asset return comovements and monetary integration should consider both participating andnon-participating countries into the process of monetary integration. Focusing too closely on EMUparticipants could lead us to attribute a role to the euro when other excluded factors also matter.Second, mean correlations over long time periods may not provide adequate evidence about stockreturn comovements because they could hide significant time variation. We should instead focus ontime-varying correlation coefficients (Cappiello et al., 2006). Finally, stock market return correlationscontain both a common component and an idiosyncratic component. In other words, stock marketscould be synchronized because they are affected by common international shocks, or because idio-syncratic shocks are transmitted across highly integrated economies. Therefore, any model of stockreturn comovements must allow for both sources of synchronization, and for different propagationchannels of country-specific shocks.

This paper relies on a panel specification to explain time-varying bilateral stock market returncorrelations between fifteen developed economies over the time period 1975–2006. The panel spec-ification is naturally appropriate since it allows for time fixed effects to control for common interna-tional sources of comovement, to control for propagation channels such as international trade linkagesand international financial integration, and to assess the impact of monetary integration on bilateralcorrelations. We do not restrict our analysis to the short time period surrounding the creation of theeuro. Instead, we take advantage of the European experiment with monetary integration during thepast thirty years or so, starting with the creation of the European Monetary System in 1979. Oureconometric specification also takes account of the fact that trade integration and financial integrationcould be endogenous variables. Exogenous instruments based on geography and institutions are usedas a remedy.

We find strong evidence that trade integration and international financial integration raise stockmarket return comovements. Monetary integration also increases bilateral correlations, both throughthe elimination of exchange rate volatility and through the common monetary policy and theconvergence of inflation expectations. Our sensitivity analysis shows that these results are robust tochanging the currency denomination of returns, whether returns are real or nominal, and acrossdifferent regimes of global volatility. However, the effect of joint EMU participation on stock returncorrelations is limited to countries which are already tightly integrated in terms of trade linkages andfinancial integration.

The remainder of this paper is organised as follows. Section 2 reviews the ambiguous theoreticalpredictions regarding the impact of different types of integration on synchronization. Section 3presents the econometric specification and discusses the measurement of the variables which areused in our regressions. Section 4 presents the results from our estimations and provides somefurther interpretation. Section 5 deals with robustness checks and Section 6 contains concludingremarks.

2. Theoretical background

Comovement results from global shocks – for example sharp changes in the level or the volatility ofoil prices – as well as the transmission of country-specific shocks through various economic linkages

S. Wälti / Journal of International Money and Finance 30 (2011) 96–11098

such as trade in goods and financial assets. This section reviews the predicted theoretical effects ofmonetary integration, trade integration, and financial integration. Trade and finance affect businesscycle synchronization, which in turn should translate into stock market comovements. Theoreticalpredictions remain largely ambiguous and assessing the contribution of specific economic linkages tostock market synchronization remains ultimately an empirical question.

2.1. Monetary integration

In general, wewould expect monetary integration to enhance stock market synchronization. Lowerexchange rate volatility means lower transaction costs in cross-border investment. Moreover, partic-ipation into a monetary union implies a single monetary policy and convergence of inflation expec-tations. Consequently, real risk-free rates should converge and lead to more homogeneous valuations.Finally, lower exchange rate volatility could lead to enhanced business cycle synchronization, therebyleading to higher stock market comovements. This being said, monetary authorities could also useexchange rate flexibility to reduce themacroeconomic impact of foreign real shocks, thereby deliveringlower output comovements across countries.

Bordo and Helbling (2004) find that although a fixed exchange rate induces higher output corre-lations, this result is not robust to the inclusion of other control variables such as trade linkages ora European Union dummy variable. The authors thus conclude that fixing by itself does not make anydifference for the degree of synchronization of business cycles. Rose and Engel (2002) reach theopposite conclusion and show that currency unions bring about higher business cycle synchronization,even after controlling for other factors including trade relations. Bodart and Reding (1999) distinguishbetween ERM and non-ERM countries and make use of a GARCH specification for bond and stockmarket returns. They find evidence that bond and stock markets correlations depend negatively onexchange rate variability. Overall, we would conclude that exchange rate volatility affects synchroni-zation but it remains necessary to control for other factors such as trade linkages and financialintegration.

2.2. Trade linkages

Although international trade in goods is usually thought of as fostering business cycle synchro-nization, its overall effects remain theoretically ambiguous. On the demand side, higher aggregatedemand in one country will partially fall on imported goods, thereby raising the output and incomeof trading partners and inducing output comovements across countries. On the supply side, however,there are two opposite effects which relate to two different approaches to modeling internationaltrade. Intra-industry models of trade consider economies with similar production structures andfactor endowments. To the extent that trade occurs mostly within industries, an expansion in someindustries will raise output comovements across countries. However, trade integration may also leadeconomies to specialize in the production of goods for which they have a comparative advantage,hence reducing comovements. The net impact of international trade on comovements is thereforeambiguous.

Frankel and Rose (1998) find strong evidence that closer trade linkages lead to an increase inthe correlation of business cycles. Calderon et al. (2007) find similar evidence for developingcountries, for which we could expect that specialization along the lines of comparative advantageis more important. Otto et al. (2001) and Bordo and Helbling (2004) conclude that internationaltrade affects output comovements in a positive and significant way, although it does not explainvery much. Imbs (2004) refines the analysis of the impact of international trade by estimating therespective contributions of both types of trade effects and concludes that a sizable part of theimpact of trade on bilateral correlations works through intra-industry trade, although there aresome smaller but significant inter-industry effects. Chinn and Forbes (2004) conclude that directtrade linkages remain the predominant determinant of the effect of large stock markets on otherfinancial markets.


2.3. Financial linkages

At a theoretical level financial integration carries an ambiguous impact on business cyclesynchronization. On the one hand, to the extent that equities of a given country are widely heldinternationally, a fall in that country’s stock market will trigger a negative wealth effect for assetholders in the world, thereby affecting consumer demand and in turn, output comovements. On theother hand, international diversification of portfolios allows to smooth consumption patterns withouthaving to diversify production, thereby leading to the possibility of greater specialization. The formereffect would increase business cycle synchronization, whereas the latter effect would tend to reducecomovements.

Empirical evidence on the role of financial linkages for business cycle synchronization is somewhatmixed. Bordo and Helbling (2004) conclude that financial integration does not affect business cyclesynchronization. Imbs (2004, 2006) and Kose et al. (2003) show that financial integration impactspositively on business cycle comovements. However, Kalemli-Ozcan et al. (2001) show that capitalmarket integration leads to greater specialization in production structures and thus, a lower degree ofoutput comovements.

3. Econometric specification and data

This paper assesses the impact of common international shocks, trade integration, financial inte-gration and monetary integration on bilateral time-varying correlations between the stock marketreturns of fifteen developed economies over the period 1975–2006.1 The baseline regression model isa panel specification allowing for time fixed effects in order to control for common internationalshocks. This specification is written as

ri;j;t ¼ gt þ b1Ti;j;t þ b2Fi;j;t þ b3Mi;j;t þ hi;j;t (1)

where ri,j,t is the correlation coefficient between the stock market returns of country i and country j foryear t, gt captures common international factors, Ti,j,t measures the intensity of trade relations betweencountry i and country j during year t, Fi,j,t stands for the degree of international financial integrationbetween country i and country j for year t, and Mi,j,t measures the degree of monetary integrationbetween country i and country j for year t. The remainder of this section discusses the measurement ofthese variables.

3.1. Measuring comovements

It is common practice to measure stock market comovements by the correlation coefficientbetween two series of stock returns over a given time period. We make use of weekly averages ofdaily MSCI stock market indices and calculate stock market returns by log-differentiation. Dailydata suffer from non-synchronous trading hours, while monthly data do not provide enoughinformation to compute yearly correlation coefficients. Stock market indices are retrieved fromDataStream and expressed in local-currency.2 Correlation coefficients are calculated for each yearbased on the weekly observations contained in this year, thereby avoiding potential issuespertaining to overlapping observations. Since correlation coefficients are not normally distrib-uted, we follow Otto et al. (2001) and adopt the following transformation of the dependentvariable:

1 These countries are the United States, the United Kingdom, Austria, Denmark, France, Germany, Italy, the Netherlands,Norway, Sweden, Switzerland, Canada, Japan, Spain and Australia.

2 Other studies have sometimes used returns expressed in a common currency, usually the US dollar, thereby taking theperspective of an international investor. In our case, however, this induces a bias towards finding a positive effect of monetaryintegration on pair-wise correlations. I am grateful to a referee for pointing this out.


wi;j;t ¼ ln1þ ri;j;t (2)
1� ri;j;t
3.2. Measuring trade linkages

Frankel and Rose (1998) and Bordo and Helbling (2004) measure bilateral trade intensity as thesum of exports and imports between countries i and j during year t, divided by the sum of totalexports and imports of each country. Bilateral trade data are gathered from the IMF’s Direction ofTrade Statistics. Making use of bilateral trade intensity as an explanatory variable gives rise toa problem of endogeneity. Countries will tend to link their currencies with their most importanttrading partners. We make use of a two-step approach to focus on the contribution of bilateralinternational trade to stock market comovements. The first step consists of regressing the measureof bilateral trade intensity on five exogenous determinants identified in the literature on gravityequations of international trade: the natural logarithm of the distance between the main businesscenters of each country, the logarithm of the product of country sizes as measured by grossdomestic product, a dummy variable for a common border, a dummy variable for commonlanguage, and a dummy variable for joint EU membership. The second step consists of introducingthe predicted values of bilateral trade intensity into the panel specification for bilateralcorrelations.

3.3. Measuring financial integration

Increasing the degree to which both domestic and foreign residents are allowed to acquiredomestic and foreign assets is only a necessary but not a sufficient condition for internationalfinancial integration. Other factors such as investment opportunities, institutional characteristicsand political stability are also important. Moreover, Bekaert and Harvey (2003) have noted thatthe announcement of financial liberalization may not coincide with the completion of itsimplementation, so that international capital flows will start rising some time after theannouncement of the liberalization. Therefore, the degree of financial market integration can bemeasured along several different dimensions and there is no widespread agreement abouta single correct measure (Adam et al., 2001; Baele et al., 2004). De jure measures of financialmarket integration rely on the dating of financial market liberalizations initiated by policymakers,whereas de facto measures focus instead on the outcomes of such liberalizations. Since the impactof policy decisions will develop into outcomes gradually over time, it is likely that de jure and defacto measures will provide different views about the extent of financial market integration ata given point in time.

A higher level of internationalfinancial integrationwill result in higher cross-border holdings (stocks) offoreign assets. Lane andMilesi-Ferretti (2001, 2007) propose tomeasure the level of international financialintegration for a given country bydividing the sumof its total foreign assets and total foreign liabilities by itsgross domestic product. Our measure of bilateral financial integration is calculated as the logarithm of theproductof twocountries’ respectivemeasuresoffinancial integration. Theuseof thisbilateralvariable raisesanobviousproblemofendogeneitysinceourdependentvariableconsistsofbilateral returncorrelationsandinternationalfinancial integration ismeasured according to quantities of foreign assets. Again,wemakeuseof a two-step approach where the first step consists of regressing the measure of international financialintegration on a set of exogenous determinants. Imbs (2006)makes use of (among others) the logarithm ofthe product of gross domestic products per capita, an indicator of creditor rights, and an indexof corruptiontaken from La Porta et al. (1998). Lane and Milesi-Ferretti (2008) show that bilateral cross-border equityholdings depend upon distance and a common language. Portes and Rey (2005) argue that the formervariablecouldcapture theextentof informationalasymmetries involved incross-border investment. Finally,wealso includeadummyvariable for jointEUmembershipsince therehavebeenseveraldirectivesat theEUlevelaimingatachieving the liberalizationof capitalmovements.Thesecondstepconsistsof introducing thepredicted values of international financial integration into the panel specification for bilateral correlations.

Table 1Gravity equations for bilateral trade intensities.a,b

Regressors (I) (II)

Constant term �19.60*** (0.37) �19.52*** (0.37)Distance �0.60*** (0.01) �0.59*** (0.01)Product of GDPs 0.36*** (0.01) 0.36*** (0.01)Common language 0.21*** (0.04) 0.23*** (0.04)Common border 0.61*** (0.04) 0.62*** (0.04)EU dummy 0.08*** (0.03)R-squared 0.710 0.711Number of obs. 3360 3360

a *Significant at 10%; **significant at 5% level; ***significant at 1% level.b Robust standard errors (in parentheses) are used for all estimations.


3.4. Measuring monetary integration

The European experience with monetary integration provides for an opportunity to assess itsconnection with asset return comovements. It started with the creation of the European MonetarySystem in 1979 and culminated with the adoption of a single currency in 1999. Our first measure ofmonetary integration is an indicator of exchange rate stability. It is calculated as the yearly standarddeviation of weekly bilateral exchange rate percentage changes within that year.

The central component of the European Monetary System was the Exchange Rate Mechanism. Animportant feature of this mechanism was joint intervention and mutual support. Exchange marketpressure would always affect two countries, and both countries would intervene jointly to eliminatemisalignments. Moreover, mutual support was guaranteed in the case of shortfalls in foreign exchangereserves. Consequently, monetary integration was more than simply achieving bilateral exchange ratestability. There was an underlying institutional mechanism that would support it. Accordingly, oursecond measure is a dummy variable for joint participation of two countries in the Exchange RateMechanism.

The creation of the monetary union goes beyond the coordination of national monetary policies,in that it entails a single monetary policy for all participating countries. Although it remains unclearwhether a single monetary policy will make national economies more synchronized, the conver-gence of inflation expectations and the convergence of nominal interest rates mean that valuationsshould become more homogeneous. Our third measure is a dummy variable for joint EMUmembership.

The introduction of these three variables into the regression specification provides a lot of infor-mation on the role of monetary integration. Having controlled for the reduction in exchange rate risk,the dummy variables for joint ERM membership and joint EMU membership will indicate thecontribution of the particular institutional mechanism underlying the exchange rate arrangement. It ispossible that the benefits of a currency union extend beyond the simple elimination of exchange raterisk.

4. Results

This section presents our baseline results. Since bilateral trade intensity and international financialintegration are potentially endogenous, we regress each of these variables on different sets of exog-enous variables. Table 1 and 2 contain the results for these preliminary estimations.3 Robust standarderrors are used for both regressions.

All coefficients for the gravity equation for bilateral trade flows carry the expected sign and arestatistically significant at the 1% level. Distance has a negative effect on bilateral trade flows, whereasa greater size, a common language, a common border and common EU membership contribute posi-tively to the intensity of trade relations. Together these five variables explain about 70% of the total

3 The sample contains fifteen countries for the time period 1975–2006. So we have (15*(15 � 1)/2)*32 ¼ 3360 observations.

Table 2Estimating equations for bilateral financial integration.a,b

Regressors (I) (II)

Constant term �31.94*** (0.58) �35.59*** (0.60)Product of GDPs per capita 1.86*** (0.03) 1.86*** (0.03)Distance �0.36*** (0.01) �0.19*** (0.02)Creditor rights 0.03*** (0.01) 0.01** (0.005)Corruption index �0.38*** (0.09) 0.12 (0.09)Common language 0.31*** (0.05) 0.49*** (0.05)EU dummy 0.92*** (0.04)R-squared 0.539 0.589Number of obs. 3360 3360



variation in trade intensity. The estimated equation for international financial integration alsoperforms very well empirically. All coefficients except one carry the expected sign and are statisticallysignificant. Our six explanatory variables explain about 60% of the total variation in bilateral financialintegration. Predicted values from both equations are retrieved and are used below as explanatoryvariables instead of the actual values.

Our main econometric specification is given by equation (1). The dependent variable is a trans-formed correlation coefficient since raw correlation coefficients are not normally distributed. Figs. A1and A2 in the Appendix show that the transformation brings about a statistical distribution closer tothe normal distribution. Several variants of our baseline specification are estimated to assess thestability of the results. Every estimation contains time fixed effects and makes use of robust standarderrors. Table 3 presents our results.

Our baseline specification performs well. All time fixed effects are statistically different from zeroand stable across specifications (see Table A1). Thus, part of the correlations arises from commoninternational effects. All F statistics are statistically significant, so we can always reject the nullhypothesis that the coefficients on our explanatory variables are jointly insignificant. The R2 statisticindicates that our simple specification explains about one half of the total variation in bilateralcorrelations.

Trade integration and financial integration contribute positively to stock market synchronization.Despite the ambiguous predictions from the theory, there is strong evidence that greater economicinterdependence between two countries raises bilateral stock market return correlations. Thecoefficients for these two variables remain stable across all baseline specifications. Our threeexplanatory variables measuring monetary integration are initially introduced sequentially and thensimultaneously. Joint EMU membership raises stock market comovements in a statistically signifi-cant manner. Beyond statistical significance, is this result economically significant? We can computethe marginal effect of EMU membership on the raw correlation coefficient by working backwards

Table 3Stock market comovements: baseline results.a,b

Regressors (I) (II) (III) (IV) (V)

Trade integration 0.17*** (0.01) 0.16*** (0.01) 0.18*** (0.01) 0.16*** (0.01) 0.15*** (0.01)Financial integration 0.03** (0.01) 0.04*** (0.01) 0.03** (0.01) 0.02 (0.01) 0.03** (0.01)EMU 0.37*** (0.08) 0.32*** (0.08)ERM �0.06** (0.03) �0.08** (0.03)Exrate volatility �0.09*** (0.02) �0.06*** (0.02)R-squared 0.501 0.509 0.501 0.504 0.511F statistic 101.12*** 103.67*** 98.71*** 98.45*** 99.07***Number of obs. 3360 3360 3360 3360 3360Time dummies Yes Yes Yes Yes Yes



from the transformed correlation coefficient (see the derivation in the Appendix). The coefficient forEMU membership in Specification (V) of Table 3 is equal to 0.32, implying that joint EMUmembership increases the raw correlation by 0.16. We would think that this magnitude iseconomically significant.

Lower exchange rate volatility is associated with greater stock market comovements. The jointsignificance of EMU membership and exchange rate volatility indicates that monetary integrationaffects stock market synchronization in two different ways. On the one hand, lower exchange ratevolatility reduces transaction costs arising from uncertainty about future exchange rates. On the otherhand, the common monetary policy and the convergence of inflation expectations lead to morehomogeneous valuations, thereby increasing comovements. The common currency may also raisebusiness cycle synchronization, although the evidence remains mixed (Rose and Engel, 2002; Camachoet al., 2006).

In contrast with our expectations, the coefficient for joint participation in the Exchange RateMechanism is negative and statistically significant. Although we do not have a definite explanation forthis result, frequent realignments in the first half of the 1980s and the 1992–1993-currency crisis haveaffected various subsets of participating countries and these bouts of instability could have prompteddivergent stock market movements.

5. Sensitivity analysis

The robustness of our baseline results is examined along various dimensions. The results ofthis sensitivity analysis are presented in Table 4. First, we check whether the currency denomi-nation makes a difference by using returns in US dollars (Specification (VI)). We also test whetherexpressing returns in real terms produces different results (Specification (VII) for local-currencyreturns, Specification (VIII) for US dollar returns).4 The sensitivity analysis delivers a clearconclusion. Expressing returns in a common currency or using real instead of nominal returnsdoes not affect our baseline results. Interestingly, the effect of bilateral exchange rate volatility onstock return correlations is larger when returns are expressed in a common currency. Thus, itappears that using the same currency denomination for all returns induces higher comovementby construction.

Second, Catao and Timmermann (2009) show that stock market returns are more correlatedduring episodes of high global volatility. Since there was a shift to high global volatility around thesame time as the introduction of the euro, one hypothesis is that the dummy variable for jointEMU membership could pick up the effect of higher global volatility on stock return correlations.This should not be an issue given our econometric specification. Shifts in global volatility, whichare common to all country pairs, will be captured by the time fixed effects and therefore, the EMUdummy variable will adequately capture the EMU effect on stock return correlations. Nevertheless,we have investigated the role of global volatility further by using three different measures ofglobal volatility: a dummy variable taking a value of one for years of high global volatility,5 theannual variance of the world stock market return (based on daily data from DataStream), and thespread between Moody’s Baa corporate bond yield and the three-month Treasury bill yield. Sincethese three variables are common to all country pairs, they are introduced one at a time and timefixed effects are omitted (Specifications (IX), (X) and (XI)). All three variables carry the expectedsign and are statistically significant. Higher global volatility is associated with higher returncorrelations. However, the effect of joint EMU membership remains positive (even higher) andhighly statistically significant.

4 Real returns are computed as the difference between weekly nominal returns and “weekly” inflation rates (constructed onthe basis of monthly inflation data from the OECD Main Economic Indicators). The formula used to obtain weekly inflation ratesfrom monthly inflation rates is given by (1 þ m)0.25 � 1, where m is the monthly inflation rate.

5 Catao and Timmermann (2009) have identified years of high global volatility using a regime-switching framework: 1975,1980, 1982–1983, 1986–1987, 1991, and 1998–2002. The correlation coefficient between our dummy variable for high volatilityyears and world stock market volatility is quite high, i.e. 0.64.

Table 4Stock market comovements: sensitivity analysis.a,b,c

Regressors (VI) (VII) (VIII) (IX) (X) (XI) (XII) (XIII) (IXV) (XV)

Tradeintegration

0.12*** (0.01) 0.15*** (0.01) 0.13*** (0.01) 0.09*** (0.01) 0.10*** (0.01) 0.08*** (0.01) 0.15*** (0.01) �0.10** (0.04) �0.08*** (0.03) 0.34*** (0.05)

Financialintegration

0.04*** (0.01) 0.03*** (0.01) 0.04*** (0.01) 0.24*** (0.01) 0.20*** (0.01) 0.24*** (0.01) 0.03*** (0.01) 0.55*** (0.08) �0.02 (0.02) �0.14*** (0.03)

EMU 0.28*** (0.08) 0.27*** (0.08) 0.29*** (0.07) 0.59*** (0.09) 0.53*** (0.09) 0.59*** (0.09) 0.36*** (0.08) 0.52*** (0.10) 0.02 (0.10) �0.45*** (0.16)ERM �0.05* (0.03) �0.07** (0.03) �0.05* (0.03) �0.16*** (0.04) �0.18*** (0.04) �0.17*** (0.04) �0.05* (0.03) �0.02 (0.07) 0.06 (0.04) �0.09 (0.08)Exrate

volatility�0.29*** (0.02) �0.05** (0.02) �0.29*** (0.02) �0.10*** (0.02) �0.12*** (0.02) �0.11*** (0.02) �0.08*** (0.02) �0.23*** (0.09) �0.03 (0.04) �0.01 (0.08)

Englishcommonlanguage

0.31*** (0.04)

Commonlegal origin

�0.07** (0.03)

Highvolatilitydummy

0.14*** (0.02)

Global stockvolatility

0.78*** (0.04)

Yield spread 0.02*** (0.008)R-squared 0.495 0.519 0.510 0.278 0.327 0.270 0.519 0.806 0.601 0.662F statistic 91.00*** 102.40*** 97.00*** 160.87*** 190.01*** 159.97*** 97.15*** 68.09*** 47.81*** 21.99***Number of

obs.3360 3360 3360 3360 3360 3360 3360 480 960 320

Timedummies

Yes Yes Yes No No No Yes Yes Yes Yes

a *Significant at 10%; **significant at 5% level; ***significant at 1% level.b Robust standard errors (in parentheses) are used for all estimations.c (VI): USD returns; (VII): real local-currency returns; (VIII): real USD returns; (XIII): both countries in the core; (IXV): one country in the core and the other in the periphery; (XV): both

countries in the periphery (see text for the definition of the core and the periphery).

S.Wälti

/Journal

ofInternationalM

oneyand

Finance30

(2011)96

–110104


Third, we add additional regressors to make sure that our explanatory variables are really capturingthe impact of monetary integration. Catao and Timmermann (2009) find that return correlations areconsistently stronger between English-speaking countries. A dummy variable taking a value of onewhen two countries have English as a common language is included (Specification (XIII)). Our evidenceconfirms that correlations are indeed higher among English-speaking countries but the impact ofmonetary integration remains the same. Moreover, a common origin of the legal system could meanmore homogeneous legislations and reduced transaction costs. We control for this effect by intro-ducing a dummy variable for a common legal origin taken from La Porta et al. (1998). Again, ourbaseline results do not change.

Finally, aggregating all country pairs together could hide important coefficient heterogeneity.Thus, we have estimated our baseline specification for three different sub-groups of country pairs.Splitting the sample should be motivated by some relevant, interesting question pertaining toheterogeneity. We investigate whether joint EMU membership has a greater impact on stock returncorrelations among already tightly integrated economies compared to less integrated countries. Wedefine two groups of European countries depending on whether they belong to the core or theperiphery of Europe. Looking at Europe from the sky at night, a blue banana crosses Europe fromLancashire to Tuscany. Accordingly, countries belonging to the core are the United Kingdom, theNetherlands, France, Germany, Switzerland and Italy. Countries belonging to the periphery areAustria, Denmark, Norway, Sweden and Spain. Including both EMU and non-EMU participants in eachgroup is crucial to identify the impact of EMU on stock market correlations. We estimate our baselinespecification for three sub-samples separately: country pairs involving two core countries, countrypairs involving two periphery countries, and country pairs involving one core country and oneperiphery country.6 Table 5 displays the average bilateral trade intensity and the average level ofinternational financial integration for each sub-sample of country pairs. Our core-periphery splittingstrategy captures different levels of economic integration very well. Countries in the core are moreintegrated between themselves than with the periphery. Countries in the periphery are even lessintegrated between themselves.

The estimation results for these three sub-samples show that coefficient heterogeneityshould be taken seriously (Specifications (XIII), (IXV) and (XV)). Joint EMU membership raisesstock return correlations for countries belonging to the core, but reduces stock marketcomovements for countries belonging to the periphery. Therefore, joint EMU participation raisesstock return correlations among already tightly integrated economies, but this result does notextend to the entire euro area. It is also interesting to note that trade and financial integrationhave different effects depending on whether countries belong to the core or the periphery.Trade integration raises stock return correlations among periphery countries but decreases themamong core countries. The opposite result obtains for international financial integration. Finally,time fixed effects are found to matter more among periphery countries than among corecountries.

6. Concluding remarks

This paper studies the connection between stock market return comovements and monetaryintegration. The recent European experience with monetary integration offers a unique opportunity toassess its impact on stock return correlations. We make use of a panel specification to control forcommon international shocks with time fixed effects, and to control for bilateral trade and financialintegration. Our sample contains fifteen advanced economies over the time period 1975–2006. Greatertrade linkages and stronger financial integration contribute to greater stock market comovements. Atleast part of this effect could be attributed to the effect of these two variables on business cyclesynchronization, in turn leading to stock market synchronization.

6 There are other possibilities, e.g. one country belonging to the core and the other to the rest of the world. But these casesare not interesting because the EMU dummy variable has only zeros. So we cannot obtain a regression coefficient for these.

Table 5Average level of integration for three sub-samples of country pairs.a

Core–core Core-periphery Periphery–periphery

Trade integration 0.038 0.014 0.016Financial integration 10.388 6.667 4.025Number of obs. 480 960 320

a Core: United Kingdom, Netherlands, France, Germany, Switzerland and Italy. Periphery: Austria, Denmark, Norway, Swedenand Spain.


Lower exchange rate volatility and joint EMU membership are associated with stronger stockmarket comovements. The joint significance of these two variables indicates that monetary integrationraises return correlations by reducing transaction costs coming from exchange rate uncertainty, andthrough the common monetary policy and the convergence of inflation expectations leading to morehomogeneous valuations. However, the effect of monetary integration applies selectively to countrieswhich are already tightly integrated.

One implication of our results pertains to globalization and the international transmissionof shocks. More and more countries are integrating into the global economy and country-specific shocks are therefore very likely to be transmitted internationally with greater scopeand strength. Not only will stock market comovements increase during episodes of financialturbulence, as the recent global financial crisis testifies, but cross-country linkages lead totransmission of country-specific shocks at all times. Our results also relate to the debateabout the benefits of international portfolio diversification. There is widespread historicalevidence that country factors dominate industry factors in explaining stock market returns(Heston and Rouwenhorst, 1994; Griffin and Karolyi, 1998; Rouwenhorst, 1999). Our findingthat monetary integration leads to stronger stock market return comovements does notnecessarily imply that the traditional approach of diversifying across countries should beabandoned in favor of diversification across industries. First, we should also assess the impactof monetary integration on industry return comovements for the same sample period. Second,some studies have concluded that the introduction of the euro coincides with a greaterdominance of industry factors (Brooks and Del Negro, 2002; Flavin, 2004). Yet, it remainsunclear whether the relative prevalence of industry factors is permanent or temporary. Thesample period in these studies includes only very few years after the introduction of the euro.Moreover, Catao and Timmermann (2009) have shown that industry factors were especiallyimportant between 1997 and 2002, a period characterised by significant stock market vola-tility. Since international stock returns alternate between times of low and high volatility, itwould not be surprising that country factors have become again more important after 2002.In fact, some recent evidence concludes that the dominance of industry factors around theturn of the century was a temporary phenomenon (Adjaouté and Danthine, 2004; Bekaertet al., 2009).

Acknowledgements

The views expressed in this paper are those of the author and do not necessarily represent those ofthe Swiss National Bank. I am grateful to an anonymous referee, Hans Genberg, Philip Lane, RobertoRigobon and Charles Wyplosz for useful discussions and suggestions, and to Philip Lane for providingdata on international financial integration. Part of the work on this paper was carried out while theauthor was at the Department of Economics, Trinity College Dublin, Ireland.

Appendix. Transformation of the dependent variable

Otto et al. (2001) suggest a transformation of the dependent variable given as

wi;j;t ¼ ln1þ ri;j;t1� ri;j;t


Figs. A1 and A2 show that the transformation brings about an empirical distribution that is closer tothe normal distribution. The coefficient b3 in equation (1) measures the effect of the presence ofa monetary union and/or a system of fixed exchange rates on the transformed dependent variable.However, the variable of interest is the original dependent variable, that is the bilateral raw corre-lation coefficient. It is therefore necessary to unscramble the transformation of the dependent vari-able to interpret the contribution of monetary integration to bilateral correlations. Starting from thetransformation,we obtain

050

100

150

200

Freq

uenc

y

−.5 0 .5 1Correlation coefficient (domestic currency returns)

Fig. A1. Distribution of raw correlation coefficients.

wi;j;t ¼ ln1þ ri;j;t1� ri;j;t

exp�wi;j;t

� ¼ 1þ ri;j;t1� ri;j;t

1þ ri;j;t ¼ exp�wi;j;t

��1� ri;j;t

�

ri;j;t ¼ exp�wi;j;t

�� 11þ exp

�wi;j;t

�

010

020

030

0Fr

eque

ncy

−1 0 1 2 3 4Transformed correlation coefficient (domestic currency returns)

Fig. A2. Distribution of transformed correlation coefficients.

Table A1. Baseline results: time effectsa.

(I) (II) (III) (IV) (V)

D1975 1.89*** 1.81*** 1.92*** 1.85*** 1.82***D1976 1.65*** 1.56*** 1.67*** 1.66*** 1.61***D1977 0.98*** 0.89*** 1.00*** 0.96*** 0.92***D1978 1.45*** 1.36*** 1.48*** 1.45*** 1.40***D1979 1.25*** 1.16*** 1.28*** 1.22*** 1.19***D1980 1.30*** 1.21*** 1.33*** 1.28*** 1.24***D1981 1.19*** 1.10*** 1.22*** 1.20*** 1.16***D1982 1.39*** 1.31*** 1.42*** 1.41*** 1.37***D1983 1.13*** 1.04*** 1.16*** 1.11*** 1.08***D1984 1.45*** 1.36*** 1.48*** 1.42*** 1.39***D1985 1.05*** 0.96*** 1.08*** 1.06*** 1.01***D1986 1.21*** 1.12*** 1.24*** 1.22*** 1.17***D1987 2.01*** 1.91*** 2.03*** 1.99*** 1.95***D1988 1.48*** 1.39*** 1.51*** 1.46*** 1.42***D1989 1.27*** 1.18*** 1.30*** 1.26*** 1.22***D1990 1.93*** 1.84*** 1.97*** 1.92*** 1.89***D1991 1.84*** 1.75*** 1.88*** 1.84*** 1.80***D1992 1.41*** 1.31*** 1.45*** 1.44*** 1.40***D1993 1.30*** 1.20*** 1.33*** 1.32*** 1.26***D1994 1.59*** 1.50*** 1.62*** 1.59*** 1.54***D1995 1.60*** 1.50*** 1.63*** 1.61*** 1.56***D1996 1.73*** 1.63*** 1.77*** 1.71*** 1.68***D1997 2.06*** 1.96*** 2.10*** 2.07*** 2.02***D1998 2.02*** 2.10*** 2.24*** 2.21*** 2.17***D1999 1.96*** 1.81*** 1.98*** 1.96*** 1.86***D2000 1.59*** 1.44*** 1.61*** 1.60*** 1.49***D2001 2.40*** 2.25*** 2.42*** 2.40*** 2.30***D2002 2.32*** 2.17*** 2.34*** 2.31*** 2.21***D2003 2.20*** 2.05*** 2.22*** 2.20*** 2.10***D2004 2.18*** 2.02*** 2.20*** 2.17*** 2.06***D2005 2.11*** 1.95*** 2.13*** 2.09*** 1.99***D2006 2.51*** 2.36*** 2.53*** 2.49*** 2.39***

a *Significant at 10%; ** significant at 5% level; *** significant at 1% level.


Table A2: Sensitivity analysis: time effectsa.

(VI) (VII) (VIII) (IX) (X) (XI) (XII) (XIII) (IXV) (XV)

D1975 2.24*** 1.71*** 2.28*** – – – 1.81*** 1.32*** 0.58*** 2.15***D1976 1.71*** 1.52*** 1.75*** – – – 1.61*** 1.00*** 0.19 1.91***D1977 1.13*** 0.89*** 1.18*** – – – 0.92*** �0.05 �0.46*** 2.24***D1978 1.78*** 1.36*** 1.82*** – – – 1.40*** 0.11 0.14 2.10***D1979 1.52*** 1.16*** 1.53*** – – – 1.18*** 0.23 �0.20 1.65***D1980 1.76*** 1.16*** 1.87*** – – – 1.23*** 0.11 �0.08 1.65***D1981 1.57*** 1.16*** 1.64*** – – – 1.15*** �0.14 �0.14 1.87***D1982 1.91*** 1.36*** 2.07*** – – – 1.36*** �0.06 0.03 1.88***D1983 1.51*** 1.05*** 1.53*** – – – 1.07*** �0.21 �0.21 1.80***D1984 1.62*** 1.32*** 1.61*** – – – 1.38*** 0.27 0.04 1.78***D1985 1.67*** 0.97*** 1.71*** – – – 1.00*** �0.17 �0.23 1.86***D1986 1.51*** 1.17*** 1.60*** – – – 1.17*** �0.18 �0.06 1.83***D1987 1.88*** 1.92*** 1.89*** – – – 1.94*** 0.67*** 0.81*** 2.82***D1988 1.44*** 1.38*** 1.46*** – – – 1.41*** 0.14 0.36** 2.25***D1989 1.49*** 1.21*** 1.57*** – – – 1.21*** �0.01 0.14 1.97***D1990 1.88*** 1.88*** 1.97*** – – – 1.87*** 0.40 0.88*** 3.05***D1991 2.03*** 1.69*** 2.06*** – – – 1.79*** 0.31 0.72*** 2.66***D1992 1.49*** 1.37*** 1.55*** – – – 1.39*** 0.06 0.34** 2.36***D1993 1.47*** 1.23*** 1.48*** – – – 1.26*** �0.04 0.33** 2.35***D1994 1.60*** 1.49*** 1.63*** – – – 1.53*** 0.15 0.59*** 2.70***D1995 1.57*** 1.57*** 1.63*** – – – 1.56*** 0.07 0.69*** 2.80***D1996 1.51*** 1.62*** 1.53*** – – – 1.66*** 0.04 0.60*** 2.69***D1997 1.87*** 2.00*** 1.90*** – – – 2.01*** 0.58* 1.16*** 3.18***D1998 2.15*** 2.12*** 2.15*** – – – 2.16*** 1.00*** 1.27*** 3.13***D1999 1.64*** 1.82*** 1.66*** – – – 1.84*** 0.35 1.07*** 2.98***D2000 1.63*** 1.48*** 1.67*** – – – 1.48*** �0.19 0.54*** 2.47***D2001 2.29*** 2.26*** 2.35*** – – – 2.28*** 1.44*** 1.39*** 3.08***D2002 2.09*** 2.15*** 2.09*** – – – 2.19*** 1.01*** 1.20*** 3.04***D2003 1.91*** 2.09*** 2.08*** – – – 2.08*** 0.92*** 1.15*** 3.09***D2004 2.35*** 1.99*** 2.41*** – – – 2.05*** 0.72** 1.18*** 3.15***D2005 2.16*** 2.00*** 2.38*** – – – 1.97*** 0.45 1.18*** 3.13***D2006 2.74*** 2.31*** 2.78*** – – – 2.37*** 0.98*** 1.70*** 3.68***


a *Significant at 10%; **significant at 5% level; ***significant at 1% level.

References

Adam, K., Jappelli, T., Menichini, A., Padula, M., Pagano, M., 2001. Analyse, Compare, and Apply Alternative Indicators andMonitoring Methodologies to Measure the Evolution of Capital Market Integration in the European Union EuropeanCommission Report.

Adjaouté, K., Danthine, J.P., 2003. European financial integration and equity returns: a theory-based assessment. In: Gaspar, V.,Hartmann, P., Sleijpen, O. (Eds.), The Transformation of the European Financial System. European Central Bank, Frankfurt,pp. 185–245.

Adjaouté, K., Danthine, J.P., 2004. Equity returns and integration: is Europe changing? Oxford Review of Economic Policy 20,555–570.

Baele, L., Ferrando, A., Hördahl, P., Krylova, E., Monnet, C., 2004. Measuring Financial Integration in the Euro Area. ECB. Occa-sional Paper 12.

Bekaert, G., Harvey, C., 2003. Emerging markets finance. Journal of Empirical Finance 10, 3–55.Bekaert, G., Hodrick, R., Zhang, X., 2009. International stock return comovements. Journal of Finance 64, 2591–2626.Bodart, V., Reding, P., 1999. Exchange rate regime, volatility and international correlations of bond and stock markets. Journal of

International Money and Finance 18, 133–151.Bordo, M., Helbling, T., 2004. Have national business cycles become more synchronized? In: Siebert, H. (Ed.), Macroeconomic

Policies in the World Economy. Springer-Verlag, Berlin, pp. 3–39.Brooks, R., Del Negro,M., 2002. International Stock Returns andMarket Integration: a Regional Perspective. IMF.Working Paper 202.Calderon, C., Chong, A., Stein, E., 2007. Trade intensity and business cycle synchronization: are developing countries any

different? Journal of International Economics 71, 2–21.Camacho, M., Perez-Quiros, G., Saiz, L., 2006. Are European business cycles close enough to be just one? Journal of Economic

Dynamics and Control 30, 1687–1706.Catao, L., Timmermann, A., 2009. Volatility regimes and global equity returns. In: Bollerslev, T., Russell, J., Watson, M. (Eds.),

Volatility and Time Series Econometrics: Essays in Honor of Robert Engle. Oxford University Press, Oxford.Cappiello, L., Hördahl, P., Kadareja, A., Manganelli, S., 2006. The Impact of the Euro on Financial Markets. ECB. Working Paper

598.


Chinn, M., Forbes, K., 2004. A decomposition of global linkages in financial markets over time. Review of Economics andStatistics 86, 705–722.

Flavin, T., 2004. The effect of the euro on country versus industry portfolio diversification. Journal of International Money andFinance 23, 1137–1158.

Frankel, J., Rose, A., 1998. The endogeneity of the optimum currency area criteria. The Economic Journal 108, 1009–1025.Griffin, J., Karolyi, A., 1998. Another look at the role of industrial structure of markets for international diversification strategies.

Journal of Financial Economics 50, 351–373.Heston, S., Rouwenhorst, G., 1994. Does industrial structure explain the benefits of international diversification? Journal of

Financial Economics 36, 3–27.Imbs, J., 2004. Trade, finance, specialization, and synchronization. Review of Economics and Statistics 86, 723–734.Imbs, J., 2006. The real effects of financial integration. Journal of International Economics 68, 296–324.Kalemli-Ozcan, S., Sorensen, B., Yosha, O., 2001. Economic integration, industrial specialization, and the asymmetry of

macroeconomic fluctuations. Journal of International Economics 55, 107–137.Kose, M., Prasad, E., Terrones, M., 2003. How does globalization affect the synchronization of business cycles? American

Economic Review 93, 57–62.Lane, P., Milesi-Ferretti, G.M., 2001. The external wealth of nations: measures of foreign assets and liabilities for industrial and

developing countries. Journal of International Economics 55, 263–294.Lane, P., Milesi-Ferretti, G.M., 2007. The external wealth of nations mark II: revised and extended estimates of foreign assets and

liabilities, 1970–2004. Journal of International Economics 73, 223–250.Lane, P., Milesi-Ferretti, G.M., 2008. International investment patterns. Review of Economics and Statistics 90, 538–549.Lane, P., Wälti, S., 2007. The euro and financial integration. In: Cobham, D. (Ed.), The Travails of the Eurozone: Economic Policies,

Economic Developments. Palgrave Macmillan, pp. 208–230.La Porta, R., Lopez-de-Silanes, F., Shleifer, A., Vishny, R., 1998. Law and finance. Journal of Political Economy 106, 1113–1155.Otto, G., Voss, G., Willard, L., 2001. Understanding OECD Output Correlations. Research Discussion Paper 05. Reserve Bank of

Australia.Portes, R., Rey, H., 2005. The determinants of cross-border equity flows. Journal of International Economics 65, 269–296.Rose, A., Engel, C., 2002. Currency unions and international integration. Journal of Money, Banking and Credit 34, 1067–1089.Rouwenhorst, K., 1999. European equity markets and EMU. Financial Analysts Journal 55 (3), 57–64.

stock market synchronization and monetary integration

Documents