پویایی حرکت شرکت بازار سهام اروپای مرکزی

This paper examines short-term and long-term comovements between developed European Union (EU) stock markets and those of three Central European (CE) countries which recently joined the EU. Dynamic cointegration and principal components methods are applied, in addition to static tests. While we find no evidence of cointegration for the period July 1995–February 2005 as a whole, dynamic tests reveal alternating period of cointegration disrupted by episodes dominated by short-term domestic factors. Principal components analysis reveals that a stable factor explains a large proportion of return variances. Ultimately, despite the decade-long process of alignment by CE countries with the EU, evidence of steadily increasing convergence of equity markets is lacking.

Investors in developed countries have been diversifying into emerging equity markets for some time, in search of potential benefits from international portfolio diversification. However, currency, banking, and other economic crises experienced in the last decade in a number of Asian and Latin American countries have led investors to branch out into other areas, including the markets of Central Europe (CE). A history of moderate correlations and relatively high returns, particularly in countries such as the Czech Republic, Hungary, and Poland, suggest that diversification benefits in these equity markets can be substantial. Further, the process of accession to the European Union (EU) has increased prospects of growing economic and political stability in this area. However, the increasing association of the economies of the CE countries with those of developed EU members raises the question of whether closer linkages between these equity markets have resulted in significantly reduced diversification benefits for international investors. Equity market comovements can arise from international trade, increasing capital mobility, relaxation of controls on international capital movements, as well as the various forms of policy alignment associated with the creation of economic unions. Since 1989 the economies of the Czech Republic, Hungary, and Poland have undergone a rapid and largely successful transformation from Communist to market economies. Also, trade ties with EU countries have increased during this period. By 2004 the EU accounted for approximately 65% of Czech exports and 73% of imports, 79% and 72%, respectively for Hungary, with 67% and 74% for Poland. Roughly 40% of these figures are accounted for by Germany in recent years (International Monetary Fund, 2000 and International Monetary Fund, 2005). Capital flows to the region had begun by 1993, with several equity funds for the region created by 1996 (Sobol, 1996). In 2002–2003, during the run-up to EU accession, increased investments were made by American and other foreign emerging market funds, and for 2004 net private portfolio inflows for the European emerging markets as a group were estimated to be $7.6 billion (Economic Commission for Europe, 2005 and Institute of International Finance, 2006; New York Times, 2005). Along with other aspiring new members, the Czech Republic, Hungary, and Poland have for some years pursued a deliberate policy of aligning their economic systems with the EU. During 1994–1995 they became associate members, which involved creation of a free-trade area, cooperation in the areas of industry, environmental protection and transport, and alignment of some national legislation. Accession negotiations began in 1998 and were concluded in December 2002 at the Copenhagen European Council meeting. On May 1, 2004, these three countries, along with seven others, became full members of the EU. Each country also “participates” in the European Monetary Union (EMU) from the date of accession “as a Member State with a derogation” (European Central Bank, 2005). The euro is to be adopted by each country in the future, as the necessary conditions are fulfilled. Short-term linkages between Central European equity markets and those of developed countries during the late 1990s have been examined in several studies. Gelos and Sahay (2000) found only weak evidence of changing linkages as a consequence of the 1997 Asian crisis but did establish higher correlations and volatility spillovers after the 1998 Russian crisis. Comparing the periods 1994–1996 and 1996–1998, Chelley-Steeley (2005) found increasing correlations between the CE equity markets and those of the UK, Germany, and other developed countries. In addition, a variance decomposition methodology showed that nearly 40% of variation in equity market returns for Hungary and Poland were due to non-domestic factors in the latter period, as opposed to about 10% for 1994–1996; there was little difference for the Czech equity market across the two periods. Cointegration studies involving the Central European equity markets have yielded differing findings that may in part be sample dependent. Early studies largely support the idea of significant long-run comovements involving the CE markets in the mid 1990s, while also yielding some evidence that those linkages may have weakened under the impact of the Asian and Russian crises in 1997–1998. Linne (1998) finds some evidence of cointegration involving the major CE markets and developed markets from the early 1990s through 1997 but concludes that the markets in the transition countries were mainly driven by domestic factors. Examining a similar time period but with a different methodology, Scheicher (2001) notes some degree of integration between the Hungarian and the Polish markets during 1995–1997. Jochum, Kirchgässner, and Platek (1999) similarly detect evidence of cointegration during the 1995–1997 period. However, after late 1997, as the Asian crisis spread and the Russian equity market began to weaken, an increase in volatilities in the CE markets caused short-run dynamics to overwhelm the earlier common trending by 1998. Röckinger and Urga (2001) also investigate the relative importance of the German and UK equity markets for the CE markets from 1994 to 1997 using an extension of the methodology of Bekaert and Harvey (1997). They conclude that the influence of London was stronger for the Czech and Polish markets than that of Frankfurt. The Hungarian market was little affected by either. Additionally, MacDonald (2001) finds evidence of long-term relationships between the CE equity markets as a group and three developed markets, Germany, the UK, and the US, individually. Gilmore and McManus (2003), however, report a lack of bilateral and multilateral cointegration of the equity markets of the Czech Republic, Hungary, and Poland with the German market for the years 1995 through early 2002. These static studies, which yield somewhat different conclusions, assume stability in the long-run relationships. However, this assumption may not be warranted. Instead, linkages between equity markets may be time-varying and episodic. One method which has been used to adjust for possible instabilities is that devised by Gregory and Hansen (1996). Their method detects structural breaks that can reveal evidence of long-run relationships that is not identified by static cointegration tests. Voronkova (2004) applies their methodology to test for structural changes over the 1993–2002 time period in the relationships between the three major CE equity markets and those of France, Germany, the UK, and the US. Using local-currency indices from the various national equity markets, she obtains stronger evidence of cointegration within the CE markets and between them and the more developed markets. The present study contributes to the understanding of short-run and long-run financial market comovements by employing several static and dynamic methods of analysis and offers additional insights into the time-varying nature of equity market linkages. It also provides evidence as to whether comovements can be viewed as a gradual, unidirectional progression resulting from increasing ties between economies over time or as a more sporadic phenomenon which may at times be overwhelmed by domestic forces. We apply various static and dynamic methodologies to examine the comovements of the major CE equity markets with those of London and Frankfurt, the dominant developed equity markets in the EU. For the period 1995–2005 as a whole, a static cointegration analysis does not reveal any evidence of a long-run relationship between the CE equity markets as a group and those of either Frankfurt or London. However, dynamic tests reveal periods of cointegration as well as instances where short-run behavior overpowers the long-run equilibrium relationship. This instability in comovements over time between the CE and more developed EU markets has implications for international investors. The rest of this paper is organized as follows. Section 2 presents characteristics of the three CE equity markets. The data are described in Section 3. Section 4 briefly explains the various static and dynamic methodologies being used and presents the empirical results. Finally, Section 5 contains the concluding remarks.

In this paper we examined the comovements between CE and developed EU markets using both static and dynamic methods of analysis. The static analyses for the period July 1995 to February 2005 as a whole indicate relatively low levels of short-term correlations as well as a lack of statistically significant cointegration. Dynamic analyses using a rolling-window approach yield a more complex portrait of unstable short-run correlations and intermittent long-run cointegration. A dynamic principal component analysis shows nevertheless a single stable factor capable of explaining a large proportion of the behavior of the five markets as a group. Overall, there is little evidence of a fairly steady progress toward a long-term equilibrium relationship of the CE equity markets with those of the UK and Germany. This is somewhat surprising, considering the strong trade links with EU countries, Germany in particular, and the decade-long process of alignment of CE economies, political and financial institutions with the EU. Rather, the changing interactions of these markets point to intermittent periods of (statistically significant) cointegration with episodes during which short-run dynamics of the various markets, such as different economic growth rates or periods of instability, emerge to dominate the long-run relationship. This leads us to conclude that the process of EU accession has not changed the comovements between the CE equity markets in ways that would materially reduce diversification benefits for dollar-based investors.