ادغام بازارهای اوراق قرضه اروپا

I investigate the time variation in the integration of EU government bond markets. The integration is measured by the explanatory power of European factor portfolios for the individual bond markets for each year. The integration of the government bond markets is stronger for EMU than non-EMU members and stronger for old than new EU members. For EMU countries, the integration is weaker the lower the credit rating is. During the recent crisis periods, the integration is weaker, particularly for EMU countries.

I investigate the time variation in the integration of the European bond markets. I consider European Union (EU) countries, both EMU members (the euro countries) and non-EMU members. For the latter, I distinguish between old and new EU member states. The analysis begins in 1994 and covers the recent years of turmoil, first during the financial crisis and subsequently during the European sovereign debt crisis. I use the new approach to measure integration brought forward by Pukthuanthong and Roll (2009) and further applied by Berger et al., 2011 and Berger and Pukthuanthong, 2012. They measure the integration of stock markets by the explanatory power (the R -squared value) of certain factor models: For a given year for a given country, they regress the daily stock market return on the daily returns of a number factor portfolios. The factors are constructed from the principal components of all stock markets using the weights from the previous year. An advantage of the new measure of integration is that it is simple, both conceptually and calculation-wise. Moreover, the new approach conveniently provides a measure of integration for each country for each year. Thereby, the method provides a panel of integration values that allows me to study the differences in integration across countries and across time. The new method takes into account that markets can be integrated even if there are differences in their exposure to the factors. This paper is (to the best of my knowledge) the first to apply this methodology outside the stock market setting. The EU government bond markets lend themselves to a number of interesting testable hypotheses. One: Before the recent crises, the bond markets of the EMU countries are effectively one market that is strongly integrated and the bonds are almost perfect substitutes. Two: During the crises, the EMU bond markets become less integrated and act more as individual bond markets. This is due to the differences in the governments’ situation across countries where the differences have become stronger during the crises periods. Three: There are differences between the level of integration of the bond markets depending on the type of the EU membership. The integration is stronger for the EMU countries than for non-EMU countries due to common monetary policy for EMU countries. The integration is weaker for the new EU countries than for the old EU countries because they have had less time to adjust to being an EU member state, i.e. less time to implement EU policies and regulations. Four: the integration of the bond markets is related to the credit rating of a country, so that the stronger the credit rating is, the stronger the integration is because for countries with lower credit ratings country specific information is more important. The credit rating of a country is an important piece of country specific information. The credit rating is most important for EMU members because they are not able to conduct monetary policy, so the country specific information is particularly relevant here. It is not possibly to provide any quantitative implications of the stated hypotheses. I apply daily returns from indexes of 17 EU government bond markets. The empirical findings are overall in accordance with the presumptions outlined above. The Pukthuanthong and Roll (2009) methodology for investigating the time variation in the integration of financial markets is thereby shown to be applicable to the bond market setting as well. The current paper is related to previous research about the integration of European financial markets. Moerman (2008) finds that there are better diversification possibilities across industries than across countries for European stock markets. Mylonidis and Kollias (2010) find that the European stock markets become more integrating during the first decade after the introduction of the euro. Cappiello et al. (2010) use quantile regression analysis and find that the comovement between the European stock markets increase after the introduction of the euro. Bekaert et al. (2013) find that the integration for European stock markets is greater for EU member states than for non-EU member states. The integration of the EU stock markets is independent of their EMU membership. Pozzi and Wolswijk (2012) consider the integration of the government bond markets of five old EU countries. They find that the markets are fully integrated until the financial crisis after which they become less integrated. Abad et al. (2010) consider the integration of 15 old EU countries’ bond markets. They show that there are differences between the integration of EMU and non-EMU countries where the former are the most integrated with the German bond market. Ehrmann et al. (2011) consider the convergence of the EU government bond markets. They find that there is one common government bond market for EMU countries. The reason for the convergence is the adoption of the euro currency. Christiansen (2007) uses volatility spillover analysis to show that for the bond markets of the old EU countries, the EMU countries are more integrated than non-EMU countries and that bond markets become more integrated of the introduction of the euro. Cappiello et al. (2006) find some evidence of increased integration of the government bond markets of the new EU member states. Beber et al. (2009) show that the main reason for differences in the yield of euro government bonds is differences in credit quality. Still, at times of distress liquidity is also important. Gerlach et al. (2010) find that the size of the banking sector is an important determinant of the government bond yield differences across EMU countries. The current paper is also related to previous research about global term structure factors and global sovereign credit risk factors. Edwards (1984) investigates the determinants of interest rate risk premia for emerging markets. Generally, the higher the probability of default is, the higher the risk premium is but still some of the interest rate risk premia remains unexplained. Sutton (2000) finds that 10-year government bond yields for five countries are positively related across markets and that the time variation has an international component. Mauro et al. (2002) compare emerging market bond yields before WWI with those of the 1990s. They find that the comovement is much stronger in the most recent time period. They conjecture that investors pay less attention to country specific events in the most recent period. Jotikasthira et al. (2012) find that global macroeconomic factors influence bond yield across three countries. Remolona et al. (2008) analyze credit default spreads (CDS) and find that sovereign risk is driven mainly by country specific factors whereas risk premia are driven by global factors. Longstaff et al. (2011) also consider sovereign credit risk using CDS. They find that sovereign credit risk is driven more by global factors than by country specific factors. The default risk component is more strongly related to global factors than is the risk premium component. The remaining part of the paper is organized as follows. Section 2 explains how integration is measured. Section 3 describes the data. Section 4 contains the empirical results. Section 5 concludes.

In this paper I use the methodology of Pukthuanthong and Roll (2009) to analyze the integration of the EU government bond markets. The integration is measured by the explanatory power of certain factor portfolios for the bond market returns for the individual country for a given year. The main empirical findings are as follows: The EMU countries are more integrated than the non-EMU countries. The new EU countries are less integrated than the old EU countries. The lower the credit rating, the less integrated are the bond markets. However, this relationship for credit ratings is only valid for EMU members. For non-EMU members the credit rating is irrelevant for explaining the bond market integration. The integration is lower during the recent crisis period, and this in particular applies to EMU members.