The purpose of this paper is to determine the cross-market liquidity and price spillover effects across euro area sovereign bond markets. The analysis is carried out with the constructed minute frequency order-book dataset from 2011 until 2018. This derived dataset covers the six largest euro area markets for benchmark 10-year sovereign bonds. To estimate the cross-market spillover effect between sovereign bonds, it was decided to use the empirical approach proposed by Diebold and Yilmaz (2012) and combine it with the vector error correction model (VECM). We also employed the panel regression model to identify why some bond markets had a higher spillover effect while others were smaller. The dependent variable was the daily average spillover effect of a particular bond. As the spillover effects vary highly across different bonds, country-specific fixed effects were used, and the clustered standard errors were calculated for robustness reasons. Lastly, the cross-market spillovers were analyzed daily to compare them with the results of the model with intraday data. The analysis was performed with rolling 100-day window variance decompositions and a 10-day forecast horizon for six sovereign bonds and the overnight indexed swap (OIS) market. The results of the created time-series model revealed that intraday cross-market spillovers exist but are relatively weak, especially in the case of liquidity spillovers. As the cross-market linkages became much more robust with the model using daily data, the liquidity or price disbalances between different markets are usually corrected on longer intervals than minutes. Distance between countries is the most important explanatory variable and is negatively linked to the magnitude of both liquidity and price spillovers. These findings should be of particular interest to bond market investors, risk managers, and analysts who try to scrutinize the liquidity and price transmission mechanism of sovereign bonds in their portfolios.
Long-Term Returns in Distressed Sovereign Bond Markets
