Version numérique PDF

This paper analyses the cross sectional differences in individual stock?s exposure to a world sovereign credit risk factor. While sovereign credit risk has been shown to spill over to equity markets at the aggregate level, we examine the sovereign credit risk impact on individual stock returns. In particular, we estimate the beta of a stock?s return with respect to a world sovereign credit risk factor using Bayesian estimation techniques. Our sovereign credit risk factor is constructed from market-based information, notably from spreads of sovereign credit default swaps. We then examine the out-of-sample properties of portfolios of stocks sorted on their sovereign betas. We find that low sovereign beta equity portfolios have significantly higher returns than high sovereign beta portfolios in times of high sovereign risk and vice versa, which confirms that our measurement of sovereign risk exposure is robust out of sample. In addition, returns of equity portfolios during periods with high sovereign risk decrease almost monotonically with their estimated sovereign credit risk exposure, i.e. stocks with a higher estimated exposure indeed suffer higher losses from bad news on sovereign risk. Our approach to estimating sovereign risk exposure of stocks thus makes it possible to construct global equity portfolios with limited exposure to sovereign credit risk.JEL Classification: G11Keywords: sovereign risk, bayesian beta, stock returns.

Auteurs :Goltz Felix
Extrait de la revue BMI 124