Distribution specific dependence and causality between industry-level U.S. credit and stock markets

Abstract

This paper examines the dependence and causal nexuses between ten U.S. credit default swaps and their corresponding stock sectoral markets, using the Quantile-on-Quantile (QQ) approach and the nonparametric causality-in-quantiles tests. The results, using the QQ approach, show asymmetric negative association between credit and markets for all industries and that the link depends on both the sign and size of the stock market shocks (i.e., bullish or bearish conditions in the CDS and/or stock markets). The sensitivity of CDS returns to stock markets shocks is higher in the extreme quantiles. Using the nonparametric causality-in-quantile tests, we find evidence of causality-in-mean from stock to CDS only for the Financial (in average and upper quantiles), Consumer Services and Oil & Gas sectors (only for the middle quantile i.e., 0.5). In addition, the causality-in-mean from the CDS to stock markets is only found for the Financial and Telecommunication sectors in the extreme lower quantiles. Finally, we find a bidirectional Granger causality-in-variance for all the CDS-equity sector pairs.