Determinants of Stock Market Correlation. Accounting for Model Uncertainty and Reverse Causality in a Large Panel Setting

Abstract

We examine 22 determinants of stock market correlations in a panel setting with 651 country pairs of developed economies over the 2001-2018 period, while accounting for model uncertainty and reverse causality. On the one hand, we find, that a number of determinants, well established in the literature, e.g. trade, institutional distance, and exchange rate volatility fail the robustness test. On the other hand, we find strong evidence supporting several others: (1) inertia, with current correlation being the best single predictor of the future stock market correlation (2) positive impact of the market size (3) imperative role of the interconnected financial factors: capital mobility, financial development, and portfolio equity flows. With the expected future growth of economies and their capital markets as well as deepening financial liberalization, this paper brings strong support to the hypothesis of diminishing international diversification potential.  

Keywords: stock market correlations, stock market comovement, financial development, Bayesian model averaging, OECD countries

JEL Classification: G100, G110, G150, F620