Stock market volatility, business cycles and the recent financial crisis: evidence from linear and non-linear causality tests

Abstract

The relationship between stock market volatility and the business cycle is macrofinancial as it links the fields of financial markets and macro-economics. This relationship links to theories of rational expectations/efficient market hypotheses and asset pricing theory. This thesis investigates the long run relationship between stock market volatility and business cycles by means of linear and non linear bivariate and multivariate causality tests. Moreover, it investigates the impact of the recent global financial crisis on the stock market volatility (SMV) and business cycles (BC) relationship. The contributions of this research to the literature include: a) analyzing the non-linear causal relationship between stock market volatility and the business cycle; b) exploring the cross-country causality between these variables; and c) looking at the impact of the financial crisis on the said relationship. To the best of our knowledge this is the first time that any of these three aspects of the relationship between stock market volatility and the business cycle have been studied. The countries investigated are the US, the UK, Canada, and Japan (among the developed countries) and Brazil, Malaysia and Turkey (from the developing countries). Monthly data from January 1990 to December 2011 are applied in the empirical investigation. Stock market volatilities are estimated using the GARCH model, and industrial production indices are used for the business cycles. Bivariate non-linear causality tests are conducted by means of the Diks and Panchenko (2006) and Hiemstra and Jones (1994) methods. Non-linear multivariate tests are conducted by means of the Bai et al. (2010) method. Multivariate tests investigate the cross-country spill-over between two countries, with the US as the main country. The results indicate that a non-linear causal relationship does exist between stock market volatility (SMV) and business cycles (BC). There is strong evidence to suggest that the SMV-BC relationship is not limited to within country only, as we find significant cross-country causal relationships. Both linear and non-linear bivariate causality tests indicate evidence of a stronger causality between variables when the financial crisis is taken into consideration. Also, both the linear and non-linear multivariate tests indicate that the US has a greater impact on the SMV and BC of developed countries than developing countries. And this impact has further increased during the recent financial crisis. The findings from this research have implications not only for investors and portfolio managers, but also for economists and policy makers. In addition, the research results signal that in countries such as the UK, inclusion of the US stock market as a business cycle indicator, in addition to the UK’s own stock market, may be beneficial in identifying the turning points of the UK’s business cycle, and vice versa.

More information

Identifiers

ORCID for Taufiq Choudhry: orcid.org/0000-0002-0463-0662

Catalogue record

Export record