Global integration and emerging stock market excess returns

Abstract

This article studies the effects of the global integration process on emerging stock market excess returns in a dynamic context. I improve the existing literature in four main directions. First, I show that the average excess returns rise as the level of financial and real integration rises. Second, I find overwhelming evidence that the financial liberalizations (i.e. de jure integration) of the late 1980s and early 1990s have not been simultaneously accompanied by a de facto integration. Third, I find that the percentage of variation in emerging excess returns explained by non-traded global risk factors rises as the level of market openness rises. Last, at the country level, I show that the correlation coefficient does not represent a robust measure of integration. Results also suggest that there are substantial cross-country differences in the dynamics of the degree of financial integration.

Keywords:

• emerging stock markets
• market openness
• financial integration

Acknowledgements

I am indebted to Alessia Varani and Paolo Vitale. I thank an anonymous referee, Nicola Borri, Lauren Persha and Lorenzo Prosperi for their suggestions. I also gratefully acknowledge comments from seminar participants at LUISS Guido Carli, UNC at Chapel Hill, Toulouse School of Economics, International Conference in Applied Business and Economics at University of Piraeus (Athens) and 1st PhD Student Conference in International Macroeconomics and Financial Econometrics at Université Parìs Ouest. All errors are the author's responsibility.

Notes

1. For a detailed discussion on emerging crises, see Joyce (2011).

2. For example, in Panchenko and Wu (2009), the sample period goes from January 1995 to December 2005, and in De Jong and De Roon (2005) it runs from January 1988 (or later) to May 2000. Pukthuanthong and Roll (2009) employ data running from January 1973 (or later) to February 2008. In contrast, my sample period goes from January 1988 (or later) to December 2011.

3. For a detailed discussion on the de jure and the de facto integration, see Bekaert, Harvey, Lundblad and Siegel (2009).

4. A similar argument can be found in Carrieri, Errunza and Hogan (2007) and Kritzam et al. (2011).

5. I justify the choice of monthly data as follows: (i) the choice is based on a desire to avoid missing daily data in the time series as well as the time zone issue; (ii) the monthly frequency is also motivated by the need to match macroeconomic variables' frequency.

6. Similar results can be found in Bekaert et al. (1998), Estrada (2000) and Grootveld and Salomons (2003), among many others.

7. The IMF database does not provide the Czech Republic Gross Domestic Product series for the years 1990–1994. To complete the series, I borrow the Czech Republic Gross Domestic Product from the OECD data library. OECD GDP is denominated in US$ at current prices and at current exchange rates.

8. Gross domestic product, current prices (US$): values are based upon GDP in national currency converted to US$ using market exchange rates (yearly averages). Source: IMF.

9. For the UKthe last observation corresponds to June 2011. Therefore, from July 2011 to December 2011 the UK trade share is assumed to be constant and equal to the last available observation.

10. Gross domestic product – expenditure approach (millions of US$, current prices, current PPPs, annual levels, seasonally adjusted). The OECD Total covers the following 34 OECD Member countries: Australia, Austria, Belgium, Canada, Chile, Czech Republic, Denmark, Estonia, Finland, France, Germany, Greece, Hungary, Iceland, Ireland, Israel, Italy, Japan, Korea, Luxembourg, Mexico, Netherlands, New Zealand, Norway, Poland, Portugal, Slovak Republic, Slovenia, Spain, Sweden, Switzerland, Turkey, United Kingdom and United States. Source: OECD.

11. All time series are I(0). The ADF test for [(R_m,t – R_t ^f ) · (GlobIntIndex _t‒1)] rejects the null hypothesis of a unit root (i.e. the series is stationary). The test is available upon request.

12. For a detailed discussion on stock markets liberalizations, see Bekaert and Harvey (1995, 1997, 2000).

13. Similar results can be found in De Jong and De Roon (2005) and Donadelli and Prosperi (2012).

14. Pukthuanthong and Roll (2009) estimate the covariance matrix for each calendar year using daily data from 1973 to 2006. Once eigenvectors are computed and sorted from the largest to the smallest eigenvalue, they estimate principal components from return in the subsequent calendar year (e.g. the weightings (eigenvectors) computed from the 1973 covariance matrix are applied to the returns during 1974).

15. Note that all national stock indexes are total return indexes (i.e. I use only MSCI TRI to compute returns). Donadelli and Prosperi (2012) point out that price indexes (i.e. MSCI PI) might generate different returns.

16. The limited amount of data does not allow us to make robust comments on the following African stock markets: Kenya, Nigeria, Morocco and Tunisia.

17. Using industrial stock market indices, Donadelli (2013) obtains a similar result.

18. A similar set of variables reflecting global economic conditions can be found in Ciarlone, Piselli and Trebeschi (2007).