International portfolio diversification: an ICAPM approach with currency risk

Abstract

This article investigates international stock market integration in four major developed economies, namely the United States, the Economic and Monetary Union of the European Union, Japan and the United Kingdom, and two Asian emerging, countries namely China and India, over the period from June 1994 to June 2009. To model stock market integration we estimate a dynamic version of the international capital asset pricing model (CAPM) in the absence of purchasing power parity. Conditional variance is modelled via a multivariate GARCH specification. To investigate the evolution of integration overtime we estimate the CAPM in sub-periods. In addition, we connect our results to the timing of world financial crises. Our findings show that the stock markets tend to move in parallel after June of 2002, although from 2002 to 2006 there have not been crises events. These results support the increasing globalization and interdependence of both emerging and developed markets in the recent decade, reducing the benefits of portfolio diversification.

Keywords:

• international markets
• market integration
• financial crises
• MGARCH-M specification

JEL Classification:

• F36
• C52
• G15

Acknowledgements

We wish to thank an anonymous referee of this journal for his/her useful comments, which improved the paper. The usual disclaimer applies.

Notes

1. According to the CIA World Factbook, on 1 January 2011, the EMU, the United States, Japan and the United Kingdom are in first four positions among developed countries when compared on nominal GDP and GDP (PPP). Also, from the above source, China was second after the United States and India fourth after the United States, China and Japan compared on GDP (PPP). Compared on nominal GDP, China is second after the United States and India eleventh after seven developed countries and China, Brazil and Russia.

2. For early studies documenting the benefits of international diversification, see Solnik (1974) for developed markets and Errunza (1977) for emerging markets. For more recent evidence, see for example Erb, Harvey and Viskanta (1994), De Santis and Gerard (1997) and Bekaert and Harvey (2000).

3. In a diagonal system with assets, the number of unknown parameters in the conditional variance equation is reduced from under full BEKK specification to under the diagonal BEKK specification.

4. We also estimated different specifications of the model. However, the results were qualitatively equivalent. The results are available upon request.

5. As suggested by Bekaert and Harvey (1995), calculating the returns in US dollars eliminates the local inflation.

6. According to Burns' (2002) study, he found evidence that the test starts to deteriorate as the number of lags exceeds 5% of the length of the series; however the number of lags should not exceed 15% of the length of data. Under these results, the percentage test is valid. In our case the 571 lags are approximately 14% of the length of our data.

7. Engle and Ng (1993) asymmetric tests include the sign bias, the negative size bias, and the positive size bias tests. The sign bias test examines the impact of positive and negative innovations on volatility not predicted by the model. The squared standardized residuals are regressed against a constant and a dummy that takes the value of unity if is negative, and zero otherwise. The test is based on the t-statistic for . The negative (positive) size bias test examines how well the model captures the impact of large and small negative (positive) innovations, and it is based on the regression of the squared standardized residuals against a constant and . The computed t-statistic for is used in this test.

8. For the process in to be covariance stationary, the condition has to be satisfied (see, for example, Bollerslev 1986; De Santis and Gerard 1997, 1998).