Mastering Investment 2001 / Part Three

The logic that lies behind overseas diversication
By Robert Hodrick
Published: May 25 2001 11:41GMT | Last Updated: May 31 2001 16:29GMT

Robert J. Hodrick is Nomura Professor of International Finance at Columbia Business School and Columbia University's School of International and Public Affairs. He is also a research associate of the National Bureau of Economic Research.

How internationally diversified should a portfolio be? Recommendations vary widely, from 5 per cent to 50 per cent. This article reviews the logic behind diversification, starting with the use of probability distributions to assess returns.

Using data from 12 country portfolios between 1970 and 2000, the trade-off between risk and return is analysed. Results demonstrate the benefits of diversification. A US investor would improve the prospects for a portfolio by adding equity from 10 of the 11 countries analysed.

The analysis continues by examining how to generate a portfolio that has an optimal risk-return trade-off. The global investor would have an increase of almost five per cent in expected return.

The weakness of the analysis is reliance on historical data, which might misrepresent the case for diversification by:

• errors in estimating returns;

• some countries' past returns being poorer representations of what to expect than others;

• not constraining the portfolio positions to be positive, which many investors do;

• not allowing for evidence that correlations of country returns are increasing.

Another way to diversify internationally is to choose companies based on a measurable criterion and not just by country, for example by the ratio of a company's book value to its equity market value.