ACTIVE PORTFOLIO MANAGEMENT AND THE GAINS FROM INTERNATIONAL PORTFOLIO DIVERSIFICATION

The thesis estimates the benefits of international portfolio diversification using an 'active' portfolio management policy. Such a policy was desirable on two counts. First, the expected returns and/or the variance-covariances of international asset returns were found to be non-stationary. In active portfolio management, it is possible to 'internalize' the non-stationarity of the stochastic process generating the data. Second, the international capital market was found to be inefficient by some earlier studies. In such a market, there exists super risk-premium which may be obtained through active management strategy. Using ex ante ' data, the additionaL benefits of ' international' over 'domestic' diversification was demonstrated by comparing the risk-return characteristics of some ' international' portfolios with those of a domestic benchmark portfolio. The portfolio selection model used throughout was the single-period Markowitz Mean-Variance model. Several portfolio strategies were investigated and it was seen that for a representative U.S. investor there were significant additional gains to be enjoyed from holding internationally diversified portfolio rather than t'he domestic portfolio. The average returns of the actively managed portfolios over the period were then compared to those of some 'passive' buy-and-hold portfolios after allowing for some arbitrarily selected transaction costs. It was seen that the actively managed portfolios outperformed the buy-and-hold portfolio for almost all the holding period cases. This result supports the 'partial' segmentation view of the international capital market and its consequent inefficiency. Incorporating Pratt-Arrow measure of relative risk-aversion in the expected utility function, the effect of investor's risk tolerance on portfolio risk and return was explored. It was seen that between two classes of risk-averse investors, the more risk-averse class earned less risk-adjusted return on average than their less risk-averse counterparts. Finally, the effect of the exchange factor on portfolio risk and return was analysed by decomposing portfolio return and variance into constituent parts. It was seen that under the fixed exchange rates system, the exchange rate changes led to a reduction of portfolio variance for some periods while under the flexible rates system they raised it. Their effects on portfolio return were mixed; however, on average, over a large number of periods, they contributed positively. Also, the exchange factor was found to affect portfolio choice -a result at variance with the view held by the proponents of the PPP theory of exchange rate determination.