Working Paper, Federal Reserve Bank of Atlanta 2002-23
We estimate a model with country- and industry-specific shocks that extends the dummy variable model used in the portfolio diversification literature by relaxing the restriction that all stocks with exposure to a given shock have the same exposure to that shock. We find that: i) This restriction is strongly rejected by the data. ii) Many industry betas are negative, while almost all country betas are positive. This difference in within-group heterogeneity may explain why country shocks have historically outweighed industry shocks in explaining international return variation. iii) We use the betas to construct portfolios whose volatility is substantially below that of the world market, both in and out of sample.