Industry and Country Effects in International Stock Returns

that international returns are predominantly driven by industry factors. Managers who believe that domestic market factors are more important for returns than industry factors decide on a country allocation first, then in the second stage select the most promising stocks from each country. This article presents a simple model to measure country and industry effects in international stock returns, and provides a quantitative framework for analyzing these two approaches to portfolio selection.' We show that there are three reasons for portfolio managers to pay more attention to the geographical than to the industrial composition of an international portfolio. Each of these reasons is based on the finding that country effects in international stock returns are larger than industry effects. First, tilting an international portfolio geographically leads, on average, to larger and more variable tracking errors than tilting the industrial composition of the portfolio. Second, stocks from the same domestic market but in different industries are closer substitutes than stocks from the same industry but in different countries. Finally, the benefits of international I