This paper proposes a conditional measure of capital market integration that allows us to characterize both the cross-section and time-series of expected returns in developed and emerging markets. The measure, which arises from a conditional regime-switching model, allows us to describe expected returns in countries that are segmented from world capital markets in one part of the sample and become integrated later in the sample. Our results suggest that a number of emerging markets exhibit time-varying integration. Interestingly, some markets appear to be more integrated than one might expect based on prior knowledge of investment restrictions. Other markets appear segmented even though foreigners have relatively free access to their capital markets. The paper presents a model that allows for the degree of market integration to change through time. While this method can be applied to a general multifactor model, the intuition can be readily obtained in a one-factor setting. We allow conditionally expected returns in any country to be affected by their covariance with a world benchmark portfolio and by the variance of the country return. In a perfectly integrated market, only the covariance counts. In segmented markets, the variance is the relevant measure of country risk. Our integration measure is a time-varying weight which is applied to these two moments. The model allows for differing prices of variance risk across countries which depends on country-specific information and a world price of covariance risk which depends on global information. The model is conditional in the sense that predetermined information is allowed to affect the expected returns, covariances, variances and the integration measure. Our procedure allows us to recover fitted values for the integration measure so that the degree and trend of a particular market's integration can be depicted through time. The paper analyzes the results of the model applied to 33 national equity markets. The results show that the degree of market integration varies over time and that some markets are more integrated than others. The paper also finds that the integration of markets is influenced by economic and financial market policies followed by their governments or other regulatory institutions. The paper concludes that the model provides a useful tool for understanding the time-varying nature of market integration.This paper proposes a conditional measure of capital market integration that allows us to characterize both the cross-section and time-series of expected returns in developed and emerging markets. The measure, which arises from a conditional regime-switching model, allows us to describe expected returns in countries that are segmented from world capital markets in one part of the sample and become integrated later in the sample. Our results suggest that a number of emerging markets exhibit time-varying integration. Interestingly, some markets appear to be more integrated than one might expect based on prior knowledge of investment restrictions. Other markets appear segmented even though foreigners have relatively free access to their capital markets. The paper presents a model that allows for the degree of market integration to change through time. While this method can be applied to a general multifactor model, the intuition can be readily obtained in a one-factor setting. We allow conditionally expected returns in any country to be affected by their covariance with a world benchmark portfolio and by the variance of the country return. In a perfectly integrated market, only the covariance counts. In segmented markets, the variance is the relevant measure of country risk. Our integration measure is a time-varying weight which is applied to these two moments. The model allows for differing prices of variance risk across countries which depends on country-specific information and a world price of covariance risk which depends on global information. The model is conditional in the sense that predetermined information is allowed to affect the expected returns, covariances, variances and the integration measure. Our procedure allows us to recover fitted values for the integration measure so that the degree and trend of a particular market's integration can be depicted through time. The paper analyzes the results of the model applied to 33 national equity markets. The results show that the degree of market integration varies over time and that some markets are more integrated than others. The paper also finds that the integration of markets is influenced by economic and financial market policies followed by their governments or other regulatory institutions. The paper concludes that the model provides a useful tool for understanding the time-varying nature of market integration.