Time-Varying Risk Premia in Large International Equity Markets

Abstract

We use an estimation methodology tailored for large unbalanced panels of individual stock returns to address key economic questions about the factor structure, pricing performance of factor models, and time-variations in factor risk premia in international equity markets. We estimate factor models with time-varying factor exposures and risk premia at the individual stock level using 62,320 stocks in 46 countries over the 1985-2018 period. We consider market, size, value, momentum, profitability, and investment factors aggregated at the country, regional, and world level. We find that adding an excess country market factor to world or regional factors is sufficient to capture the factor structure for both developed and emerging markets. We do not reject asset pricing restriction tests for multifactor models in 74% to 91% of countries. Value and momentum premia show more variability over time and across countries than profitability and investment premia. The excess country market premium is statistically significant in many developed and emerging markets but economically larger in emerging markets.