The Cross–section of Risk Premia in the Italian Stock Market

Abstract

Our analysis compares multi–factor models with Italian stock market data for the period 1990–2000. The first, the simple CAPM, is the relevant benchmark because of its simplicity. The second, the extended Fama–French model (including the momentum portfolio), is the best candidate for substituting the one–factor model. The third is a multi–factor model including sectors; and the fourth is a multi–factor model including the change in short–term interest rates as an extra factor. The results of our research are mildly positive. The Fama–French multi–factor model behaves rather well in time series tests. However, in the cross–section, the average premia are not significantly different from zero, supporting the idea that they are not able to explain the cross–section of returns in the Italian market. Moreover, there is weak evidence that the factor portfolios have predictive power for macroeconomic variables characterizing the state of the economy. What may explain the results? There are two main components: first is the presumably large size of shocks to returns in the Italian case, which makes it difficult to explore the relation among expected returns; second is the presence of extra factors which are not accounted for in our analysis.(J.E.L.: G11, G12).