Brazilian equity risk premium analysis: a macroeconomic approach

Abstract

This research studies the role of fluctuations in the aggregate consumption-wealth ratio 𝑐𝑎𝑦 proposed by Lettau and Ludvigson (2001) as a predictor of stock returns in the Brazilian economy. Using quarterly data, evidence for predictability of asset growth was found with an 𝑅̅2 of over 45% and a highly significant coefficient as expected, in contrast to absence of statistical evidence for predictability of stock returns or excess returns. Regressions containing those fluctuations also resulted in worse 𝑅̅2. For the data used, dividend yield was not capable of showing predictive power also. The predictability of the returns on the Brazilian economy is not rejected but data fails to show the expected results. Finding macroeconomic data that represent the same agent was a big obstacle. After testing many different datasets and different model specifications, data still failed to show any explanatory power over returns or excess returns.