Consumption risk exposures based on firm characteristics in the Korean stock market

Abstract

The main goal of this thesis is estimating consumption risk exposures at the firm-level. To avoid existing problems in estimating consumption betas at the firm-level such as measurement errors in low-frequency consumption data or covariance structures, I map out the relation between consumption betas and a set of firm characteristics using panel regressions. The decile portfolios sorted by consumption risk exposures calculated by firm characteristics yield monotonically increasing average returns which are statistically significant and robust to risk adjustment. Moreover, I show that consumption betas implied by firm characteristics capture the cross-sectional variation in expected returns of industry portfolios. Finally, I examine Fama-MacBeth regressions and the results indicate that risk exposures implied by characteristics dominate other alternative factor models at explaining cross-sectional variation in mean returns.