Predicting Aggregate Stock Returns with Aggregate Accrual Components

• In Event: Capital Markets Research II

Abstract

I investigate whether the insignificant return forecasting power of aggregate normal accruals documented in Kang et al. (2009) is driven by possible misspecifications of accrual decomposition models or is due to different return forecasting powers of various components in normal accruals. I find that aggregate normal accruals related to growth (investment) component negatively (positively) predict one-year-ahead aggregate stock returns. The return forecasting powers of aggregate accruals related to growth and investment components are robust to choices of accrual decomposition models used to estimate the growth and investment components in normal accruals. I also find that GDP growth and excess market returns are negatively correlated, and growth (investment) component in the aggregate normal accruals positively (negatively) predicts one- year-ahead GDP growth. My findings suggest that growth and investment components in aggregate normal accruals predict the aggregate stock returns via their ability to signal the economy conditions, providing possible fundamental links between aggregate normal accruals and one-year-ahead aggregate stock returns.

Author

• Liya Hou, University of Texas at Dallas