Economic concepts, methods, and variable classifications of personal consumption expenditures (PCE) are regularly revised to accurately reflect evolving economy; however, history is rewritten when these updates are adopted retrospectively to measure past economy. Consequently, latest-vintage PCE is significantly less informative about future stock market returns than real-time-vintage PCE. Moreover, in revised data only the fourth quarter consumption growth predicts market returns. The seasonality, which reflects revisions of seasonal adjustment factors, disappears when using real-time data. These findings are consistent with the countercyclical equity premium hypothesis and lend support to using real-time data in empirical studies.
主 講 人:Dr. Yu-Jou (Abby) Pai