with Nikolay Gospodinov, 2021, Journal of Finance, Replications and Comments. [New Version] [New Code and Data] [Previous Version] [Previous Online Appendix] [Previous Code and Data].
Using long-horizon beta estimates, Lettau, Ludvigson, and Ma (2019) document striking pricing ability and cross-sectional explanatory power for a capital share growth factor across major asset classes. We revisit their findings and show that the statistical significance of their results is likely due to the interaction between the lack of identification of the proposed single-factor model and the persistence induced by overlapping the data to obtain long-horizon beta estimates. This casts doubts on whether capital share betas are truly priced in the cross-section of expected returns and calls for alternative methods to validate the existence of capital share risk in U.S. asset prices.