Change in Cash-Holding Policies and Stock Return Predictability in the Cross Section

The author found that stocks with a positive change in company cash holdings have significantly higher risk-adjusted returns than stocks with a negative change in cash holdings (CCH). Moreover, the return predictive power of CCH is (1) distinct from the effect of cash holdings (CH), (2) absent among cash-rich companies, (3) stronger among small-cap stocks, and (4) limited to non-January months. The CCH anomaly appears to be more “contaminated” than the CH effect by mispricing.A considerable amount of research has been published on the determinants of a company’s cash holdings and the holdings’ time-series properties over time. Only recently, however, have excess cash holdings been linked to stock returns and a precautionary savings policy to expected returns. This study investigates the cross-sectional relation between a company’s changes in cash holdings (CCH) and subsequent equity returns. Changes in cash holdings are, by definition, different from the level of cash holdings or excess cash holdings in the sense that a company can increase its cash holdings and still have a low cash-holding level or a company can decrease its cash holdings and still maintain an excess cash-holding level. I found evidence that CCH, as a signal for changes in future investment opportunities, predicts subsequent stock returns. The abnormal returns associated with CCH remain significant in the presence of size, book-to-market, momentum, asset growth, illiquidity, and idiosyncratic volatility. Specifically, stocks with positive CCH significantly outperform stocks with negative CCH. The dispersion in risk premiums is not explained by the capital asset pricing model, the Fama–French three-factor model, or the Carhart four-factor model. Importantly, the Sharpe ratio of a CCH-based factor-mimicking portfolio is higher than that of the market factor or the Fama–French SMB and HML factors and is very close to that of the momentum factor. Moreover, the return predictive power of CCH is (1) distinct from the effect of the level of cash holdings (CH)—that is, the abnormal returns associated with CCH remain after controlling for CH, and vice versa—(2) absent among cash-rich companies, (3) stronger among small-cap stocks, and (4) limited to non-January months. I explored the implications of arbitrage risk for both the CCH and the CH effects and found that unlike accrual and asset growth anomalies, both effects are pervasive across arbitrage risk groups (proxied by idiosyncratic risk) but that the CCH anomaly appears to be more influenced than the CH effect by mispricing. Finally, I found a surprisingly negative CCH effect for January months, which is, however, driven entirely by mispricing.