This study investigates REIT's momentum returns in different market states, and explains the momentum phenomenon with a risk-based dividend growth theory of Johnson (2002). Results show that momentum returns on REITs are higher during up markets, consistent with Cooper, Gutierrez, and Hameed (forthcoming). Moreover, this study finds that winners' dividend/price ratios are higher than those of losers, and that conditioning on different market states, momentum returns are positively correlated with the difference between winners' and losers' dividend/price ratios, consistent with Johnson's prediction that momentum portfolios differ in dividend growth rates. This study also finds that momentum returns are higher after the legislation change of REITs in 1992, and dividend/price ratios of REITs are also higher after 1992. Therefore, the persistent shock to REIT's dividend/price ratios in 1992 explains REIT's higher momentum returns after 1992, consistent with Johnson's prediction that infrequent shocks to business conditions cause persistent shocks to dividend growth rates. However, results of this study are not consistent with Johnson's theory for cyclical stocks, which states that cyclical stocks may have higher dividend growth rates during recessions, resulting in higher momentum returns in bear markets. Contrary to Johnson's prediction, this study finds that REIT's momentum returns are lower during bear markets. The explanation is that REIT's dividend/price ratios increase with market returns. In general, findings of this study support Johnson (2002) and Cooper, Gutierrez, and Hameed (forthcoming).
展开▼
