Reconciling the Return Predictability Evidence, with Martin Lettau
 

ABSTRACT:

Evidence of stock return predictability by the dividend-price ratio is still controversial as documented by inconsistent results for in-sample and out-of sample regressions. This paper shows that these seemingly incompatible results can be reconciled if the assumption of a fixed steady state mean of the dividend-price ratio is relaxed. We find strong empirical evidence in support of shifts in the steady state and propose simple methods to adjust the dividend-price ratio for such shifts. The forecasting relationship of adjusted dividend-price ratios and returns is statistically significant, stable over time and performs well in out-of-sample tests. We also show in simulations that shifts in the steady state are responsible for parameter instability and poor out-of-sample performance of the unadjusted dividend-price ratio that is found in the data. Our conclusions hold when we use earnings as a cash flow measure and are robust to changes in the econometric technique used to estimate shifts in the steady state.