We consider whether sentiment affects the profitability of momentum strategies. We hypothesize that news that contradicts investors’ sentiment causes cognitive dissonance, slowing the diffusion of such news. Thus, losers (winners) become underpriced under optimism (pessimism). Short-selling constraints may impede arbitraging of losers and thus strengthen momentum during optimistic periods. Supporting this notion, we empirically show that momentum profits arise only under optimism. An analysis of net order flows from small and large trades indicates that small investors are slow to sell losers during optimistic periods. Momentum-based hedge portfolios formed during optimistic periods experience long-run reversals.

This study presents a theoretical model that links chief executive officer (CEO) overconfidence to the value loss of corporate diversification. Consistent with the model's prediction, the findings show that diversified firms run by overconfident CEOs experience value loss compared to diversified firms run by their rational counterparts. Empirically, the value loss is economically significant and ranges between 12.5% and 14.1%. In addition, the model predicts heightened corporate refocusing activity by overconfident CEOs who pursued diversified investments in the past once realized returns fail to match initial expectations. The empirical odds of corporate refocusing decisions are 67% to 98% higher when past diversifications are undertaken by overconfident rather than rational CEOs. Another prediction of the model is that overconfident CEOs exhibit preference for diversified investments, especially in the presence of ample internal funds. This prediction is also strongly supported by the data. Overall, this study proposes CEO overconfidence as a unified and consistent explanation of why firms pursue value-destructive corporate diversification policies and later adopt refocusing policies aiming to restore value.