No document available.
Abstract :
[en] Why is investment in stocks so persistently weak after a rare disaster? Connecting disaster episodes with post-disaster expectations seems crucial for such post-disaster forecasting and also policymaking, but rational-expectations models with variable disaster risk often fail to achieve this connection. To this end, while retaining full rationality, we introduce limited information and learning about rare-disaster risk and show that the resulting stock-investment behavior seems similar to persistent investor fear after a rare disaster. We study, (a) rational learning for state verification (RLS), with investors knowing the data-generating process of disaster riskiness but being unable to observe whether the economy is in a riskier state (regime) or not, and (b) rational learning about the data-generating process (RLP) of disaster risk, with investors also being unaware of the data-generating process of disaster riskiness. We analytically show that both RLS and RLP synchronize disaster events with post-disaster expectations and asset prices, and create persistence in price-dividend ratios even if data-generating processes of disaster risk have no persistence. Using De Finetti's theorem we show that RLP offers an explanation for global spells of pessimism and weak investment after a disaster.