WHY BUBBLE-BURSTING IS UNPREDICTABLE: WELFARE EFFECTS OF ANTI-BUBBLE POLICY WHEN CENTRAL BANKS MAKE MISTAKES

This paper examines the effect of bubble-bursting policy in the case where the central bank sometimes tries to deflate an asset which is not, in fact, overpriced. We consider the case of a “semi-bubble,” where some traders know that an asset is overpriced, but others do not. Unlike most previous papers on bubble policy, our framework assumes rational traders. We also assume a finite time horizon, to rule out infinite horizon type bubbles. The market’s “fulfilled expectations” equilibria are derived, and standard tools of welfare economics are applied to evaluate the effect of anti-bubble policy. Under the assumption that the announcements of the financial authority can help less informed traders to learn more about a risky asset, market equilibria are presented and compared. We show that, if sellers care relatively more about the states where the central bank makes a negative bubble-bursting announcement, an announcement policy interferes with the asset’s ability to share risks. Conversely, if sellers care relatively less about the announcement states, an announcement policy improves risk sharing. “Information leakage” plays an important role in our analysis. Because of this leakage, central bank announcements can initiate further information revelation between traders. That is, the leakage effect can reveal information that the central bank, itself, does not have. However, this information leakage may not be welfare improving. Also, this leakage effect makes it difficult to predict the effects of bubble-bursting policy. This may complicate both private investment strategies and public policy analysis.