Time series methods for identifying structural economic disturbances often require disturbances to satisfy technical conditions that can be inconsistent with economic theory. We propose replacing these conditions with a less restrictive condition called recoverability, which only requires that the disturbances can be inferred from the observable variables. As an application, we show how shifting attention to recoverability makes it possible to construct new identifying restrictions for technological and expectational disturbances. In a vector autoregressive example using post-war U.S. data, these restrictions imply that independent disturbances to expectations about future technology are a major driver of business cycles.

The literature on belief-driven business cycles treats news and noise as distinct representations of agents' beliefs. We prove they are empirically the same. Our result lets us isolate the importance of purely belief-driven fluctuations. Using three prominent estimated models, we show that existing research understates the importance of pure beliefs. We also explain how differences in both economic environment and information structure affect the estimated importance of pure beliefs.

This paper exploits a data rich environment to provide direct econometric estimates of time-varying macroeconomic uncertainty. Our estimates display significant independent variations from popular uncertainty proxies, suggesting that much of the variation in the proxies is not driven by uncertainty. Quantitatively important uncertainty episodes appear far more infrequently than indicated by popular uncertainty proxies, but when they do occur, they are larger, more persistent, and are more correlated with real activity. Our estimates provide a benchmark to evaluate theories for which uncertainty shocks play a role in business cycles.