Abstract: This paper combines a data-rich environment with a machine learning algorithm to provide new estimates of time-varying systematic expectational errors ("belief distortions") embedded in survey responses. We find sizable distortions even for professional forecasters, with all respondent-types over-weighting the implicit judgmental component of their forecasts relative to what can be learned from publicly available information. Forecasts of inflation and GDP growth oscillate between optimism and pessimism by large margins, with belief distortions evolving dynamically in response to cyclical shocks. The results suggest that artificial intelligence algorithms can be productively deployed to correct errors in human judgment and improve predictive accuracy.
Abstract: We build a model in which the Fed and the market disagree about future aggregate demand. The market anticipates monetary policy "mistakes," which affect current demand and induce the Fed to partially accommodate the market's view. The Fed expects to implement its view gradually. Announcements that reveal an unexpected change in the Fed's belief provide a microfoundation for monetary policy shocks. Tantrum shocks arise when the market misinterprets the Fed's belief and overreacts to its announcement. Uncertainty about tantrums motivates further gradualism and communication. Finally, disagreements affect the market's expected in ation and induce a policy trade-off similar to "cost-push" shocks.
摘要：我们建立了一个模型，其中美联储和市场对未来的总需求有不同意见。市场预期货币政策会出现 "错误"，从而影响当前的需求，并促使美联储部分地适应市场的观点。美联储预计将逐步实施其观点。揭示美联储信念发生意外变化的公告为货币政策冲击提供了微观基础。当市场误解了美联储的信念并对其公告作出过度反应时，就会出现发泄性冲击。对不确定性激励了进一步的渐进主义和沟通。最后，分歧影响了市场的预期，诱发了类似于 "成本推动 "冲击的政策权衡。
Abstract: Smallholder farming in many developing countries is characterized by low productivity and low quality output. Low quality limits the price farmers can command and their potential income. We conduct a series of experiments among maize farmers in Uganda to shed light on the barriers to quality upgrading and to study its potential. First, we document that quality is low but partly observable. Second, we show that the causal return to quality is zero, suggesting that the market for quality maize is effectively missing. Third, we generate experimental variation in access to a market for premium quality maize, combined with training on agricultural best-practices, and document large increases in both farm productivity and income. Fourth, we show that agricultural training alone does not affect agricultural outcomes. Our findings reveal the importance of demand-side constraints in limiting rural income and productivity growth.