A recent study from the Federal Reserve indicates that consumer sentiment and the tone of news coverage can forecast economic recessions similarly to traditional data like job statistics and prices. The working paper, released on July 17, was authored by economists from the Federal Reserve Bank of San Francisco, including Nicolas Petrosky-Nadeau, Yeji Sung, and Daniel J. Wilson.
The researchers examined whether 'soft' data, such as measures of consumer sentiment and economic-policy uncertainty, could predict recessions as effectively as 'hard' statistics. Their findings suggest that sentiment models may even outperform traditional data in some cases, particularly in identifying recession risks one month in advance. However, these models also generated more false alarms.
The authors emphasize that while sentiment data is valuable, it should complement traditional economic indicators rather than replace them. AcadeResearch noted that the study quantifies the predictive power of soft data over short time frames, accounting for various factors such as publication lags and revisions.
The research utilized data from August 1999 to May 2026, covering three recessions, with sentiment inputs from sources like the University of Michigan consumer surveys and the San Francisco Fed's Daily News Sentiment Index. While the findings provide insights for households and businesses about economic trends, the authors caution that the paper reflects their views and does not predict an imminent recession.
