Newspapers Can Predict the Economy Better than Confidence Indexes

The phrase "news you can use" is taking on a whole new meaning.

Textual analysis of articles and editorials about the economy can be scrutinized to track current economic conditions pretty well: so well, in fact, that they often beat standard consumer sentiment surveys as forecasting tools. That's the finding in a new Federal Reserve Bank of San Francisco study, which leads today's economic research wrap. We also take a look at common inequality misconceptions, the link between unemployment and depression, and GDP measurement in a digital age.

Check this column every Tuesday for new and interesting studies from around the globe.

San Francisco Fed research advisers Adam Hale Shapiro and Daniel J. Wilson and co-author Moritz Sudhof have come up with a technique for tracing economic sentiment in newspaper articles based on textual analysis. Even better, it seems to work. The researchers applied their method to financial news articles published from January 1980 to April 2015, and in most cases, the news sentiment indexes proved better at predicting economic activity than consumer sentiment measures in head-to-head comparisons.

"These methods of sentiment text analysis hold great promise for improving our understanding of news sentiment shocks and how they affect the economy," they write.