Tuesday, September 15, 2020

News sentiment index from SF Fed is useful


We have seen over the last few years indices aggregating news stories associated uncertainty. There are now complete suites of uncertainty indices, see Economic Policy Uncertainty. The uncertainty index methodology is based on three components. One component quantifies newspaper coverage of policy-related economic uncertainty. A second component reflects the number of federal tax code provisions set to expire in future years. The third component uses disagreement among economic forecasters as a proxy for uncertainty.

There is now a new index from the San Francisco Fed Research Department that focuses on sentiment from news stories. This Daily News Sentiment Index is a high frequency measure of economic sentiment based on lexical analysis of economics-related news articles from leading newspapers. The index looks at news stories from a wide range of publication and uses a sentiment scoring method based on economic news lexicon to make predictions on consumer sentiment. This is different than the focus on policy uncertainty indices. 


The news sentiment index shows a steep sell-off during recessions consistent with a fall in consumer sentiment; nevertheless, there are also some sharp declines during non-recession periods. 

The link between the news sentiment index and the stock market is mixed. The news sentiment is a high frequency index, so it shows significant variation versus the trend in stocks. As usual, the market often follows its own mind and does not match news sentiment; however, sharp extreme drops below zero suggest weakness in stock prices. 

No comments: