Equity Market Uncertainty Index
News Coverage about Equity Market-related Economic Uncertainty
From August 7, 2013, we are featuring an improved version of our daily equity uncertainty index. The new version offers greater accuracy in the most recent data, and it better conforms to the measurement practices used in our monthly news-based EPU index. The old version of the daily equity uncertainty index remains available here and will continue to be updated. In addition, on August 16th, we fixed a bug that caused a small number of dates to display inflated values (approx 50x) for the true level of equity uncertainty.
We construct a measure of Equity Market-related Economic Uncertainty through an analysis of news articles containing terms related to equity market uncertainty. For this exercise, we use newspapers from Access World New's NewsBank service. The NewsBank Access World News database contains the archives of thousands of newspapers and other news sources from across the globe. While NewsBank has a wide range of news sources, from newspapers to magazines to newswire services, we conduct our analysis only utilizing newspaper sources.
We restrict our attention to newspapers in the United States, of which NewsBank covers well over 1000. These newspapers range from large national papers like USA Today to small local newspapers across the country. The data spans from 1985 to 2013. To construct the index, we perform month-by-month searches of each paper, starting in January of 1985, for terms related to economic and policy uncertainty. In particular, we search for articles containing the term 'uncertainty' or 'uncertain', the terms 'economic' or 'economy' and one or more of the following terms: 'equity market', 'equity price', 'stock market', or 'stock price'. In other words, to meet our criteria for inclusion the article must include terms in all three categories pertaining to uncertainty, the economy, and the stock market.
The number of newspapers that NewsBank covers over our sample period has drastically increased from 18 in 1985 to over 1800 by 2008. To correct for this growth, we must normalize our index. To do this, we take daily counts of the total number of newspaper articles and normalize the equity market-related article count by the total number of articles. We then normalize the series to an average value of 100 from 1985-2010.
The resultant series has a contemporaneous daily correlation with the VIX of over 0.3. When averaged at a monthly frequency, the correlation with the monthly VIX rises to over 0.5.