Economic news causes markets to reassess the economy’s current state and its likely future evolution. It should therefore prompt changes in the prices of financial assets, including interest rates, exchange rates, and stock prices. Some announcements generate price responses that are economically significant and measurably persistent through the business day, whereas others cause little change. The authors’ analysis suggests that the magnitude and sign of the response depends on the type of surprise, but it generally supports the view that higher-than-expected growth or inflation prompts a rise in bond yields and the dollar value of stocks.
Earlier studies of the impact of announcements on asset prices relied on estimates of expected values of an economic indicator, typically defined in terms of predictions from an empirical forecasting model. This practice, however, carries with it several potential sources of error. One is the fact that survey data are conducted well in advance of actual indicator releases, with leads ranging from a few days to a week or more. This lag implies that by the time of the actual release, a large amount of information about the indicator already has accumulated, and some portion of the “measured news” may be redundant or already known.
To address these problems, the authors develop an alternative measure of “true” news that cleans away measurement errors and irrelevant information accumulated between surveys and data releases. They then compare the estimated asset price response to this new measure to estimate the real impact of an economic announcement. The results show that the Rigobon-Sack approach produces estimates of asset price reactions that are comparable in sign with those produced by standard methods, but that are free of the signaling and measuring errors that plague the standard approach.
About the Author
