Economic news causes investors to reassess the economy’s current state and its likely future evolution, and this reassessment prompts changes in asset prices. These effects typically vary across assets and over time, but they tend to be economically significant and measurably persistent. They also tend to be stronger for interest-bearing assets, such as bond yields and foreign exchange rates, than for stock prices. The strongest responses come from announcements of stronger-than-expected growth and inflation, while the weakest and most erratic are those that announce weaker-than-expected growth or inflation.
Economic surprise announcements are typically measured by comparing the change in an asset price or bond yield with its corresponding change in the indicator being announced, minus the indicator’s expected value at the moment of its release. However, many earlier studies of the impact of economic news defined the measure of news using the predictions of an empirical forecasting model, a practice that made the resulting estimate dependent on the model used to generate the forecasts. This can lead to spurious findings of weak or nonexistent effects.
To address these concerns, Rigobon and Sack developed a methodology for measuring the impact of economic news that cleans the estimates of asset price responses from survey data measurement errors and informational noise that accumulates between the survey and the data release. The results of our analysis confirm that the Rigobon-Sack methodology produces estimates of the immediate effects of economic news on asset prices that are both valid and considerably larger than those produced by the standard approach.