Economic news can cause sudden and lasting shifts in asset prices. In particular, market participants reassess their expectations about future growth and inflation in response to changes in the data. Basic economic thinking suggests that the resulting adjustment should prompt higher bond yields and the exchange rate value of the dollar, while stock prices rise less briskly.
However, earlier studies estimating announcement effects typically defined the new information as a change in the predictions of an empirical forecasting model. This approach made it difficult to tell whether a finding that news had little impact was indicative of weak economy-wide impacts or simply a result of poor forecasts.
The authors use an alternative method that eliminates the need for a forecasting model and provides more reliable estimates of asset price responses to economic data. The approach, known as the Rigobon–Sack method after its developers, measures the true new information contained in a survey of market expectations by subtracting the indicator value released from the survey value expected one instant before the release. The Rigobon–Sack estimates of the impact of each announcement are comparable to those based on standard survey measures but clean of measurement errors and informa- tional noise that accumulated between the survey and the data release.
The results suggest that a significant amount of economic news is essentially noise, while a small number of announcements generate significant and measurably persistent asset price responses. The strongest effects are observed for announcements affecting interest rates, with those involving GDP advance releases and private sector manufacturing reports having the weakest impact.
About the Author
