Economic news typically causes market participants to reassess their expectations about the economy’s current and future condition. This reassessment usually results in adjustments in prices for financial assets. Thus, the release of a given economic indicator can be expected to affect interest rates, exchange rates, and stock prices in different ways. In particular, a small unexpected change in one of these indicators may rock bond yields and exchange rates but have little impact on stocks.
Earlier studies of news effects usually measured the effects of news on asset prices by comparing them with predictions from an empirical forecasting model. This made the results highly dependent on the model used and made it difficult to discern whether weak impacts of news were due to an inadequate forecasting model or a lack of true economic significance.
We use a simple but powerful new methodology to estimate the direct and indirect effects of economic news on key U.S. asset prices–bond yields, stock prices, and the value of the dollar against the euro. The analysis suggests several important lessons. First, only a few economic announcements–nonfarm payrolls and the GDP advance release–generate price responses that are economically significant and measurably persistent through the day. Bond yields exhibit the strongest response and stock prices the weakest.
We also show that the standard approach to measuring asset price responses to economic news–which uses survey data to capture prerelease expectations–may be flawed. This is because the surveys are typically conducted with a lead of a few days or more, which means that by the time the data is released, much of the “measured news” has already been incorporated into prices through a process called informed noise. The Rigobon-Sack approach cleans this information out, resulting in estimates of asset price responses that agree in sign with those using the standard measure but that are typically larger in magnitude.