A1387
Title: Impacts of macroeconomic data revisions on economic activities
Authors: Naoko Hara - Seikei University (Japan) [presenting]
Abstract: Uncertainty necessitates data-driven decision-making. While sound decisions require timely and accurate economic information, major official statistics are often substantially revised long after their first release. These data revisions imply the presence of significant noise in preliminary data. The purpose is to examine the economic impacts of the noise in U.S. macroeconomic data. Noise components are identified in U.S. GDP by estimating structural vector autoregression models using different data vintages. It is found that the noise shocks significantly affect economic fundamentals and professional forecasts in the short run. Moreover, the results suggest that the GDP growth is likely to be overestimated in real time, and forecasters tend to respond differently when uncertainty increases. These findings highlight a need for caution when interpreting early releases of macroeconomic data, particularly amid heightened uncertainty.
