Cleveland Fed researcher finds real-time reliability of monthly indicators has changed over time, cautions policymakers to assess revisions
Economic policymakers carefully assess all recent economic data for clues about the current state of the economy and the direction it may be headed, even as a unique feature of economic data is that they are subject to revisions. In this Economic Commentary, Cleveland Fed researcher Mark Bognanni looks at the extent to which the real–time reliability of six monthly macroeconomic indicators important to policymakers has remained stable over time by studying the time-series properties of their short-term and long-term revisions.
“The model estimates imply that both short–term and long–term revisions exhibit systematic deviations from zero that policymakers could build into their real-time assessments,” says Bognanni. “I also find some indicators' revision series have seen substantial variation over time in the size of the typical revision, which suggests that some series have become less useful real–time indicators than they once were.”
When examining the long–term revisions, Bognanni finds that substantial revisions tend to occur indefinitely after the initial data release, a result that suggests a certain degree of caution is in order when using even thrice–revised monthly data in policymaking.