Revisions to BLS quarterly labor productivity estimates: How large are they?

Abstract

BLS quarterly estimates of labor productivity, or output per hour worked, are revised as more data become available and seasonal factors are updated. BLS publishes three regularly scheduled estimates of labor productivity growth: the preliminary estimates, the first revised estimates, and the second revised estimates. Most of the revisions result from updates to source data. This article analyzes revisions to the preliminary and first revised estimates relative to the second revised estimates and describes the new BLS methodology for constructing intervals that better informs data users about the likely size of future revisions. As part of our analysis, we identify the sources of revisions and determine whether systematic patterns exist that could be exploited to construct intervals. We find no evidence of such trends or systematic patterns.