Have revisions to GDP over time (from new source data and improved methods) raised GDP growth?
No. Average revisions from the current quarterly estimates—advance, preliminary, and final—to the latest available estimates are not significantly different from zero. That is, even though GDP revisions are non-zero, they are “unbiased.” In other words, on average they neither raise nor lower GDP growth. Most GDP revisions result from the incorporation of new or revised source data that were not available at the time of the earlier estimates. In addition, annual revisions to quarterly estimates include the effects of periodic improvements in estimation methods that are necessary to keep the GDP attuned to the changing economy. Methodological improvements typically are put in place as part of once-every-five-year “comprehensive” GDP revisions. The most recent such revision, in 2003, yielded long-run growth rates of current-dollar and real GDP, measured from 1930-2002, that were virtually the same as the pre-revision growth rates. For more information on characteristics of GDP revisions,
