I would characterize quarterly GDP growth estimates as being of medium reliability. They are subject to frequent, and sometimes sizable, revisions. As another check on the GDP growth numbers I just cited, we can look at a second method that calculates gross domestic income, or GDI. In theory, GDI should equal GDP, but that's rarely the case in practice.

Economics is a highly sophisticated field of thought that is superb at explaining to policymakers precisely why the choices they made in the past were wrong. About the future, not so much.

Significant scope exists to improve the quality of price data as well--for example, by using the wealth of information available from checkout scanners or finding better ways to adjust for quality change.  I encourage researchers to become more familiar with the strengths and shortcomings of the data that they routinely use.  Besides leading to better analysis, attention to data quality issues by researchers often leads to better data in the longer term, both because of the insights generated by research and because researchers are important and influential clients of data collection agencies.

Data revisions have important implications for policymaking and policy evaluation.  Potentially, they can erode central bank credibility.  Partly because of data revisions, it is hard for us to spot turning points in the economy.  The fact is, statistical agencies fill in missing data with extrapolations that are especially likely to be wrong at these turning points because the estimates and probabilities they use are biased toward the latest trend.  The result is that we are likely to underestimate slowdowns and pickups at precisely the moment when we need to take corrective action.  

Innovations in economic statistics must keep pace with innovations in the economy. Barriers to useful sharing of information across statistical agencies should be removed to reduce costs and enhance benefits, but in no way should we compromise the high standards of privacy protection embodied in CIPSEA and other statutes. And allocating resources to better measure the most dynamic parts of the economy will help us get the most bang per buck in economic statistics by enhancing our ability to spot trends and improve forecasts of the direction of the economy.

I draw two lessons from that late '90s experience [the changing view of productivity growth]. The first is that you don't just look at the conventional measures, you look at the data quite deeply and try to understand how the data are constructed and how they relate to each other, because there may be anomalies that'll be instructive. And that was the case in the late '90s. The other is that published government data is not the only source of information. It's also important to talk to people in the marketplace, to talk to business people.

The Chicago Purchasing Managers’ report indicates that the overall index as well as the prices paid and the supply and delivery components moved down in December, but all three components remained well above the 50 percent level. I would remind you that these data should be treated confidentially, as they won’t be released until December 30th.

Remark made at December 21 FOMC meeting