Distinguished Lecture on Economics in Government: Data Quality and Public Policy
The goal of a government statistical agency must be to produce data that are objective, relevant, accurate, and timely. But none of these criteria is absolute, and sometimes tradeoffs must be made among them. Timeliness and accuracy compete when we decide to issue preliminary results that are subject to revision. Accuracy and relevance compete when we decide whether to introduce data for states or local areas at the expense of reducing the reliability of national estimates. In making these choices, and in developing our statistical programs, we must manage a broad array of tasks. These can be divided into three broad areas of activity: 1) the management of economic design; 2) the management of statistical design; and 3) the management of the process by which we produce our final product. Let me review how we are addressing these issues at the Bureau of Labor Statistics.