130 likes | 432 Views
Rapid Estimates of U.S. GDP:. Timeliness, Estimating Methods & Accuracy. Dave Wasshausen. International Seminar on Timeliness, Methodology and Comparability of Rapid Estimates of Economic Trends Ottawa, Canada May 27-29, 2009. Timeliness.
E N D
Rapid Estimates of U.S. GDP: Timeliness, Estimating Methods & Accuracy Dave Wasshausen International Seminar on Timeliness, Methodology and Comparability of Rapid Estimates of Economic Trends Ottawa, Canada May 27-29, 2009
Timeliness • “Advance” estimates of GDP released toward the end of the first month after the end of the quarter. • Based on partial source data and BEA assumptions • Complete set of expenditures estimates; partial set income estimates • Corporate profits and net interest are not prepared due to insufficient source data
Timeliness • Tradeoff between quality and timeliness • Try to strike the appropriate balance between accuracy and timeliness • BEA releases several “vintages” of GDP estimates, each subsequent vintage based on better source data
Source Data for Advance Estimate • Three months of source data are available for: • Consumer spending on goods • Shipments of capital goods excluding aircraft • Motor vehicle sales and inventories • Manufacturing durables inventories • Federal government outlays • Consumer, producer and international prices • Only two months of data are available for most other key source data (including construction spending, foreign trade, & remaining inventory series)
Estimating Methods • Retail Control Method • Commodity Flow Method • Price-Times-Quantity Method • Judgmental Trend • Imputations
Revisions and Accuracy • Revisions and accuracy are related, but not the same • Some of the most inaccurate data have never been revised • Some of the most accurate data are subject to regular revisions • Revisions must be judged against average growth and volatility in growth • Publish regular revision studies (Feb. 2008)
Revisions • Single largest source of revisions is updating of seasonal factors • Largest source of revisions in trend growth is changes in concepts, methods, and source data intended to update the accounts and improve accuracy.
Accuracy • Current quarterly estimates of real GDP correctly indicate: • Direction of change 98 percent of the time • Whether accelerating or decelerating 74 percent of the time • Whether growth is above, near, or below trend growth more than three-fifths of the time. • Estimates have also been shown to be un-biased, with no identifiable bias from the introduction of real time data
Revisions to Real GDP estimates Real GDP Growth Rate: Quarters, 2008
Revisions to GDP for QII, QIV • Large changes in energy prices (upward in QII, downward in QIV). • Wedge between GDP and Gross domestic purchases prices. • Inventory valuation adjustment. • Difficulty of projecting data for the missing month. • Importance of publishing “key assumptions.”
Conclusion • BEA’s goal: Early estimates that provide users with a useful “snapshot” of the economy (in the aggregate and by component) • BEA’s early estimates do a relatively good job providing an overall picture of economic activity. In general, they tell us: • If the economy is expanding or contracting • If growth is accelerating or decelerating • If growth is high or low relative to trend • What major components are contributing to growth • The timing of business cycle peaks and troughs