For more than three decades, CBO has prepared economic forecasts that underlie the agency’s projections for the federal budget and cost estimates for proposed federal legislation. In particular, forecasts of output, income, inflation, and interest rates play a significant role in the agency’s budgetary analysis; for example, projections of wages and salaries are used to forecast individual income tax receipts.
CBO regularly evaluates the quality of its economic forecasts by comparing them with the economy’s actual performance and with forecasts by the Administration and the Blue Chip consensus—an average of about 50 private-sector forecasts. Such comparisons indicate the extent to which uncertainty and imperfect information may have caused CBO to “miss” patterns or turning points in the economy. They also identify areas where CBO has tended to make larger errors or less accurate forecasts than other analysts.
CBO’s forecasts generally have been comparable in quality with those of the Administration and the Blue Chip consensus. When CBO’s projections have proved inaccurate by large margins, the errors have tended to reflect difficulties shared by other forecasters, as can be seen in the figure below regarding two-year forecasts of real (inflation-adjusted) output growth.
A simple and widely used indicator of statistical bias is the mean error—the average tendency of a forecast to be low or high over an entire period. In general, CBO’s forecasts and those by the Administration and the Blue Chip consensus have had similar mean errors. Specifically, CBO’s evaluation finds this:
Accuracy is the degree to which forecast values are dispersed around actual outcomes. One widely used measure of accuracy is the root mean square error. By that measure, the forecasts by CBO, the Administration, and the Blue Chip consensus have been about equally accurate over two-year periods as well as over five-year periods. CBO’s evaluation finds this:
Sources of large forecasting errors have included the difficulty of predicting:
In addition, revisions to the historical data (on output and income, for example) that forecasters use for economic projections can complicate the task of interpreting forecasting errors. CBO used current vintages of historical data to compute the forecasting errors and statistics. Had the revised data been available to forecasters, rather than the original information that was available when the forecasts were produced, the forecasts themselves would have been different. Despite that complication, recently published data present a simple and consistent point of comparison for evaluating forecasts by CBO and others.
CBO constructs its economic projections under the assumption that federal fiscal policy will follow current law, thereby providing a benchmark for lawmakers as they consider potential changes in the law. In contrast, the Administration’s forecasts assume the adoption of policies reflected in the President’s proposed budget. Forecasters in the private sector (represented in the Blue Chip consensus) form their own assumptions about the future stance of federal fiscal policy, which may anticipate changes in law.
Differences between forecasts, and thus differences in forecasting errors, sometimes arise from different assumptions about fiscal policy, particularly when policymakers are considering major changes to current law. For example, in 2009 and 2010, different fiscal policy assumptions caused CBO’s two-year forecasts of real output growth to diverge noticeably from those of the Administration and the Blue Chip consensus.
This report was prepared by Stephanie Burns of CBO’s Macroeconomic Analysis Division.