BEA's strategic plan for 2001-2005

Survey of Current Business, May, 2002

I have just completed a new paper giving detailed production accounts by industry. These incorporate the BEA interindustry transactions accounts. ("Information Technology, Higher Education, and the Sources of Economic Growth across U.S. Industries," with Mun S. Ho and Kevin J. Stiroh, to be presented to the Conference on Research in Income and Wealth, Washington, DC, April 26-27, 2002.) For each industry the output is BEA's "gross output," and the input is broken down by capital, labor, and intermediate inputs. Each of these is presented in current and constant prices.

The key innovation in this production account is the introduction of the concept of the flow of capital services. This is employed in the NIPA's in measures of the rental value of housing. The new architecture extends this idea to all categories of assets included in the BEA capital stock study. A parallel concept of the flow of labor services is broken down by age, sex, education, and class of employment with individual components weighted by total compensation per hour worked. The detailed architecture is laid out in Paul Schreyer's Productivity Manual, published by the Organisation for Economic Co-Operation and Development in 2001.

What are the next steps in developing a new architecture for the production account? The first order of priority should be development of a conceptual framework for integrating the NIPA's and the BEA interindustry transactions accounts. This has been done by Robert Yuskavage ("Priorities for Industry Accounts at BEA," paper presented to the BEA Advisory Committee, November 17, 2000). A very important detail is providing a time series link between the industry accounts before and after the introduction of the North American Industry Classification System (NAICS).

A longer-term issue is consideration of production of annual interindustry transactions tables on the same schedule as the NIPA's. This is already done by the Office of Occupational Statistics and Employment Projections at the Bureau of Labor Statistics (BLS), but using less detailed data than in BEA's annual tables. However, the BLS tables are available at the same time as the NIPA's. Unfortunately, they do not incorporate the latest information from the annual revisions of the NIPA's. A system for producing the two data sets simultaneously is already in place in Australia and Canada and has been adopted by the United Kingdom. This should be considered by BEA.

The third step would be construction of a production account at both aggregate and industry levels along the lines I have suggested. Fortunately, much of the required work is already available, at least in prototype, in the papers I have written on the production account. These are carefully integrated with the NIPA's and other data sets produced by BEA, such as gross product originating, the capital stock study, and hours worked. Unfortunately, my papers inherit some of the gaps in the BEA data sets, such as the inconsistency between the NIPA's and the interindustry transactions accounts.