Evaluation of labor force projections to 1990 (EXCERPT), Monthly Labor Review Online, Aug. 1992

As the final step in the projection process, the Bureau of Labor Statistics assesses its labor force projections.¹ Such evaluations help persons making projections better understand the types of problems and errors that could occur and allow users to focus on the accuracy of projections for a specific group in the labor force or on overall accuracy.

This article examines the errors in the labor force projections to 1990 and their sources. It does this by examining projected levels of the labor force and the rates of labor force participation of specific age groups for men and women, and for whites and blacks and others. Where appropriate, the accuracy of the 1990 labor force projections is compared with evaluations of BLS projections of the 1975, 1980, and 1985 labor force.²

The 1990 projections
Each of the six projections to 1990 had three alternatives: high, moderate, and low. This analysis, for the most part, focuses on the middle or "moderate" growth projection in each series. (See table 1.) The following tabulation shows the projections to 1990 (in millions) and the level and percentage point errors made in each year the projections were developed.³

Projection for 1990 made in:	Labor force	Error	Percent error
1973 ..........	110.6	-14.2	11.4
1976 ..........	113.8	-10.9	8.8
1978 ..........	119.4	-5.4	4.3
1980 ..........	122.4	-2.4	1.9
1983 ..........	125.0	.2	.1
1985 ..........	122.6	-2.1	1.7

1990 Labor force (actual)	124.8	--	--

The overall error became progressively smaller through the 1983 projection, when it was 0.1 percent, or fewer than a quarter of a million persons, but increased in the next projection (1985) to 1.7 percent, to near that of 1980. What were the sources of labor force error and why, with one exception, did the error fall as time passed?

This excerpt is from an article published in the August 1992 issue of the Monthly Labor Review. The full text of the article is available in Adobe Acrobat's Portable Document Format (PDF). See How to view a PDF file for more information.

Footnotes
¹ For the 1990 evaluations, see the following Monthly Labor Review articles: Denis F. Johnson, "The U.S. labor force: projections," July 1973, pp. 3-*13, reprinted as Special Labor Force Report 156; Howard N. Fullerton, Jr. and Paul O. Flaim, "New labor force projections to 1990," December 1976, pp. 3-13, reprinted as Special Labor Force Report 197; Paul O. Flaim and Howard N Fullerton, Jr. "Labor force projections to 1990: three possible paths," December 1978, pp. 25-35, reprinted in Employment Projections for the 1980's, Bulletin 2030 (Bureau of Labor Statistics, 1979); Howard N Fullerton, Jr., "The 1995 labor force: a first look," December 1980, pp. 11-21, reprinted in Economic Projections to 1990, Bulletin 2121 (Bureau of Labor Statistics, 1982); Howard N. Fullerton, Jr., and John Tschetter, "The 1995 labor force: a second look," Monthly Labor Review, November 1983, pp. 3-10, reprinted in Employment Projections for 1995, Bulletin 2197 (Bureau of Labor Statistics, 1984); and Howard N Fullerton, Jr., "The 1995 labor force: BLS's latest projections," Monthly Labor Review,, November 1985, pp. 17-25, reprinted in Employment Projections for 1995: Data and Methods, Bulletin 2253 (Bureau of Labor Statistics, 1986).

² See Sol Swerdloff, "How good were manpower projections for the 1960's," Monthly Labor Review,, November 1969, pp. 17-22; Paul Ryscavage, "BLS labor force projections: a review of methods and results." Monthly Labor Review, April 1979, pp. 14-22; Howard N Fullerton, Jr., "How accurate were the projections of the 1980 labor force?" Monthly Labor Review, July 1982, pp. 15-21; and Howard N Fullerton, Jr., "An evaluation of labor force projections to 1985," Monthly Labor Review, November 1988, pp. 7-17.

³ In this analysis, we compare the projected labor force numbers for 1990 with the annual average estimates of the labor force derived from the Current Population Survey, using weights from the 1980 census. We call such estimates "the actual."