Measurement of Productivity Growth in U.S. Manufacturing
Gullickson, William, Monthly Labor Review
The indexes of multifactor productivity for two-digit manufacturing sectors prepared by the Bureau of Labor Statistics for sectors in manufacturing have been revised and extended to cover the 1949-92 period. These indexes, also called the "KLEMS" multifactor measures, compare changes in output to changes in a composite of all the inputs used in production--capital, labor, energy inputs, nonenergy material inputs, and business services.(1) Because of this comprehensive input list, these indexes give an indication of advances in technology and production efficiency in these broad sectors, important topics as the economy emerges from the recession of the early 1990's.
This article discusses the measurement of multifactor productivity for manufacturing and analyzes growth trends within the sector. Through the years, a wide variety of productivity statistics have appeared in the literature, distinguished by the concepts underlying the measurement of output, the methods of aggregation, and the inputs included for analysis. Recent additions of "superlative" indexes of gross domestic product (GDP) by industry to the U.S. National Income and Product Accounts, prepared by the Bureau of Economic Analysis, U.S. Department of Commerce, have enhanced available alternatives for measuring manufacturing productivity. Planned changes in the way BLS measures manufacturing productivity are also discussed.(2)
Multifactor productivity growth trends are then examined for the overall manufacturing sector and for 19 two-digit SIC manufacturing subsectors.(3) In particular, early postwar and more recent productivity growth trends are compared. When this comparison was last discussed in 1992, data were available through 1988, covering a period of rapid growth following emergence from the 1982 recession.(4) Because of this growth, multifactor productivity growth seemed to have regained much of its early postwar momentum. It is now possible to examine recent trends more comprehensively because the extended series cover the 1990 business cycle peak, the brief recession that followed in 1991, and a recovery period in 1992. These trends indicate that the productivity growth rates of the early postwar period were not entirely regained during the 1980's.
Issues in measurement
Until recently, the Bureau of Labor Statistics produced two distinct and fairly different measures of multifactor productivity for the manufacturing sector. One measure was a comparison of "net" output to capital and labor inputs.(5) The other was the KLEMS multifactor measure, issued along with multifactor productivity measures for broad (two-digit SIC) manufacturing industries, which compares "sectoral" output to capital, labor, and "intermediate" inputs.
In the future, BLS will use the measure based on sectoral output for both purposes, while continuing to use a somewhat modified net output-type multifactor productivity measure for its international comparisons of multifactor productivity. Some further background on measures which help explain these changes are provided in the following sections. The main issues have to do with which inputs and outputs should be included in a multifactor productivity ratio and how heterogeneous inputs and outputs should be weighted together.
Basic principles. BLS is engaged in efforts to insure that its measures conform, as nearly as possible, to some basic principles of productivity measurement which have been developed in the economics literature.(6) One of the basic principles is that inputs be as comprehensive as possible, so that productivity growth does not merely reflect changes through time in unmeasured inputs. Thus, the multifactor productivity measures for manufacturing industries presented later in this article take into account all intermediate inputs (energy and other materials and business services). The importance of intermediates first gained prominence in the literature because of the events of the 1970's. …