Economic growth and productivity in the United States, Canada, United Kingdom, Germany, and Japan in the post-war period

Abstract

T HIS is a brief summary of a long report on the rates of growth of outputs, inputs, and factor productivities in the United States, Canada, United Kingdom, Germany, and Japan in the post-war period, for the countries as a whole and for major economic sectors. Like many empirical studies, this paper shows numerous scars from battling statistical data: frequent use of the n. a. abbreviation, insufficient disaggregation (particularly for Germany), heavy reliance on ingenuity in bridging statistical gaps, and finally a rather weak conceptual framework chosen under duress. For all these reasons, the reader is urged to take our findings with a good dose of salt. Space does not permit us to discuss the numerous qualifications, sources, and statistical procedures. These are available, however, in the unabridged report which will be furnished on request.1 Since there exists a large literature on the methodology of such studies, we can be brief here.2 the whole, we have used the Kendrick method in obtaining what he calls the of Total Factor and what is called here the as well as in measuring specific factor productivities, with the following major modifications: (1) the outputs (in their several variants) are expressed gross rather than net of depreciation; (2) labor input is aggregated without being weighted by the average wage of each industry, as Kendrick did; (3) imports (when present) are treated as inputs. Much as we wanted to deviate from Kendrick and to include materials among the inputs (in the several sectors), lack of data forced us to follow him in subtracting material inputs from both sides of the production equation and to express output as value added (in constant prices). This exclusion of material inputs is unfortunate: it reduces the usefulness of Kendrick's Index and of our Residual, the most novel feature of both studies, because labor endowed with a large weight dominates the input side (capital playing a rather minor role) and frequently pushes the Residual rather close to the conventional measure of labor productivity (see below). One special qualification of our results should be mentioned. Neither Kendrick's nor our indexes of productivity have been corrected for the degree of utilization of inputs. During slack periods, even labor, although paid for, is not fully utilized, nor of course is capital. Hence both the rates of growth of the index of labor productivity and of the Residual will usually rise during the expansion phase of the economy or a sector and fall during a contrac* The original study was sponsored by the Organization for Economic Cooperation and Development, whose financial support is gratefully acknowledged. The study does not necessarily reflect the views of, or the approval by, the sponsor. We are grateful to Professor John M. Kendrick and his assistant, Mrs. Maude Pech of the National Bureau of Economic Research, for their generous help; they are not, of course, responsible for our errors and conclusions. Mr. Carl Riskin and Mrs. Marilyn Wright helped with computations; Miss Sharon Dewar deserves thanks for her aid in typing and computations, and Mrs. Juliet Raventos for preparing the final copy. We are also grateful to the Center for Advanced Study in the Behavioral Sciences at Stanford, where the final version of this paper was written, for the use of its facilities and for many other things. 1 Please contact E. D. Domar, Department of Economics, Massachusetts Institute of Technology, Cambridge 39, Massachusetts. 2 The most important recent work on factor productivities belongs to John W. Kendrick, Productivity Trends in the United States, a study by the National Bureau of Economic Research (Princeton, 1961). See also E. D. Domar's methodological comments in On the Measurement of Technological Change, The Economic Journal, LXXI (December 1961), 709-29, and in On Total Productivity and All That, The Journal of Political Economy, LXX (December 1962), 597-608. The first paper also contains a partial bibliograDhv on this subject.