Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Introduction, Findings, and Recommendations INTRODUCTION During the past several years, there has been a great deal of concern about the performance of the American economy, particularly about productivi- ty. Measured output per hour of all persons engaged in production in the private business sector has been rising much less rapidly in the past decade than it did in the two previous decades. This apparent slowdown in productivity growth causes serious concern for several reasons. A slowdown would mean a slower rate of growth in real income per capita in the standard of living. A slowdown in productivity growth could also contribute in important ways to the problem of the current high and persistent rate of inflation and to the imbalance in international payments. However, it was not the main purpose of the Panel and of this report to discuss either the causes or the consequences of the apparent slowdown in productivity growth; our task was rather to consider the measurement of productivity. Whenever any measure of economic performance becomes a source of concern, several sets of questions naturally arise. First, just what does the measure really mean? What are the underlying concepts? How well are they understood and used by economists, public officials, journalists, and the public? Second, how accurate and timely are the measures? What are the sources of error in them? How, if at all, can errors be reduced or eliminated? If one is to recommend policies to deal with a slowdown in the 3
4 REPORT OF THE PANEL measured rate of productivity growth, it is important to be sure that the measures are reasonably accurate, just as one would want to be sure before treating a patient for fever that the patient's temperature had been taken with an accurate thermometer. Third, what additional data should be collected to improve the measure? It is to these questions that this report is addressed. The Panel's first charge was to consider the major purposes for which productivity statistics are used and to recommend the most appropriate concepts to achieve those purposes. The Panel was to recommend how productivity might best be defined in the actual measurement process. The Panel was charged specifically with appraising present methods, identi- fying gaps, suggesting appropriate statistical methodology, and recom- mending new directions for research for statistical compilation. In carrying out this large assignment, the Panel decided to concentrate its attention on the statistical measurement of productivity as a tool for monitoring and promoting efficient production of goods and services. The Panel's work thus is directly related to the concepts of the national economic accounts, while broader welfare issues were considered more briefly. (For a fuller discussion of welfare issues, see Moss [in this volume].) Several members of the Panel have had previous experience as members of groups dealing with the relevance and accuracy of other sets of economic statistics. We have found the evaluation of productivity measures more difficult in one respect than that of most other economic statistics. Very few of the data used for productivity measures are collected for that purpose. Rather, most productivity measures are composites of other measures and data collected from a wide variety of sources for many different kinds of uses. The task of reviewing productivity measures thus comes uncomfortably close to that of reviewing the entire body of economic statistics collected by the federal government. Clearly, we could not do that task with the time and resources at our disposal. We have had to select what seemed to us to be the crucial aspects, and some may disagree with our selection. Nor did we want to duplicate the work of other groups that have reviewed in greater depth some of the components of productivity measures. Instead, we refer to their work where it is relevant and on occasion endorse some of their recommendations where they are applica- ble to our task. In particular, we do not discuss in this report the measures of productivity in agriculture produced by the U.S. Department of Agriculture; those measures are being reviewed by a task force of the American Agricultural Economics Association. We begin in Chapter 2 with a discussion of the uses, misuses, and
Introduction, Findings, and Recommendations limitations of productivity measures. This discussion provides an introduc- tion for those without much previous exposure to the measurement of productivity. Chapter 3 reviews the basic concepts underlying the measures of productivity used by government agencies and by private researchers. It extends the nontechnical exposition of these concepts as presented in Chapter 2 by introducing the concept of a production function and by deriving some of the basic statistical formulas used in productivity measurement. Chapter 4 reviews the present official measures of productivity. This review is intended to be factual and descriptive and to serve as background to the rest of the report. Some problems raised by existing measures are mentioned, but the main body of evaluation and criticism is reserved for later chapters. Chapters 5 and 6 are in a sense the heart of the report: they contain most of our discussion of shortcomings in existing measures and possible ways of improving them. Chapter 5 discusses the measurement of output, the numerator of the productivity ratio. The first part of Chapter 5 establishes criteria for the inclusion or exclusion of specific types of activity in the measure of output used in measuring productivity; the second part discusses problems presented by changes in the quality of output. 5 Chapter 6 discusses the measurement of inputs (labor, capital services, and intermediate inputs), one or more of which make up the denominator of the productivity ratio. The problems in measuring labor input include the appropriate concept of hours of work and whether labor input should be differentiated by some measure of its composition, such as skill, experience, or education. The problems of measuring the input of capital services include the treatment of depreciation, obsolescence, and changes . . . In Inventories. Chapters 7 through 9 deal with three important special aspects of productivity measures: the ways in which measures can be used to make inferences about the possible sources of changes in productivity (Chapter 7~; the measurement of productivity at the level of the individual plant or company (Chapter 8~; and international comparisons of productivity (Chapter 9~. Throughout the report, we have made recommendations where we thought them appropriate. These recommendations and our major findings comprise the rest of this chapter. Most of our recommendations are addressed to the agencies of the federal government that prepare productivity measures or supply the data that enter into them. Some of these recommendations are intended to improve existing productivity measures; others would broaden their scope.
6 FINDINGS AND RECOMMENDATIONS REPORT OF THE PANEL This section contains a summary of the principal findings and all the recommendations of the Panel. Each recommendation is followed by a reference to the chapter or chapters of the report and, in some cases, a paper in this volume, that contain a fuller explanation of the findings and the evidence. The order of the findings and recommendations follows the order of the chapters from which they are drawn. Therefore, the recommendations are not listed in order of priority. Although there have not been any explicit dissents from any of the recommendations, it should not be assumed that every member of the Panel agrees with every statement in the report or with every recommendation. A supplementary statement to the Panel's report by Markley Roberts follows Chapter 9. The Panel has not attempted to estimate the cost of implementing its recommendations, both because this task was not included in its charge and because it does not have access to the information needed to do so. In thinking about the possible costs of implementing the recommendations, the reader should keep in mind that almost all of the recommendations would have benefits broader than improving productivity measures and would also help to improve other parts of the federal statistical system. THE PRESENTATION AND INTERPRETATION OF PRODUCTIVITY MEASURES Changes in the Bureau of Labor Statistics (BES) measures of output per hour of labor input are subject to misinterpretation by users, who may incorrectly associate changes in these measures solely with the changing skill and effort of the work force. Output per hour is also affected by changes in technology, the capital/labor ratio, changes in inputs of intermediate goods, rates of capacity utilization, and many other factors as well as by the interaction of these factors. The BES now publishes cautionary statements warning users not to attribute changes in output per hour to labor alone, but these statements appear only in the annual bulletin on productivity for detailed industries and in selected articles in the Monthly Labor Review. Such statements do not appear in the quarterly releases on productivity measures for broad sectors of the economy, which are the measures most closely followed by the press.
Introduction, Findings, and Recommendations Recommendation 1. The Panel recommends that the Bureau of Labor Statistics give more prominence in its publications and press releases to cautionary statements warning against misinterpretation of its output-per-hour measures. (Chapter 2) 7 Output-per-hour measures are constructed from many data sources, some of which become available only after preliminary, and even later, productivity measures have been published. The incorporation of addition- al data from these sources is the reason that published estimates of output per hour are revised. Occasionally, these revisions are large, as is shown in De Sha (in this volume). To give users some idea of the reliability of early measures of output per hour, it may be useful to provide them with an estimate of the possible size of expected revisions on the basis of past experience. Recommendation 2. The Panel recommends that the Bureau of Labor Statistics study the size, direction, and other characteristics of past revisions in estimates of output per hour and, on the basis of its findings, consider publishing in its press releases a range of probable revisions (based on historical experience) for the preliminary esti- mates. (Chapter 4 and De Sha) Even after the final revisions are made, incorporating the most reliable and comprehensive sources of information available, a certain margin of error remains in the estimates of output per hour. This error may be due to incomplete coverage of data sources, inaccurate reporting, sampling error, and the like. Existing techniques for estimating the distribution of the overall margin of error rely heavily on the opinions of experts those who prepare or study the estimates-concerning the distribution of the errors in various components of the index. The feasibility of using these methods is examined in Holland and King (in this volume). Further study of the overall error in productivity measures and the sensitivity of overall error to reductions of error in components of the measures will aid in setting priorities for improving the measurement of those components. Recommendation 3. The Panel recommends that the Bureau of Labor Statistics and the Bureau of Economic Analysis explore methods for estimating the implications of error reduction in component measures for the reduction of overall error in productivity measures beyond that corrected by routine revisions. (Chapter 4 and Holland and King)
8 MEASUREMENT OF OUTPUT REPORT OF THE PANEL There are two important criteria for measuring output for use in productivity ratios: data on output and inputs should be taken from the same universe, and output and inputs should be measured independently of each other. The scope of the output measure used in the current official measure of productivity in the private business sector, for the most part, satisfies these criteria. The aggregate measure excludes some large and growing sectors for which there is as yet no satisfactory way to measure output independently of inputs, such as general government and not-for- profit institutions. However, it includes some sectors for which estimates of the change in real output are estimated either entirely or mainly from changes in inputs, including finance and insurance and portions of the services and construction sectors. About 5 percent of the current official productivity output universe does not meet the second criterion. It would be valuable, though difficult, to develop satisfactory ways to measure output directly in these questionable areas. A major problem in measuring output stems from changes in quality. Real output is usually measured by deflating the current-dollar value of output by a price index. To be an appropriate deflator, this index must distinguish changes in prices that correspond to changes in quality from true price changes. Current techniques make that distinction when the changes in quality result from changes in costs of production, but those techniques are not adequate for dealing with changes in quality that do not result from changes in costs. Recommendation 4. The Panel recommends that the Bureau of Economic Analysis (BEA) and the Bureau of Labor Statistics (BES) seek to improve their existing price indexes and to develop auxiliary measures of price change. These new auxiliary measures should take into account more adequately the types of quality change that are not now measured. They should be a collaborative effort of BEA and the productivity and price research divisions of BES. Among the adjustments that could be incorporated in the new measures are adjustments (through better linking procedures) for changes in value to users resulting from the introduction of improved products; estimates of the value to users of improvements in performance that are achieved without increases in real costs; and estimates of the present value of future savings in operating efficiency made possible by design changes and improvements. Since many of the proposed adjustments would require background studies and additional research, they could not all be incorporated
Introduction, Findings, and Recommendations simultaneously into a single new output measure to be used for productivity analysis. Instead, the output measure could be revised periodically, perhaps every 5 years, as new research evidence becomes available on the importance of unmeasured quality change both in the current and in the previous periods. (Chapter 5) 9 Observers have criticized the output measure used in existing productiv- ity measures from yet another point of view, saying that it attempts to measure only changes in welfare that operate through the goods and services sold in the market and therefore that it does not measure many important changes in welfare. Some of these changes are directly related to the economic system; for example, the costs of relocating workers laid off because of technological change and the environmental damage done by industrial pollution. Others, like changes in health and safety, while less closely related to the economic system, are clearly of great importance. The Panel concludes that at present the scope of official productivity measures should be restricted to sectors in which changes in real output can be adequately measured. In addition, the Panel thinks that it is important to develop separate output measures for sectors now excluded from the private business sector or from gross national product (GNP). The largest GNP sector excluded from productivity measures is general government. For the federal government, BES has developed productivity measures that cover about 65 percent of the federal civilian work force; these measures are not used to expand the output measure for the private business sector, however, because of their limited coverage and the difficulty of defining the outputs of federal agencies. For state and local governments, private researchers have developed measures of output and productivity for some of the more readily identifiable services, such as police protection and education. Another major GNP sector excluded from the broadest productivity measures is not-for-profit institutions, principally private not-for-profit hospitals and colleges and universities. It may be possible to measure outcomes and productivity for some of these institutions. Scott (in this volume) finds that the use of more inputs by hospitals leads to higher- quality outcomes in some dimensions of their output, such as reduction in postsurgical mortality rates. However, other organizations, such as schools and churches, that lack empirically demonstrable techniques of production use instead what may be termed socially validated techniques. A major area of non-market activities not covered by GNP is home production, which includes child care, meal preparation, and home repairs by owners. The BEA, which prepares GNP estimates, is now beginning to
10 REPORT OF THE PANEL develop estimates of the value of home production. The BEA iS also seeking to estimate environmental pollution and other external effects. Recommendation 5. The Panel recommends that research on the measurement of the output and productivity of the resources in excluded sectors be expanded. However, there should be no prema- ture selection or foreclosing of any of the alternative measures of output for such systems as health care and higher education. Although progress has been made in measurement in these areas, we do not yet know enough about the operation of such systems either to measure precisely their salient outcomes or to assume that we understand the processes that account for them. (Chapter 4 and Scott) There is growing recognition that the interpretation of productivity measures could be enhanced if more attention were given to quality-of-life issues. Quantitative data from federally funded research by BEA'S Environmental and Nonmarket Division, studies of social indicators, and work on goals accounting are intended to broaden our understanding of many critical aspects of productivity change not fully covered in the existing framework of economic measures. This broadening of understand- ing does not require that data on the quality of life be incorporated into the measures of output and productivity now in use. Recommendation 6. The Panel concludes that many useful analyses of economic and social welfare issues can be undertaken within the framework of output and input used in the current official measures. However, the Panel does agree that for the study of many important social problems for example, improvement of the health status of the population-definitions of output and input that go well beyond those currently used to measure productivity are required. The Panel believes that for each of these special problems, partial and special measures of output and input should be developed when possible. However, the Panel concludes that it is not now possible to add or subtract such special measures from the conventional measures of output and productivity to construct a meaningful single index of overall welfare for the nation. (Chapter 5 and Moss)
Introduction', Findings, and Recommendc~tior~s MEASUREMENT OF INPUTS Labor Input 11 The current official measure of labor input is the unweighted sum of labor- hours in all the enterprises whose output is included in the numerator of the productivity measure. Although this measure is satisfactory for many purposes, it supers from some drawbacks. First, for a substantial fraction of the work force, nonproduction and supervisory workers, there are no direct data on hours. Second, the current official measure of hours of employees uses data on hours paid for rather than hours spent at the workplace to measure changes in labor input. Hours paid for include paid leave as well as hours spent at the workplace, and the ratio of hours at work to hours paid for has been gradually falling. Finally, the official measure, since it is an unweighted sum of hours, does not take into account the effect on measured productivity of changes in labor composi- tion, such as changes in skill, education, and experience. Recommendation 7. The Panel recommends that the Bureau of Labor Statistics develop an annual supplement to the current employment statistics (CES) program requesting data on the actual hours of nonproduction and supervisory workers. (Chapter 6) Recommendation 8. The Panel concurs with the recommendation of a 1976 BES task force that recommended that a separate annual survey of hours at the workplace be administered to a subsample of the establishments currently responding to the CES. (Chapter 6) Recommendation 9. The Panel recommends that the Bureau of Labor Statistics devote more resources to studying the use of weighted labor input measures. The purpose of this effort would be the preparation of one or more weighted measures of labor input for broad aggregates of economic activity, such as the private business sector. Such weighted labor input measures could be used alone and could also be used in combination with measures of capital input to measure changes in multi-factor productivity. Both of these measures could help in explaining the changes in the single-factor unweighted measure of labor productivity. (Chapter 6 and Kunze)
12 Capital Input REPORT OF THE PANEL There are now no official published measures of single-factor productivity other than those for labor input. Single-factor measures of inputs of capital services are needed to improve our understanding of the causes of productivity change in the economy. Capital stock estimates, which are needed to derive estimates of inputs of capital services, are already prepared and published by the BEA, using the perpetual inventory method. These require data on investment expenditures (including inventories), capital goods prices, and estimates of the length of life of capital goods. These perpetual inventory estimates are not currently reconciled with book-value data and are not available jointly cross-classified by functional group (e.g., buildings, trucks, etc.) and industry. Recommendations have been made in the reports of the Wealth Inventory Planning Study and the GNP Data Improvement Project of ways to improve capital stock and input estimates. On the basis of these reports, the Panel endorses several recommendations that will substantially improve capital estimates to be used in productivity measures. Recommendation 10. The Panel recommends that the Census Bu reau, in its quinquennial economic censuses and annual economic surveys, collect data on beginning and end-of-year inventories by method of valuation. The censuses and surveys should also provide fuller coverage and greater detail on capital outlays, book values of . invested capital (gross and net of depreciation), and on depreciation allowances and retirements. The data should distinguish major types of equipment by major industry. The Panel also recommends that a sample survey of the age distribution of fixed assets, by type, and of retirements, by age, during the previous year be taken in order to keep current the retirement factors used in estimating capital stocks. (Chapter 6) Recommendation 11. The Panel recommends that government agen- cies make use of available estimates of real capital stocks to develop ratios of output per unit of capital in order to determine the savings that have been achieved over time in physical capital per unit of output. (Chapter 6)
Introduction, Findings, and Recommendations Intermediate Inputs 13 Intermediate inputs are purchases of goods and services by one firm or industry from another that are used up in current production. Productivity changes at the individual firm or industry level can be due to increases in efficiency in the use of intermediate inputs as well as in the use of labor and capital. Good measures of changes in both the quantity and quality of intermediate inputs are needed to identify the industrial sources of productivity change. If measures of intermediate inputs are not available, measurement of industry productivity must rely on total output per unit of labor and capital, which can reflect both productivity change and substitution between intermediate inputs and labor and capital. For most nonmanfacturing industries, current data gaps preclude the direct measurement of intermediate inputs (see Chapter 4~. For the industries in which intermediate inputs are measured directly (e.g., manufacturing), deflators are constructed from information on the product mix of purchased inputs from input-output tables that may be outdated. This problem is important because the mix and amount of intermediate inputs used per unit of output can change sharply in an economy subject to rapid technological change. Recommendation 12. The Panel recommends that the Census Bureau in its periodic reports on real gross output for detailed industries based on successive quinquennial censuses, include estimates of the real amount of intermediate purchases of materials, including energy, for those detailed industry categories for which data are available. (Chapters 4, 5, and 6 and Myers and Nakamura) Recommendation 13. The Panel recommends that, in order to improve the quality of the annual measures of industry productivity, the Census Bureau augment the Annual Survey of Manufactures (and other annual surveys) to include subcategories of the materials input categories. (Chapters 4 and 6 and Myers and Nakamura) Recommendation 14. The Panel endorses the recommendation of the GNP Data Improvement Project calling for the Census Bureau to collect, as an integral part of each economic census, data on the purchases of intermediate services as well as materials by establish- ments. (Chapters 4 and 6 and Myers and Nakamura)
14 Multi-Factor Productivity Measures REPORT OF THE PANEL By combining two or more measures of inputs (capital, labor, and intermediate purchases) by some appropriate weighting scheme and dividing this combined measure of inputs into a measure of output, one can obtain a multi-factor productivity ratio. Changes over time in this ratio measure the net savings in the use of inputs per unit of output that are brought about by technical change and other sources of productivity growth. Multi-factor productivity measures are useful in a number of ways. They help in understanding changes in output per unweighted labor-hour. If the growth rate in the multi-factor measure is less than that of the single-factor labor-hour measure, we know that some of the increase in the labor-hour measure is due to increases in the amount of other inputs per unweighted labor-hour. Multi-factor measures are also useful in monitoring and studying the sources of output growth other than those from inputs including technical change, managerial innovations, and improved resource alloca- tion. Changes in the growth rates of multi-factor productivity measures are due entirely to these non-input sources. Recommendation 15. The Panel recommends that the Bureau of Labor Statistics experiment with combining labor and other inputs into alternative measures of multi-factor productivity. (Chapter 6) MEASUREMENT OF THE SOURCES OF PRODUCTIVITY CHANGE Existing research projects on sources of productivity growth, which are almost all done by private researchers, take a long time to complete. If they were done quickly, they would be unreliable. Thus, there is often a long lag between the occurrence of important changes in the official productivity measures (output per unweighted labor-hour) and the appearance of well-documented analyses of these changes, even in terms of forces that are relatively easy to measure, such as changes in the capital/labor ratio and various dimensions of work-force composition.
Introduction, Findings, and Recommendations Recommendation 16. The Panel recommends that the Bureau of Labor Statistics and the Bureau of Economic Analysis take joint responsibility for developing and maintaining measures of some of the sources of growth (such as changes in physical capital and work-force composition) so that policy makers can have timely and accurate information on at least the more easily measurable sources of productivity change. (Chapter 7) 15 Although the existing research on the sources of productivity change provides useful knowledge about the broad patterns of contributions to growth, it tells us relatively little about its more detailed aspects and patterns. How do the various inputs and the non-input sources interact with one another (for example, the rate of capital investment and the rate of embodied technical change)? What is the relative importance of new product and service innovation, whose output ejects are hard to measure, in total innovation? How representative is the existing evidence on the social rate of return to investments in research and development? How do institutional (legal, political, social, cultural) factors affect productivity change in various industries and in the economy? Recommendation 17. The Panel recommends that government agen- cies support research aimed at improving knowledge about the sources of productivity change. These agencies should be especially attentive to research that focuses on measuring technical and organizational change and new product and service innovation. The Panel also recommends funding micro-level studies of the returns to research and development. Micro-level studies of innovations in personnel management and other organizational behavior should also be encouraged. (Chapter 7) INTERAGENCY COOPERATION AND COORDINATION Several federal agencies collect basic economic data from establishments on the value of shipments, hours of employees, and prices of commodities and services sold. These basic data should be comparable because they are combined into secondary measures of output, hours, and productivity for industries and larger aggregates. Since by law the Census Bureau cannot share its Standard Statistical Establishment List a directory of all establishments by industry the basic data are taken by samples from different universes and are not strictly comparable.
16 REPORT OF THE PANEL Recommendation 18. The Panel supports legislation that would allow the Census Bureau to share with other federal statistical collection agencies the Standard Statistical Establishment List, so that all those agencies could sample from a common universe, making the basic economic data more comparable. (Chapter 8) Either because of legal restrictions or because of provisions regarding confidentiality for companies providing data voluntarily, the federal statistical agencies seldom interchange data. Unfortunately, the result of this lack of coordination is duplication of data collection and the underutilization of existing information within the federal government. Recommendation 19. The Panel stresses the urgency of finding a solution to the problem of coordinating data collection and allowing data interchange among the federal statistical collection agencies for statistical and research purposes, in such a manner that the rights, benefits, or privileges of individual respondents are not violated. (Chapter 8) Several agencies of the federal government prepare estimates of output for identical sectors of the economy. These estimates sometimes diverge widely because the agencies use different concepts and data sources or different procedures for deflation and weighting. For recent benchmark years, substantial differences were found in the four estimates of output in the pulp and paper industries prepared by BEA, the Federal Reserve Board (FRB), BES, and the Census Bureau (see Myers and Nakamura in this volume) and between the indexes of output of manufacturing industries prepared by FRB and BEA (see Popkin in this volume). Reconciling differences between output measures may shed further light on measurement problems and on observed changes in productivity. The agencies involved should therefore try to determine the sources of differences in their estimates. Recommendation 2a The Panel recommends that the relevant agencies try to reconcile their different output measures that cover the same industry or sector to improve the measures and to acquire a better understanding of measurement problems associated with weighting, deflation, and other procedures. This can be achieved by strengthening the existing mechanisms in government that bring the agencies together, such as committees formed by the Once of Federal Statistical Policy and Standards. (Chapter 8 and Myers and Naka- mura)
Introduction, Findings, and Recommendations INTER-FIRM COMPARISONS 17 Many companies in the United States measure their productivity and use these measures for finding ways to reduce costs, to plan future production, and to keep track of their long-run performance. In many other countries, companies participate in industry-wide productivity comparison programs that permit them to assess their performance vis-a-vis their competitors. Productivity measures add a new dimension to the financial information that managers typically rely on. Recommendation 21. The Panel recommends that companies investi- gate whether having measures of productivity would improve their performance, planning, and evaluation. It encourages the U.S. Department of Labor and the U.S. Department of Commerce to continue to inform companies of the potential benefits of productivity measurement programs. (Chapter 8) INTERNATIONAL COMPARISONS OF OUTPUT AND PRODUCTIVITY Reliable comparisons of national income and product levels in different countries are a prerequisite for international comparisons of productivity levels. The widely published comparisons of output of different countries obtained by converting the nominal value of output (expressed in yen, marks, dollars, etc.) by use of exchange rates are virtually meaningless. Meaningful comparisons require comparing the prices of similar goods and services in all countries for a given year. The International Comparisons Project (~cP) has made such comparisons for several industrialized and developing nations for 1970 and 1973. This project is particularly successful because of its association with the United Nations and its contact with the national governments concerned. Recommendation 22. The Panel recommends continued and increas- ing support for the work of the International Comparisons Project, along the lines suggested by the Statistical Office of the United Nations. We recommend that international organizations provide the financial support necessary to implement the proposals and urge U.S. representatives to these organizations to support the proposals. (Chapter 9) The ice data do not permit comparisons of output or productivity at the detailed industrial level. There have been some international productivity comparisons by industry, but they are limited in the number of industries
18 REPORT OF THE PANEL or countries covered. The next logical step in expansion of the ice is industry estimates for industrial nations, for which these comparisons are most meaningful. Perhaps the ice will soon be able to expand its research in this direction. In this country the analysis of international productivity differences has been largely the work of private researchers supported by federal agencies. Direct research by the agencies has been confined in the postwar period to rather simple compilations of series on output per worker, or per hour, for manufacturing, farming, and steel. Explanation of international productivity differences also requires study of institutional factors, including social, cultural, and political factors, as well as economic conditions. This could become an important field of research for researchers in many disciplines. Recommendation 23. The Panel endorses continued support for private research on international productivity comparisons, with some enlargement of its scale as opportunities for useful projects arise. The Panel also believes that research within government should be expanded along lines to be determined by the relevant agencies and adjusted on the basis of their experience. As an important component, we suggest, once the Bureau of Labor Statistics and the Bureau of Economic Analysis have established a program to provide growth accounting series on a regular basis for the United States, they extend their series to other industrial countries. (Chapter 9)