Productivity in the U.S. Services Sector

Productivity in the U.S. Services Sector: New Sources of Economic Growth

Jack E. Triplett
Barry P. Bosworth
Copyright Date: 2004
Pages: 401
  • Cite this Item
  • Book Info
    Productivity in the U.S. Services Sector
    Book Description:

    The services industries -which include jobs ranging from flipping hamburgers to providing investment advice -can no longer be characterized, as they have in the past, as a stagnant sector marked by low productivity growth. They have emerged as one of the most dynamic and innovative segments of the U.S. economy, now accounting for more than three-quarters of gross domestic product. During the 1990s, 19 million additional jobs were created in this sector, while growth was stagnant in the goods-producing sector. Here, Jack Triplett and Barry Bosworth analyze services sector productivity, demonstrating that fundamental changes have taken place in this sector of the U.S. economy. They show that growth in the services industries fueled the post-1995 expansion in the U.S. productivity and assess the role of information technology in transforming and accelerating services productivity. In addition to their findings for the services sector as a whole, they include separate chapters for a diverse range of industries within the sector, including transportation and communications, wholesale and retail trade, and finance and insurance. The authors also examine productivity measurement issues, chiefly statistical methods for measuring services industry output. They highlight the importance of making improvements within the U.S. statistical system to provide the more accurate and relevant measures essential for analyzing productivity and economic growth.

    eISBN: 978-0-8157-9663-3
    Subjects: Business, Political Science

Table of Contents

  1. Front Matter
    (pp. i-iv)
  2. Table of Contents
    (pp. v-vi)
  3. Foreword
    (pp. vii-viii)
    Strobe Talbott

    After many years of slow growth, the U.S. economy has experienced a remarkable improvement in labor productivity that began in the mid-1990s but continued through the 2001–02 recession and has now lasted for nearly a decade. This book undertakes a detailed assessment of the resurgence of productivity growth. It demonstrates the dominant role played by the services producing industries in recent U.S. productivity performance. This is a great turnaround from the past view of services industries. In the old view of services, they were stagnant and unprogressive sectors that slowed overall economic growth. In the new view presented in...

  4. Acknowledgments
    (pp. ix-xii)
  5. CHAPTER ONE Introduction
    (pp. 1-5)

    The United States has a services-based economy. Over the past half-century, the share of the nation’s output accounted for by goods-producing industries has fallen by nearly half, and the services-producing industries now account for more than three-quarters of GDP and a comparable proportion of total employment.

    Yet services industries have long been disparaged as sources of low-skill, low wage jobs, and they often are characterized as part of a stagnant sector marked by low productivity growth and only limited opportunities for innovation. Some economists have contended that, over the long term, the need to devote an ever-increasing share of employment...

  6. CHAPTER TWO Overview: Industry Productivity Trends
    (pp. 6-40)

    In this chapter we report measures of labor and multifactor productivity (MFP) for U.S. industries at roughly the two-digit SIC (Standard Industrial Classification) level. The productivity estimates are based on gross output, using a combination of government databases from the Bureau of Economic Analysis (BEA) and the Bureau of Labor Statistics (BLS)—and implicitly the Census Bureau, since the other two agencies’ compilations rest heavily on data originally collected by the Census Bureau’s “economic directorate.” Our database is a relatively new one that has not previously been employed for productivity research. With the new database, we can compare productivity trends...

  7. Discussion of Productivity Trends and Measurement Issues in Services Industries
    (pp. 41-45)

    The last decade has been an exciting time for productivity analysts. After more than twenty years of relatively slow productivity growth, the U.S. economy suddenly reversed course in 1995 and productivity accelerated substantially. This allowed economists to turn from the decidedly dismal task of explaining the productivity slowdown to the much more appealing job of explaining the productivity revival. Given the issue’s importance, potential answers poured in: fundamental technological progress in the production of information technology, capital deepening related to information technology use, cyclical effects, changes to workplace practices and firm organization, and competitive forces driven by deregulation and globalization...

  8. CHAPTER THREE Output and Productivity in the Transportation Sector
    (pp. 46-70)

    The transportation sector has been of considerable interest in productivity research. Major portions of the sector were long subject to regulation, which resulted in the collection of a wealth of data on their operations. More recently, much of the industry has been deregulated and economists have been very interested in the impact of that change on the efficiency of its three major components—rail, truck, and air transport.¹ Here we examine the nature of productivity change in the transportation industry as revealed by the existing data. What are the patterns of change in labor productivity and multifactor productivity? What can...

  9. CHAPTER FOUR Output and Productivity Growth in the Communications Industry
    (pp. 71-86)

    Since the historic court ruling in 1984 that opened the long-distance telephone market, long dominated by AT& T, to competition, the communications industry has experienced dramatic economic and technological change. The Telecommunications Act of 1996 went further, promoting competition in all telecommunications sectors, including the provision of local telecommunications services. Technological and economic changes have included the introduction of fiber-optic cable, the expansion of cable services from 10 million subscribers in 1975 to 68 million in 2000, the surge in cellular phone subscribers from only 5 million in 1990 to 128 million in 2001, and regular access to the Internet...

  10. Discussion of the Transportation and Communications Industries
    (pp. 87-94)

    The Triplett-Bosworth chapters on transportation and communications are admirable. The authors do the best job to date of untangling what the BEA and BLS data actually show and why they differ, and they bring together evidence on the reasons for acceleration and deceleration of productivity growth at the sectoral level. Moreover, their book is impressive in recognizing past research and linking their results to this inheritance. The main thrust of these comments is to praise the progress that the authors have made in reconciling measurement differences but to criticize them on two quite different grounds. First, they display output and...

  11. CHAPTER FIVE Overview: Productivity and Measurement in the Finance and Insurance Sector
    (pp. 95-122)

    In some services industries, the concept of real output is unclear. When it is difficult to measure the output of an industry, it is also difficult to measure its price change and productivity. The finance and insurance sector is filled with those difficult-to-measure industries.

    For goods, the concept of output is usually straightforward. What the industry produces—an automobile or a computer, for example—is usually obvious and implies the unit in which output is measured. Of course, measuring the output of automobiles or computers involves much more than just counting the units, because in both cases quality change is...

  12. CHAPTER SIX Price, Output, and Productivity of Insurance: Conceptual Issues
    (pp. 123-176)

    As noted in chapter 5, the insurance industries have not experienced high productivity growth rates, at least as measured by using the data in the Bureau of Economic Analysis industry accounts. Insurance carriers appear to have had negative labor productivity growth over the whole 1987–2001 period (table 5-4), and both the insurance carrier and the insurance agent industries had negative measured multifactor productivity (MFP) for most of that interval.

    In April 1998, the Brookings Program on Output and Productivity Measurement in the Services Sector convened a workshop, “Measuring the Price and Output of Insurance,” that included participants from both...

  13. CHAPTER SEVEN Measuring Banking and Finance: Conceptual Issues
    (pp. 177-210)

    In contrast to insurance, where measured productivity growth is low, labor productivity and multifactor productivity in banking and financial institutions are relatively high—for example, using the BEA-BLS industry file, labor productivity in banking grew 3.1 percent a year after 1995, and banking MFP grew 1.5 percent a year (table 5-4). These rates exceed recent economywide productivity growth rates. The nondepository institutions industry has a bit slower labor productivity growth but faster MFP (1.9 percent and 2.1 percent, respectively). As currently measured in the BEA industry accounts, the banking and financial services industries seem to show healthy performance.

    Despite the...

  14. Discussion of Banking Output
    (pp. 211-216)

    The current methodology the Bureau of Economic Analysis uses to approximate the value of banking output provides a testable hypothesis, which is outlined here. Independent of any test results, BEA would be well advised to try to determine the robustness of alternative indicators of the value of banking output. One possibility would be to use the Bureau of Labor Statistics transactions-based banking quantity index of loan, deposit, and trust services and multiply that output quantity indicator by a corresponding price-cost index of the same banking services.

    To illustrate the usefulness of the suggested transactions-based output quantity approach, I briefly note...

  15. Discussion of Output Measurement in the Insurance and the Banking and Finance Industries
    (pp. 217-229)

    These chapters provide a comprehensive discussion of the Brookings workshops on measuring financial service output, in particular the output of the insurance industry and the banking industry. Jack Triplett should be commended for providing the detail of the discussion. I also want to thank Jack for the opportunity to comment on the chapters. The output of each of the two industries is among the most difficult to measure and certainly the subject of much debate. I will comment on each industry separately.

    Though the chapter raises many issues regarding the measurement of the output of the insurance industry, I will...

  16. Reply to Dennis Fixler
    (pp. 230-232)

    There are only a few differences between our position and Fixler’s, though there are miscommunications and misunderstandings. As only one example of the latter, we did not say that “a margin approach . . . is the proper measure of bank output”; we said that an appropriate subject for debate is whether gross margin or output is the proper measure for productivity analysis in banking as well as in insurance and wholesale/retail trade. We can only concede that in a number of places our writing apparently was not clear enough.

    The main issue, as we see it, is determining which...

  17. CHAPTER EIGHT Output and Productivity in Retail Trade
    (pp. 233-250)

    The retail trade sector has been the largest single contributor to the post-1995 resurgence of growth in productivity. This is a surprise to many—particularly those who identify productivity growth with the emergence of the newer, more glamorous high-tech economy. Retail trade is among the oldest of industries, and, at least in the United States, it has concentrated increasingly on employing the least-educated and least-skilled workers. Between 1975 and 2000, average hourly earnings in retail trade declined from 88 to 78 percent of the private sector average, while retail trade’s share of total employment fluctuated in the range of 16...

  18. Discussion of Output and Productivity in Retail Trade
    (pp. 251-255)

    Triplett and Bosworth provide an overview of major issues affecting productivity and its measurement in retail trade and e-commerce, paying special attention to answering the following question: Are the measured productivity gains in these areas real or an artifact of the method used to measure them? The chapter provides an excellent summary of the key issues affecting the measurement of retail productivity; it also provides a survey of what is known about productivity change in specific retail sectors and about the impact of information technology and e-commerce on retail productivity. I concentrate here on issues related to the measurement of...

  19. CHAPTER NINE Output and Productivity in Other Services
    (pp. 256-273)

    Measurement problems in the services industries are not exhausted by the ones we treated in previous chapters. We singled out certain industries for extended discussion partly because the research literature deals with them more extensively and partly because in a number of them (finance, insurance, and retail trade) perplexing conceptual problems need to be addressed in order to improve productivity measures and our understanding of productivity change.

    Business services, medical care, and education are three services industries that also were covered in the Brookings economic measurement workshops, and they too are large industries that present severe problems for measuring output...

  20. CHAPTER TEN High-Tech Capital Equipment: Inputs to Services Industries
    (pp. 274-320)

    We estimate that services industry labor productivity advanced at the trend rate of 2.41 percent a year since 1995. We also estimate that roughly a quarter of the services industry labor productivity growth (0.59 percentage point) came from investment in information and communications technology capital (table 10-1: these estimates were taken from tables 2-1 through 2-7 of chapter 2 and appendix table A-1). In some services industries, the IT contribution was much larger—for example, in communications IT contributed 1.3 percentage points, but out of a very rapid 6.7 labor productivity advance (see appendix table A-1).

    Estimating sources of labor...

  21. CHAPTER ELEVEN Data Needs
    (pp. 321-338)

    The U.S. statistical system has made vast strides in recent years in improving the data available for the analysis of productivity and particularly for estimating productivity in the services industries. These improvements have not been widely noticed. The title of this chapter shows that we have no doubt that measurement problems remain, and we use our research to demonstrate where resources for improvement can be allocated. But we do not want our list of improvements to obscure the fact that the situation today is far better than it was when Baily and Gordon (1988) reviewed consistency of industry data for...

  22. APPENDIX A Industry Productivity Accounts, 1987–2001
    (pp. 339-364)
  23. APPENDIX B Workshops in the Brookings Program on Output and Productivity Measurement in the Services Sector
    (pp. 365-374)
  24. References
    (pp. 375-388)
  25. Index
    (pp. 389-402)
  26. Back Matter
    (pp. 403-405)