信息技术与美国生产率增长

Information technology and U.S. productivity growth: evidence from a prototype industry production account Dale W. Jorgenson • Mun S. Ho • Jon D. Samuels � Springer Science+Business Media, LLC 2011 Abstract The rapid productivity growth in the US during the Information Age, prior to the dot-com bust in 2000, and the large contribution of the IT producing sector, is well known. Less known are the sources of the surprisingly rapid TFP growth during the slow growth period after 2000. We construct an account of US economic growth by aggregating over detailed industries using a new data set based on the NAICS classification. We find that, post 2000, TFP originating from the IT-Producing sector decelerated relative to the IT boom, but still accounted for 40% of aggregate productivity growth. This deceleration was counterbalanced by the contribution from IT-Using sectors, which buoyed aggregate TFP growth to almost the same rate as the 1995–2000 period. For aggregate GDP, the contributions to the growth rate of 2.8% during 2000–2007 were: capital input (1.7% points), labor input (0.4) and TFP (0.7). Keywords Total factor productivity � NAICS � Growth accounting � Information technology JEL Classification O47 � D24 1 Introduction The computer equipment manufacturing industry com- prised only 0.3% of US value added from 1960 to 2007, but generated 2.7% of economic growth and 25% of produc- tivity growth. By comparison, agriculture accounted for 1.8% of US value added, but only 1.0% of economic growth during this period. This reflects the fact that agriculture has grown more slowly than the US economy, while the com- puter industry has grown thirteen times as fast. However, agriculture accounted for 15% of US productivity growth, indicating a very significant role for agricultural innovation. The great preponderance of economic growth in the US involves the replication of existing technologies through investment in equipment and software and expansion of the labor force. Replication generates economic growth with no increase in productivity. Productivity growth is the key eco- nomic indicator of innovation. This innovation accounts for less than 12% of US economic growth, despite its importance in industries like computers and agriculture. Although inno- vation contributes only a modest portion of growth, this is vital to long-term gains in the American standard of living. The predominant role of replication of existing tech- nologies in US economic growth is crucial to the formu- lation of economic policy. As the US economy recovers from the Great Recession of 2007–2009, economic policy must focus on maintaining the growth of employment and reviving investment. Policies that concentrate on enhancing the rate of innovation will have a very modest impact over the intermediate term of 10 years. However, the long-run growth of the economy depends critically on the perfor- mance of a relatively small number of sectors, such as agriculture and computers, where innovation takes place. The purpose of this paper is to present a new data set on US productivity growth by industry. This data set covers 70 D. W. Jorgenson Harvard University, Cambridge, USA M. S. Ho (&) Resources for the Future, Washington, USA e-mail: ho@rff.org J. D. Samuels Johns Hopkins University, Baltimore, USA 123 J Prod Anal DOI 10.1007/s11123-011-0229-z industries for the period 1960–2007 and uses the North American Industry Classification System (NAICS). Previ- ous industry-level data sets on US productivity provided by Jorgenson et al. (1987) and Jorgenson et al. (2005) have used the Standard Industrial Classification (SIC). The US statistical system has shifted gradually to NAICS, begin- ning with the Business Census of 1997. The national accounts converted to NAICS in the 2003 Comprehensive Revision of the National Income and Product Accounts. An important advantage of NAICS over the SIC is the greater detail available on the service industries that make up a growing proportion of the US economy. Jorgenson et al. (2007) have shown that US productivity growth has been concentrated in the service industries since 2000, especially those that make intensive use of information technology. NAICS also provides more detail on industries that produce information technology hardware, software, and services. The IT-service-producing industries, infor- mation and data processing services and computer systems design and related services, are growing in importance, relative to software and the IT hardware manufacturing industries—computer and peripheral equipment, commu- nications equipment, and semiconductor and other elec- tronic components. This paper begins with a brief summary of the method- ology for productivity measurement in Sect. 2. The tradi- tional approach of Kuznets (1971) and Solow (1970) has been superseded by the new framework presented in Schreyer’s OECD (2001) manual, Measuring Productivity. The focus of productivity measurement has shifted from the economy as a whole to individual industries like agriculture and computers. The OECD productivity manual has estab- lished international standards for economy-wide and indus- try-level productivity measurement. This focus of measuring productivity at the industry level is summarised in Sect. 3. The OECD standards are based on the production accounts constructed by Jorgenson et al. (1987). These accounts were updated and revised to incorporate invest- ments in information technology hardware and software by Jorgenson et al. (2005). The EU KLEMS (capital, labor, energy, materials, and services) study, described by O’Mahony and Timmer (2009), was completed on June 30, 2008. This landmark study presents productivity measure- ments for 25 of the 27 EU members, as well as Australia, Canada, Japan, and Korea, and the US, based on the methodology of Jorgenson et al. (2005). Current data for the participating countries are available at the EU KLEMS website: http://www.euklems.net/. The hallmark of the new framework for productivity measurement is the concept of capital services, including the services provided by IT equipment and software which is dealt with in Sect. 4. Modern information technology is based on semiconductor technology used in computers and telecommunications equipment. The economics of infor- mation technology begins with the staggering rates of decline in the prices of IT equipment used for information and computing. The ‘‘killer application’’ of the new frame- work for productivity measurement is the impact of invest- ment in IT equipment and software on economic growth. Research on the impact of this investment is summarised by Jorgenson (2009a) in The Economics of Productivity. Jorgenson et al. (2007) have traced the American growth resurgence after 1995 to sources within individual indus- tries. They have measured output and productivity for the IT-producing industries and divided the remaining indus- tries between the IT-using industries, those that are partic- ularly intensive in the utilisation of information technology equipment and software, and the Non-IT industries. How- ever, the IT-producing industries were limited to IT hard- ware and software and did not include IT services. Furthermore, the definition of the IT-using industries was based on the intensity of IT capital input, relative to total capital input. Again, the role of the IT service industries was not identified. The final section sums up the paper. 2 The new framework for productivity measurement The most serious challenge to the traditional approach to productivity measurement of Kuznets (1971) and Solow (1970) was mounted by Jorgenson and Griliches (1967) in ‘‘The Explanation of Productivity Change.’’ Jorgenson and Griliches departed radically from the measurement con- ventions of the traditional approach. They replaced Net National Product with GNP as a measure of output and introduced constant quality indexes for both capital and labor inputs. The key idea underlying the constant quality index of labor input was to distinguish among different types of labor inputs. Jorgenson and Griliches combined hours worked for each type into a constant quality index of labor input, using labor compensation per hour as weights in the index number methodology Griliches (1960) had devel- oped for US agriculture. This considerably broadened the concept of substitution employed by Solow (1957). While Solow had modelled substitution between capital and labor inputs, Jorgenson and Griliches extended the concept of substitution to include different types of labor inputs as well. This altered, irrevocably, the allocation of economic growth between substitution and productivity growth. Constant quality indexes of labor input are dis- cussed detail by Jorgenson et al. (1987, Chapters 3 and 8, pp. 69–108 and 261–300), and Jorgenson et al. (2005, Chapter 6, pp. 201–290). Jorgenson and Griliches introduced a constant quality index of capital input by distinguishing among different J Prod Anal 123 types of capital inputs. To combine these capital inputs into a constant quality index, they identified prices of the inputs with rental prices, rather than the asset prices used in measuring capital stock used by Solow and Kuznets. This further broadened the concept of substitution and again altered the allocation of economic growth between substi- tution and productivity growth. Jorgenson and Griliches employed a model of capital as a factor of production introduced by Jorgenson (1963) in ‘‘Capital Theory and Investment Behaviour’’. This made it possible to incorporate differences among depreciation rates on different assets, as well as variations in returns due to the tax treatment of different types of capital income, into the rental prices. Constant quality indexes of capital input are presented by Jorgenson et al. (1987, Chapters 4 and 8, pp. 109–140 and 267–300), and by Jorgenson et al. (2005, Chapter 5, pp. 147–200). Finally, Jorgenson and Griliches replaced the aggregate production function employed by Kuznets and Solow with the production possibility frontier introduced in Jorgenson (1966) in ‘‘The Embodiment Hypothesis’’. This allowed for joint production of consumption and investment goods from capital and labor services. This captures the fact that systems of national accounts distinguish between outputs of consumption, investment, and other goods and services. Each of these is associated with a price deflator specific to the category of output. Jorgenson used the production possibility frontier to generalize Solow’s (1960) concept of embodied technical change, showing that productivity growth could be inter- preted, equivalently, as ‘‘embodied’’ in investment or ‘‘disembodied’’. Jorgenson and Griliches (1967) removed this indeterminacy by introducing constant quality price indexes for investment goods. As a natural extension of Solow’s (1956) one-sector neo-classical model of eco- nomic growth, his 1960 model of embodiment had only a single output and did not allow for the introduction of a separate price index for investment goods. Oulton (2007) demonstrated that Solow’s model of embodied technical change is a special case of Jorgenson’s (1966) model. He also compared the empirical results of Solow’s one-sector model and a two-sector model with outputs of consumption and investment goods. Greenwood and Krussell (2007) employed Solow’s one-sector model, replacing constant quality price indexes for investment goods with ‘‘investment-specific’’ or embodied technical change. The deflator for the single output, consumption, is used to deflate investment, conflicting with national accounting conventions that provide separate deflators for consumption, investment, and other outputs. Jorgenson and Griliches showed that changes in the quality of capital and labor inputs and the quality of investment goods explained most of the Solow residual. They estimated that capital and labor inputs accounted for 85% of growth during the period 1945–1965, while only 15% could be attributed to productivity growth. Changes in labor quality explained 13% of growth, while changes in capital quality another 11%.1 Improvements in the quality of investment goods enhanced the growth of both invest- ment goods output and capital input, but the net contribu- tion was only 2% of growth. 2.1 Official statistics on productivity The final blow to the traditional framework for productivity measurement of Kuznets (1971) and Solow (1970) was administered by the Panel to Review Productivity Statistics of the National Research Council (1979). The Rees Report, Measurement and Interpretation of Productivity, became the cornerstone of a new measurement framework for the official productivity statistics. This was implemented by the Bureau of Labor Statistics (BLS), the US government agency responsible for these statistics. The BLS Office of Productivity and Technology undertook the construction of a production account for the US economy with measures of capital and labor inputs and total factor productivity, renamed multifactor productivity. A detailed history of the BLS productivity measurement program is presented by Dean and Harper (2001). The BLS (1983) framework was based on GNP rather than NNP and included a constant quality index of capital input, dis- placing two of the key conventions of the traditional framework of Kuznets and Solow. However, BLS retained hours worked as a measure of labor input until July 11, 1994, when it released a new total factor productivity measure including a constant quality index of labor input as well (BLS 1993). Meanwhile, BEA (1986) had incorporated a constant quality price index for computers into the national accounts. This index was included in the BLS measure of output, completing the displacement of the traditional framework of economic measurement by the conventions employed by Jorgenson and Griliches (1967). Jorgenson and Landefeld (2006) have developed a new architecture for the US national income and product accounts (NIPAs) that includes prices and quantities of capital services for all productive assets in the US econ- omy, as well as productivity. The incorporation of the price and quantity of capital services into the United Nations’ System of National Accounts 2008 (2009) was approved by 1 See Jorgenson and Griliches (1967), Table IX, p. 272. We also attributed thirteen percent of growth to the relative utilization of capital, measured by energy consumption as a proportion of capacity; however, this is inappropriate at the aggregate level, as Denison (1974), p. 56, pointed out. For additional details, see Jorgenson et al. (1987), especially pp. 179–181. J Prod Anal 123 the United Nations Statistical Commission at its February– March 2007 meeting. Schreyer, then head of national accounts at the OECD, prepared an OECD Manual, Measuring Capital (Schreyer 2009). This provides detailed recommendations on methods for the construction of prices and quantities of capital services. In Chapter 20 of SNA 2008 (U.N. 2009, page 415), esti- mates of capital services are described as follows: ‘‘By associating these estimates with the standard breakdown of value added, the contribution of labor and capital to produc- tion can be portrayed in a form ready for use in the analysis of productivity in a way entirely consistent with the accounts of the System.’’ The measures of capital and labor inputs and productivity in the prototype system of US national accounts presented by Jorgenson and Landefeld (2006) and updated by Jorgenson (2009b) are consistent with the OECD productivity manual, SNA 2008, and the OECD Manual, Measuring Capital. The volume measure of input is a quantity index of capital and labor services, while the volume measure of output is a quantity index of investment and consumption goods. Productivity is the ratio of output to input. The new architecture for the US national accounts was endorsed by the Advisory Committee on Measuring Inno- vation in the Twenty first century Economy to US Secre- tary of Commerce (2008, page 8) Guttierez: The proposed new ‘architecture’ for the NIPAs would consist of a set of income statements, balance sheets, flow of funds statements, and productivity estimates for the entire economy and by sector that are more accurate and internally consistent. The new archi- tecture will make the NIPAs much more relevant to today’s technology-driven and globalising economy and will facilitate the publication of much more detailed and reliable estimates of innovation’s con- tribution to productivity growth. In response to the Advisory Committee’s recommen- dations, BEA and BLS have produced an initial set of total factor productivity estimates integrated with the NIPAs. The results are reported by Harper et al. (2009) and will be updated annually. This is a critical step in implementing the new architecture. Estimates of productivity are essen- tial for projecting the potential growth of the US economy, as demonstrated by Jorgenson et al. (2008). The omission of productivity statistics from the NIPAs and the 1993 SNA has been a serious barrier to assessing potential growth. 3 Measuring productivity at the industry level A complete system of industry-level production accounts for the US economy was constructed by Gollop and Jorgenson (1980) and Jorgenson et al. (1987), using the SIC. The system incorporates a consistent time series of input–output tables and provides the basis for the industry- level production accounts presented by Schreyer’s OECD Productivity Manual (2001). Details on the construction of the time series of input–output tables are presented by Jorgenson et al. (1987, Chapter 5, pp. 149–182) and Jor- genson et al. (2005, Chapter 4, pp. 87–146). The approach to growth accounting presented by Jorgenson et al. (1987) and the official statistics on aggregate productivity published by the BLS in 1994 have been recognised as the international standard. This standard is discussed in Schreyer’s (2001) OECD Manual, Mea- suring Productivity. The expert advisory group for this Manual was chaired by Dean, former Associate Commis- sioner for Productivity at the BLS and a leader of the successful effort to implement the Rees Report (1979). Reflecting the international consensus on productivity measurement, the Advisory Committee on Measuring Innovation in the Twenty first Century Economy to the US Secretary of Commerce (2008, page 7) recommended that the Bureau of Economic Analysis (BEA) should: Develop annual, industry-level measures of total factor productivity by restructuring the NIPAs to create a more complete and consistent set of accounts integrated with data from other statistical agencies to allow for the consistent estimation of the contribution of innovation to economic growth. The principles for constructing industry-level produc- tion accounts are discussed by Fraumeni et al. (2006). Disaggregating the production account by industrial sector requires the fully integrated system of input–output accounts and accounts for gross product originating by industry, described by Lawson et al. (2006), and Moyer et al. (2006). Donahoe et al. (2010) present data for the fully integrated system for 1998–2008 on a NAICS basis. Jorgenson et al. (2005), the EU KLEMS project descri- bed by O’Mahony and Timmer (2009), and the studies presented in Jorgenson (2009a), The Economics of Pro- ductivity, present industry-level data on productivity. These data have made possible the international comparisons of patterns of structural change presented by Jorge