Click for next page ( 119


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 118
Technology and the Services Sector: America's Hidden Competitive Challenger STEPHEN S. ROACH Information technology has always played an important role in the services sector of the U.S. economy. In recent years, however, services industries have stepped up their acquisitions of computers, telecommunications equip- ment, and other such products dramatically. As a result, the broad segment of the economy that can be classified as services providers now owns about 84 percent of the total U. S. stock of information technology items. Moreover, relative to goods-producing industries, a much larger proportion of the ser- vices sector's capital budgets is spent on information technology, revealing a significantly greater dependence by services on such technology as a factor of production. This reliance underscores information technology's strategic importance in the U.S. competitive challenge: with services now the pre- dominant mode of economic activity in the United States, a productivity payback from information technology is absolutely essential to keep the economy on a longer term path of sustainable growth. So far, the services sector has little to show for its spending binge on technology. Quite simply, massive investments in information technology have failed to boost national productivity growth in the present decade. Furthermore, with manufacturing productivity now on the rebound, problems in the services sector loom increasingly large in America's broader com- petitive struggle. It is certainly not too late. New and creative applications of information technology could still enhance the productivity performance of the services sector's predominantly white-collar work force. Until that ~ Some of the findings reported in this chaper are drawn from studies previously published by Morgan Stanley; for the most recent such effort, see Morgan Stanley (1987). 118

OCR for page 118
TE:CHNOLOGY AND THE SERVICES SECTOR 119 payback begins to occur, however, the role of technology spending will be under growing suspicion (Bower, 1986; Neikirk, 1987; Schneider, 19871. In what follows, an attempt is made to provide a detailed industry-by- industry assessment of services sector spending on information technology. By way of background, the broad contours of capital formation in services industries are first examined over the post-World War II era. CAPITAL ACCUMULATION IN THE SERVICES SECTOR Over the course of contemporary economic history, the services sector has emerged as a major force in shaping capital formation in the United States. As the upper portion of Figure 1 indicates, up until about the mid-1960s capital budgets in the goods-producing and services-providing sectors were of roughly equal magnitudes in "teal," or inflation-adjusted, terms. Over the past 20 years, however, a sizable gap has opened between the investment programs of these two segments of the economy; at present, services-pro- viding industries account for nearly 60 percent of corporate America's total 300 o c\5 O ~ 200 ._ m 0 60 50 45 i:.. Services Providing Industries ~ _ Services t~ 7- as a Share of ~ ~ Total Capital ~ /\ Spendinn :,~ 1948 1952 1956 1960 1964 1968 1972 1976 1980 1984 1988 Producing Industries Hi, ~ _~ <: I I . Year FIGURE 1 Post-World War II trends in services sector capital spending. NOTE: Shaded areas indicate recessionary periods as designated by the National Bureau of Economic Research. The goods-producing sector is defined to include manufacturing, mining, and construction; the services sector from which govern- ment entities are excludedconsists of trade, communications, transportation, public utilities, finance, insurance, real estate, business services, professional services, health, and legal and educational institutions.

OCR for page 118
120 STEPHENS. ROACH outlays on plant and equipment. Moreover, as the lower portion of this figure illustrates, services sector investment budgets, in general, have tended to be relatively immune to the ups and downs of the business cycle. With one exception the contraction of 1973 to 1975 the investment share of this segment of the economy actually has risen during the major recessions of the past 35 years. This does not mean that services providers get away unscathed during recessions. However, as Figure 1 shows, investment cut- backs during periods of cyclical distress are generally smaller for these in- dustr~es than for manufacturers, with the latter still bearing the brunt of most serious shortfalls in economic activity. Table 1 provides a summary of investment trends In major segments of the services sector over the past three and half decades. In the l950s and 1960s, capital investment of services providers grew at or slightly below spending rates in goods-producing industries. Not until the 1970s did services really come to the fore as the leader of capital formation in the United States. During the 1970s decade, services providers increased capital budgets (in real terms) at a 4.6 percent average annual rate; although that performance represented a slowdown from the prior decade, it was one-third faster than the 3.4 percent yearly gains in the goods-producing segment. In the 1980s, overall investment advances have slowed further, although gains in the ser- vices segment have continued to outdistance those in goods-producing in- dustries by a considerable margin. By industry, the recent strength in services sector investment has been concentrated in finance, insurance, and real estate; wholesale trade; and retail trade segments of the economy. By contrast, over the past decade and a half, capital spending growth has slowed for trans- TABLE 1 Shifting Trends in Business Fixed Investment, Average Annual Growth Rates (percent) - Business Sector l950s 1960s 1970s 1 980sa Ace INDUSTRIES 2.8 6.9 4.0 2.2 Goods producing 3.6 6.9 3.4 1.8 Manufacturing 3.9 7.6 2.7 2.6 Other 2.0 0.5 11.4 - 3.6 Services providing 2.3 6.8 4.6 2.6 Transportation 0.4 6.6 0.8 -2.0 Communications 7.2 8.9 4.9 1.1 Public utilities 0.9 6.8 3.6 - 0.5 Trade 4.8 4.5 7.4 6.0 Finance and insurance 10.8 8.8 11.0 12.8 Personal and business services 4.5 8.2 4.2 2.4 aThrough 1985 NOTE: Figures are based on constant 1982 dollars and are taken from the plant and equipment survey of the U.S. Department of Commerce.

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 121 portation services, public utilities, personal and business services, and com- munications. Similar conclusions are apparent in examining recent trends in the services sector's stock of information technology capital, which is hardly surprising since such a stock reflects the accumulation of past investment flows. Figure 2 summarizes three and a half decades of growth in the U.S. business sector's capital stock, with an emphasis on the diverging experience of services and goods producers. While capital stock growth in both segments has slowed in the present decade relative to peak increases in the 1960s, gains in the services sector have generally held up much better, running "only" 25 percent below the record rates of two decades ago, compared to a 75 percent growth shortfall in manufacturing. Overall trends in services sector capital accumulation mask some important differences among industnes, as Table 2 illustrates. For trade establishments, capital stock growth has actually picked up in the present decade relative to the pace of the 1970s. A similar trend is observable for a collection of services sector industries referred to as "information producers" communications, finance, insurance, real estate, business services, and legal services. All in 6 4 a) a, 2 ~ Goods Producers E] Services Providers ~3 HI Industries 1 970s 1 980s 1 950s 1 960S Decade FIGURE 2 Growth in the U.S. capital stock. NOTE: Figures are average annual growth rates based on constant 1982 dollars; capital stock data are net stocks using constant-cost valuation and straight-line de- preciation patterns; the data are taken from U.S. Department of Commerce statistics (described in Musgrave, 19861. The "1980s" includes data through 1985. For our analysis, an extension of the industry detail reported in that article was used, which provides both an industry and a commodity breakdown of the private nonresidential capital stock. The resulting industry-commodity capital stock matrix contains 59 industry groupings; each industry, in turn, has up to 21 categories of equipment and structures. We are grateful to Mr. Musgrave for both providing these data and being most generous with his time in clarifying several issues regarding data construction.

OCR for page 118
122 STEPlIEAt S. ROACH TABLE 2 Postwar Trends in the Capital Stock, Average Annual Growth Rates (percent) Business Sector 1950s 1960s 1970s 1980Sa Ace INDUSTRIES 3.8 4.3 3.6 3.1 Goods producing 4.3 3.8 3.2 2.1 Manufactunng 3.5 4.7 3.3 1.7 Other 5.7 2.2 3.1 2.8 Services providing 3.6 4.5 3.8 3.6 Information producers 6.4 6.8 4.6 5.0 Communications 6.4 6.7 6.0 4.6 Finance 5.8 8.7 7.4 7.8 Insurance 7.2 6.2 3.4 6.3 Real estate 6.2 6.2 3.1 4.0 Business services 10.3 1 1.4 6.8 7.9 Legal services 5.0 2.8 1.4 6.2 Trade 3.9 6.8 5.1 5.6 Other 2.4 2.7 2.8 1.5 aThrough 1985 NOTE Figures are based on constant 1982 dollars and are derived from U.S. Department of Commerce statistics. all, whether it is investment spending or capital stock growth, the message that comes through loud and clear is that the services sector clearly has taken over as the engine of capital formation in the United States. TECHNOLOGY AND SERVICES SECTOR INVESTMENT Not surprisingly, it turns out that the services sector's impressive record on capital accumulation largely reflects a voracious appetite for information technology. Figure 3 divides services sector capital stock growth into tech- nology and "nontechnology" items. As the middle panel shows, the stock of information technology capital in the services sector has expanded at close to a 10 percent average annual rate over the past 35 years; although the growth rate slipped a bit in the 1960s and 1970s, it has accelerated signifi- cantly in the present decade to approximately a 12 percent average annual pacea record performance by historical standards. Indeed, had it not been for the rapid gains in such technology spending, total services sector capital accumulation would have slowed considerably further over the past decade and a half. As Figure 3 shows, thus far in the 1980s, nontechnology services sector capital has risen at only a 2.5 percent yearly pace, over one-third slower than average gains of the period 1950-1979 and the weakest perfor- ~mance of any decade in the postwar era. Figure 4 offers more insight into shifting trends in services sector invest-

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 15 a' a) 10 5 o 1 950S 1 960S 123 1 ~ Other Capital ~3 Information Technology Total 1970S Decade 1 980S FIGURE 3 Growth in the services sector capital stock. NOTE: Figures are average annual growth rates based on constant 1982 dollars; capital stock data are net stocks using constant-cost valuation and straight-line de- preciationpatterns. The "1980s" includes date through 1985. Information technology capital, or information-processing equipment, is defined to include computers, office and accounting machinery, communications equipment, instruments, and photo- copiers and related equipment. Industry estimates of information technology capital were obtained by a technology "sort" of the industry-commodity capital stock matrix described for Figure 1. Obviously, other nonservices industries have increased their reliance on new technologies, ranging from machine tools and motor vehicles to biogenetics and artificial intelligence. These items are not included in this analysis of information technology capital largely because of the relatively small role they play in overall business activity and also because they are not separately identifiable in the measurement framework employed, which is based on gross national product. ment from 1970 to 1985; the functional composition of the capital stock in this segment of the economy is shown. Of the four major product categories identified, only information technology has experienced an increase in its share of the total services sector's capital stock; in 1970 information tech- nology accounted for only 6.4 percent of the total services capital, whereas 15 years later the share had risen almost 2~/: times to 15.5 percent. This sharp surge in information technology purchases has been accom- modated by a significant rearrangement in the capital budgets of services- providing industries. Most significant in this regard has been a contraction of the portion of capital now invested in buildings. The transportation equip- ment share has dropped only fractionally, and industrial equipment has es- sentially maintained its relatively small portion of 15 years earlier. For comparative purposes, Figure 5 provides a parallel analysis of com- positional shifts in the capital stock of the goods-producing segment of the

OCR for page 118
124 STEPHENS. ROACH 1970 $1,343.5 Billion Buildings 64.0% Buildings 54.5% - 1 985 $2,300.4 Billion . Basic Industrial 7.9% _ Information Technology 6.4% ; ~ Other 7.0% ~ Transportation 14.6% , Basic Industrial 7.4% ,.~:::: ... _ Information _ \ Other 9.0% / - Transportation 13.6% FIGURE 4 Shifting composition of services sector capital stock. NOTE: Figures are based on constant 1982 dollars. Because of rounding, totals may not add up to 100.0 percent. Information technology capital includes office, com- puting, and accounting machinery; communications equipment; instruments; and photocopiers and related equipment. In addition to the information technology group- ing, this functional breakdown also includes the following broad categories: "basic industrial capital," which encompasses fabricated metal products, engines and tur- bines, metalworking machinery, special industry machinery, general transmission, distribution, and industrial apparatus; "buildings," which consists of commercial, office, and public utility structures; and "transportation," which is composed of automobiles, trucks, trailers, buses, commercial aircraft, and rail equipment. Technology 15.5% U.S. economy. While there have been analogous rearrangements of pro- ductive capital, goods producers remain considerably less dependent on in- formation technology as a factor of production. In 1985, approximately 6 percent of their total capital was invested in such technology items. While that is about three times the average share of the 1970s, it is less than half the portion currently found in the services sector. Without question, the most

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 1 970 $729.8 Billion _ - Basic Industrial 40.6% Buildings 47.3% 1 985 $1,098.0 Billion - - Basic Industrial 39.7% Buildings 44.4% ~' 125 \ Information Technology 1.2% '1 ~ Other 7 Q% Transportation 3.1% Information Technology 6.1% r ~ Other 6.9% _ / _ ~ ~ Transportation 3.0% FIGURE 5 Shifting composition of goods-producing sector capital stock. NOTE: Figures are based on constant 1982 dollars. Because of rounding, totals may not add up to 100.0 percent. Information technology capital includes office, com- puting, and accounting machinery; communications equipment; instruments; and photocopiers and related equipment. intensive users of information technology are in the services sector; moreover, this same segment has continued to move dramatically to increase this de- pendence. INDUSTRY DETAIL Tables 3 through 5 provide a detailed industry-by-industry assessment of shifting trends in the stock of information technology capital in the services sector. Benchmark assessments describing the concentration of. technology ownership within various segments of the services grouping can be found in Tables 3 and 4. In 1985 the last year for which complete industry and

OCR for page 118
126 STEPHEN S. ROACH TABLE 3 Where is America's Information Technology Capital? (billions of constant 1982 dollars) Business Sector 1950s 1960s 1970s 1985 Ace INDUSTRIES 28.8 61.5 142.7 423.6 Goods producing 4.7 6.4 17.2 66.7 Services providing 24.1 55.1 125.5 356.9 Transportation 1.0 1.2 1.5 3.1 Rail 0.4 0.6 0.8 0.6 Nonrail 0.5 0.6 0.7 2.4 Air 0.1 0.1 0.1 1.2 Trucking 0.1 0.1 0.0 0.2 Other 0.4 0.4 0.5 1.0 Communications 10.5 30.6 73.4 159.7 Telephone and telegraph 10.4 29.9 71.3 154.4 Broadcasting 0.2 0.7 2.1 5.3 Public utilities 0.8 1.4 3.9 13.4 Electric 0.6 1.1 3.4 11.0 Gas and other 0.2 0.3 0.5 2.5 Total trade 0.5 1.1 5.7 41.1 Wholesale trade 0.3 0.7 4.0 33.9 Retail trade 0.2 0.4 1.6 7.2 Finance, insurance, and real estate 8.1 13.7 24.6 90.3 Finance and insurance 0.8 1.3 4.2 46.4 Banks (including Federal Reserve) 0.2 0.4 1.6 20.5 Credit agencies 0.1 0.3 1.1 15.6 Securities brokers 0.0 0.0 0.1 0.9 Insurance carriers 0.2 0.2 0.7 6.1 Insurance agents 0.2 0.3 0.3 1.0 Investment holding companies 0.1 0.1 0.4 2.3 Real estate 7.3 12.4 20.4 44.0 Services 3.1 7.1 16.4 49.3 Hotels and lodging 0.0 0.0 0.1 0.9 Personal 0.1 0.8 1.6 1.5 Business 0.3 1.2 4.3 22.3 Auto repair 0.1 0.1 0.1 0.8 Miscellaneous repair 0.0 0.0 0.0 0.2 Motion pictures 0.1 0.4 1.4 2.4 Amusement and recreation 0.4 1.1 1.9 3.8 Health 1.2 2.4 5.4 13.7 Legal 0.0 0.1 0.1 0.7 Educational 0.1 0.1 0.2 0.7 Other 0.7 1.0 1.3 2.1 NOTE: Figures are averages over designated intervals and are Morgan Stanley estimates derived from the Industry-Commodity Capital Stock Matrix of the U.S. Department of Commerce.

OCR for page 118
TECHNOLOGY AND TlIE SERVICES SECTOR TABLE 4 Industry Shares of Information Technology Capital 127 Percentage Business Sector Change: 1985 vs. 1970s Percentage 1950s 1960s 1970s 1985 Points Ratioa Ale INDUSTRIES 100.0 100.0 100.0 100.0 0.0 1 .0 Goods producing 17.1 10.6 11.4 15.8 4.3 1.4 Services providing 82.9 89.4 88.6 84.2 -4.3 1.0 Transportation 3.6 2.0 1.1 0.7 - 0.4 0.7 Rail 1.6 1.0 0.6 0.1 - 0.5 0.2 Nonrail 2.0 1.0 0.5 0.6 0.1 1.1 Air 0.3 0.2 0.1 0.3 0.2 3.3 Trucking 0.3 0.2 0.0 0.0 0.0 1.7 Other 1.4 0.7 0.4 0.2 - 0.1 0.6 Communications 36.1 48.8 52.0 37.7 - 14.3 0.7 Telephone and telegraph 35.5 47.8 50.5 36.5 - 14.0 0.7 Broadcasting 0.6 1.1 1.5 1.2 -0.3 0.8 Public utilities 3.0 2.2 2.7 3.2 0.5 1.2 Electric 2.2 1.7 2.3 2.6 0.3 1.1 Gas and other 0.9 0.5 0.3 0.6 0.2 1.7 Total trade 1.8 1.8 3.7 9.7 6.0 2.6 Wholesale trade 1.1 1.1 2.6 8.0 5.4 3.1 Retail trade 0.8 0.7 1.1 1.7 0.6 1.5 Finance, insurance, and real estate 27.8 23.0 17.5 21.3 3.8 1.2 Finance and insurance 2.7 2.3 2.7 10.9 8.2 4.0 Banks (including Federal Reserve) 0.6 0.6 1.0 4.8 3.8 4.8 Credit agencies 0.5 0.5 0.7 3.7 3.0 5.5 Securities brokers 0.1 0.1 0.1 0.2 0.1 2.1 Insurance carriers 0.6 0.4 0.5 1.4 1.0 3.1 Insurance agents 0.6 0.5 0.2 0.2 0.0 1.2 Investment holding companies 0.4 0.2 0.3 0.5 0.3 1.8 Real estate 25.1 20.7 14.8 10.4 -4.4 0.7 Services 10.5 11.6 11.7 11.6 0.0 1.0 Hotels and lodging 0.0 0.0 0.0 0.2 0.2 5.9 Personal 0.5 1.2 1.2 0.4 -0.8 0.3 Business 1.1 1.9 3.0 5.3 2.2 1.7 Auto repair 0.2 0.1 0.0 0.2 0.1 4.5 Miscellaneous repair 0.0 0.0 0.0 0.0 0.0 2.6 Motion pictures 0.5 0.6 1.0 0.6 - 0.4 0.6 Amusement and recreation 1.4 1.8 1.4 0.9 - 0.5 0.7 Health 4.1 3.9 3.8 3.2 - 0.6 0.8 Legal 0.1 0.1 0.1 0.2 0.1 1.6 Educational 0.3 0.2 0.1 0.2 0.1 1.7 Other 2.3 1.7 0.9 0.5 - 0.4 0.6 al985/1970s average. NOTE: Figures are averages over designated intervals and are Morgan Stanley estimates derived from the Industry-Commodity Capital Stock Matrix of the U.S. Department of Commerce.

OCR for page 118
128 STEPHENS. ROACH TABLE 5 Information Technology Capital as a Share of Each Industry's Overall Capital Stock Percentage Change: 1985 vs. 1970s Percentage Business Sector 1950s 1960s 1970s 1985 Points Ratioa Ace INDUSTRIES 2.5 3.7 5.8 12.5 6.7 2.2 Goods producing 1.1 1.1 2.0 6.1 4.1 3.0 Services providing 3.3 5.2 7.7 15.5 7.8 2.0 Transportation 0.4 0.5 0.6 1.1 0.5 1.9 Rail 0.2 0.5 0.7 0.6 - 0.1 0.9 Nonrail 0.8 0.7 0.5 1.4 0.9 2.8 Air 1.5 0.7 0.4 3.0 2.5 7.3 Trucking 0.5 0.5 0.1 0.4 0.3 3.7 Other 0.8 0.8 0.8 1.3 0.5 1.7 Communications 20.0 30.6 40.8 53.4 12.6 1.3 Telephone and telegraph 20.8 31.9 42.1 54.5 12.5 1.3 Broadcasting 5.7 10.6 19.9 32.9 13.0 1.7 Public utilities 0.5 0.6 1.1 3.1 2.1 2.9 Electric 0.5 0.6 1.2 3.2 2.0 2.6 Gas and other 0.5 0.4 0.6 2.8 2.3 5.0 Total trade 0.7 0.9 2.5 11.1 8.7 4.5 Wholesale trade 1.6 1.8 4.9 22.5 17.6 4.6 Retail trade 0.4 0.5 1.1 3.3 2.1 2.9 Finance, insurance, and real estate 6.1 5.7 6.0 14.4 8.5 2.4 Finance and insurance 4.1 3.5 4.7 27.3 22.7 5.9 Banks (including Federal Reserve) 1.7 1.9 3.9 26.3 22.4 6.7 Credit agencies 3.1 3.1 3.3 25.0 21.7 7.6 Securities brokers 2.6 3.9 8.3 31.6 23.3 3.8 Insurance carriers 6.0 4.4 7.2 38.0 30.8 5.3 Insurance agents 17.9 15.6 12.0 32.7 20.8 2.7 Investment holding companies 14.2 8.0 10.2 31.2 21.0 3.1 Real estate 6.4 6.1 6.3 9.7 3.4 1.5 Services 5.8 6.5 8.3 16.1 7.9 2.0 Hotels and lodging 0.1 0.1 0.1 1.7 1.6 11.5 Personal 2.7 9.3 13.6 11.5 - 2.1 0.8 Business 5.1 7.9 10.9 28.4 17.5 2.6 Auto repair 0.6 0.5 0.2 1.5 1.3 7.6 Miscellaneous repair 0.5 0.4 0.5 2.7 2.2 5.3 Motion pictures 8.3 15.0 31.5 42.2 10.6 1.3 Amusement and recreation 5.3 9.7 12.3 19.7 7.4 1.6 Health 21.2 16.0 19.2 29.5 10.3 1.5 Legal 1.9 2.9 4.0 13.3 9.3 3.3 Educational 10.0 9.5 12.2 46.8 34.6 3.8 Other 19.0 10.0 6.6 10.5 3 9 1.6 a 1985/1970s average. NOTE: Figures are averages over designated intervals and are Morgan Stanley estimates derived from the Industry-Commodity Capital Stock Matrix of the U.S. Department of Commerce.

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 129 commodity detail is available the services-providing sector as a whole owned 84 percent of the U.S. economy's total stock of information tech- nology capital. Within services, about 45 percent of this capital can be found in the communications industry (Table 3), largely reflecting the sizable in- vestment in a nationwide telephone system. The finance and real estate sectors are on the next rung down the ladder, accounting in 1985 for about 25 percent of total services sector technology capital. Other large owners of such capital include wholesale trade, business services, and health care providers. To be sure, the dispersion of information technology capital across the various industries is only a helpful starting point. The key questions that need to be addressed in assessing the role of these technology products really have more to do with how industry has changed its reliance on such items as a factor of production. Table 5 deals with this issue by providing a detailed analysis of the share of each industry's (or industry grouping's) total capital stock that can be accounted for by information technology products, mea- suring what is called "technology endowment." Table 6 summarizes this analysis in a systematic and consistent fashion by estimating what can be referred to as "technology intensity measures" (TIMs). Two different TIMs have been constructed in an effort to provide a uniform assessment of relative shifts in information technology endowment. TIM1 is each industry's change in technology endowment the 1985 level relative to that which prevailed, on average, over the 1970s compared with the typical change in the economy at large; it essentially measures the relative speed by which various segments have increased the technology share of their capital stock. TIM2 is each industry's 1985 technology endowment relative to the economy-wide average. Thus, each of the TIM variants cap- tures a somewhat different dimension of the shift into information technology. TIM1 looks at movements in the technology content of each industry's capital stock, whereas TIM2 is more of a static picture of the relative dispersion of technology endowment in 1985. Communications As Table 5 indicates, no services-producing segment can match the tech- nology dependency of the communications sector. Over 50 percent of the total capital stock owned by the telephone and telegraph industry can be classified as information technology items a share that has risen steadily over the past 35 years. This sector's leading role as a high-technology user reflects, of course, the telephone industry's enormous stock of communi- cations equipment that dates back well before the advent of modern-day telecommunications and computers. Thus, this is a mature industry as seen from the standpoint of shifting technology dependence. Because of this char- acteristic, TIM1 our barometer of relative changes in technology inten-

OCR for page 118
130 TABLE 6 Technology Intensity Measures (TIMs) Business Sector TIM1 TIM2 Ace INDUSTRIES 1.0 1.0 Goods producing 1.4 0.5 Services providing 0.9 1.2 Transportation 0.9 0.1 Rail 0.4 0.0 Nonrail 1 .3 0. 1 Air 3.4 0.2 Trucking 1.7 0.0 Other 0.8 0. 1 Communications 0.6 4.3 Telephone and telegraph 0.6 4.4 Broadcasting 0.8 2.6 Public utilities 1 .3 0.3 Electric 1.2 0.3 Gas and other 2.3 0.2 Total trade 2.1 0.9 Wholesale trade 2.1 1.8 Retail trade 1 . 3 0. 3 Finance, insurance, and real estate 1.1 1.2 Finance and insurance 2.7 2.2 Banks (including Federal Reserve) 3.1 2.1 Credit agencies 3.5 2.0 Securities brokers 1.8 2.5 Insurance carriers 2.4 3.1 Insurance agents 1.3 2.6 Investment holding companies 1.4 2.5 Real estate 0.7 0.8 Services 0.9 1 .3 Hotels and lodging 5.3 0.1 Personal 0.4 0.9 Business 1.2 2.3 Auto repair 3.5 0.1 Miscellaneous repair 2.4 0.2 Motion pictures 0.6 3.4 Amusement and recreation 0.7 1.6 Health 0.7 2.4 Legal 1.5 1.1 Educational 1.8 3.8 Other 0.7 0.8 STEPHENS. ROACH NOTE: TIM1 is each industry's change in technology endowment from the 1970s to 1985, relative to the average change for all industries. TIM2 is each industry's 1985 technology endowment relative to the all-industries average. Technology intensity measures are Morgan Stanley estimates derived from the Industry-Commodity Capital Stock Matrix of the U.S. Department of Commerce.

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 131 sityranks communications at the bottom of the services sector in terms of its ability to increase its reliance on high technology as a factor of production. Quite simply, since the stock of information technology capital is so large in communications, it cannot possibly grow, for all practical purposes, at the rate experienced by other services providers. F. Inance As TIM2 shows, the finance sector is second only to communications in the degree of its present dependence on information technology. As Table 5 indicates, banks, credit agencies, security brokers, and investment holding companies currently have about 25-30 percent of their capital stock invested in such technologies; for the insurance industry the share is even higher. Also, as the results of TIM1 confirm, this sector has made by far the most rapid move into information technology over the present decade; by industry, credit agencies and banks have increased their technology endowment most dramatically, followed, in order, by insurance carriers, security brokers, and investment holding companies. These results demonstrate quite clearly that of all the services-providing industries, the finance sector has wagered the largest bet on the ultimate productivity payback from information technology. Trade As Table 5 shows, wholesale trade establishments have also experienced a sizable increase in their reliance on information technology as a factor of production; this trend stands in sharp contrast to the retail sector, whose dependence on such capital has increased only modestly in recent years, despite the growing profusion of automated end-sales and inventory control equipment. A chronic problem with trade sector data, however, has been an inability to make accurate sampling distinctions between wholesalers and retailers; consequently, for the purposes of this analysis, it seems best to treat that segment of the economy as a unified industry division. In this context, while TIM1 suggests that the trade sector has moved vigorously to acquire information technologies, TIM2 indicates that this surge has come off a low base; indeed, compared with other industries, the trade sector's endowment of such technology still appears a bit on the low side. Miscellaneous Services A final grouping of miscellaneous services providers accounts for the remainder of the services sector's spending on inflation technology. As Table 3 indicates, from the standpoint of aggregate technology outlays the only areas of significance in this collection of industries are business services

OCR for page 118
132 STEPHENS. ROACH and health care. The TIM1 measure for the former indicates that this industry has experienced a relatively rapid shift into technology; as TIM2 shows, the speed of this move has been sufficient to leave business services with one of the services sector's highest relative endowments of spending on infor- mation technologies. In health care, the shift into such technology has been less dramatic, but that is probably because this industry has always had a fairly hefty endowment of such capital as seen from a TIM2 perspective. Technology ownership in other services-providing industries whether trans- portation, public utilities, law firms, or educational institutions constitutes a very small slice of the U.S. information technology pie and thus is not a real factor in the "high-teching" of the services sector. Slight exceptions can be made for the legal and educational groupings where TIM1 findings point to very sharp increases in technology intensity during the present de- cade. THE COMPUTER REVOLUTION Not all technology is alike, of course. The computer is widely perceived to offer the greatest opportunities for improved efficiencies in the workplace, and, not surprisingly, computers are the single largest line item in the in- formation technology budgets of services-providing industries.] Excluding the communications sector and its nationwide telephone system, computers total 49 percent of all services sector info~ation-producing capital. That ratio has risen significantly over the present decade and compares with an average of only 17 percent during the entire 1970s. In fact, over the past 15 years, growth in the stock of computers for services has averaged 22 percent per year, almost three times the gains for other technology items and more than five times the growth pace of all services sector capital. By industry, the computer story actually differs little from that suggested by our previous analysis of total services sector information technology. The finance sector, which owns 49 percent of all services sector computers, also has the highest dependence on this item as a factor of production. Financial services, as of 1985, are estimated to have 27.3 percent of their total capital held in the form of computers and other office machinery a share over seven times the ratio for all industries. Within the finance sector, securities brokers, investment holding companies, and insurance companies have the highest dependence on computers, followed in turn by banks and credit agencies. Outside of finance, wholesale trade and business services appear to be most reliant on computers. However, in both of these instances, the computer share of total capital is about half the size of that in the finance sector. Under miscellaneous services, computer ownership is generally small, although both legal services and educational institutions have accelerated their purchases

OCR for page 118
TECHNOLOGY AND TlIE SERVICES SECTOR 133 greatly over the present decade. The only exception is communications where, in 1985, more than 98 percent of technology stock was held in the form of telecommunications equipment. Thus, the communications industry, with its relatively small endowment of computers, is not a major factor when the analysis is framed exclusively in terms of the computer portion of the total stock of information technology. THE PRODUCTIVITY PUZZLE There can be little debate over the services sector's mounting reliance on information technology. The big question of course is, What does the U.S. economy have to show for these efforts? The answer is not much at least, not yet. Table 7 matches up TIMs' estimates against the available industry record on services sector productivity performance. The results are uniformly dis- appointing. Services providers as a whole, with an information technology endowment that is at Present 2~/: times the size of that in the goods sector (as shown by TIM2), have experienced a clear slowdown in productivity ~ , TABLE 7 Technology and Productivity Disappointments in the Services Sector Technology Intensity Average a Measures Productivity Growth (%) Change (percentage TIM1 TIM2 1973-1979 1979 1985 pointy Ace INDUSTRIES 1.0 1.0 0.6 1.1 Goods producing 1.4 0.5 0.5 2.2 Services providing 0.9 1.2 0.7 0.4 - 0.3 Transportation 0.9 0.1 1.5 - 1.2 - 2.7 Rail 0.4 0.0 1.1 3.5 2.4 Nonrail 1.3 0.1 1.7 -1.8 -3.5 Communications 0.6 4.3 4.3 3.9 - 0.4 Public utilities 1.3 0.3 0.3 1.8 1.5 Trade 2.1 0.9 0.8 1.3 0.5 Finance and insurance 2.7 2.2 - 0.1 - 1.3 - 1.2 Real estate 0.7 0. 8 - 0.2 - 1.4 - 1.2 Miscellaneous services 0.9 1.3 0.2 0.7 0.5 aDifference in percentage points between 1973-1979 period and 1979-1985 period. NOTE: TIM1 is each industry's change in technology endowment from the 1970s to 1985, relative to the average change for all industries. TIM2 is each industry's 1985 technology endowment relative to the all-industries average. Technology intensity measures are Morgan Stanley estimates derived from the Industry-Commodity Capital Stock Matrix of the U.S. Department of Commerce; produc- tivity detail is taken from Multiple Productivity Indexes, published by the American Productivity Center based on U.S. government statistics.

OCR for page 118
134 STEPHENS. ROACH growth so far this decade. Moreover, that deterioration follows a period of relatively meager productivity increases over most of the 1970s. The tech- nology "leader"- the finance sector is estimated to have experienced an especially dramatic deterioration in productivity performance in recent years. In addition, whereas productivity growth is estimated to have nicked Fin _ ~ _ _1_ .1 _ _ ~ ~ ~ . Sony In trade and miscellaneous services, after barely rising at all during the 1970s, such improvement seems paltry when compared with the rapidly growing intensity of technology endowment in each of these segments of the economy. Of course, there are those who say in response that the problem is one of measurement. After all, can anyone really calculate services sector produc- tivity? The difficulty with such a claim is that measurement problems have always plagued estimates of services sector output. The question is: Have measurement pitfalls worsened appreciably during the past decade a period when productivity growth has deteriorated and technology spending has ac- celerated most dramatically? There is no conclusive evidence to suggest that this is the case. Most likely, measurement errors have cumulated over a long period of time and, thus, have had a much greater impact on the level of productivity than on its rate of change over the past several years. Thus, it seems an extremely remote possibility that the services sector's productivity disappointments of the 1980s would disappear if the numbers could, in fact, be corrected for statistical problems that date back well into the l950s.2 It is one thing to identify a productivity problem, but it is another task altogether to explain why it occurs.3 Where the above analysis has been substantiated largely by empirical evidence, efforts to identify the reasons behind services sector productivity disappointments will have to rely mainly on circumstantial evidence. As a consequence, one must conclude with as many questions as answers. The place to start is with the work force, the human input to any productive endeavor. The services sector's work force is unlike that of any other segment of the U.S. economy. About two-thirds of all services sector employees can be classified as working in the so-called white-collar occupations. In fact, over 80 percent of all white-collar workers in the United States are employed by services-providing industries. By contrast, goods producers have only about one-third of their work force employed in white-collar job categories. The unique occupational characteristics of white-collar workers underscore the important role played by information technology as a factor in the services sector's productivity puzzle. Services sector "production" turns out to be a process that largely takes place in the office, and white-collar workers using a rapidly growing stock of information technologyhave simply not been able to create the efficiencies that would trigger a meaningful growth in productivity.4 There is far more to this problem than just pointing a finger at computers

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 135 and other information-producing technologies. Admittedly, one of the im- ponderables simply may be time. As American managers and other "infor- mation workers" become more adept at understanding and utilizing complex new technologies, they climb the traditional "learning curve" that eventually opens the door to ever-greater efficiencies. The lack of an instantaneous payback does not necessarily preclude a transitional break-in period that allows information technology to grow naturally into its full productive po- tential.S Another facet of the productivity problem may also be the unique nature of the technology itself. The computer revolution is, in fact, the first major technological breakthrough in the U.S. economy that is being applied primarily to the generation of that seemingly amorphous commodity called services sector "output." By contrast, in the past, other technological break- throughs have been applied mainly to the production or delivery of goods. Evaluating such technology investment must, therefore, be done in the ab- sence of well-established criteria and guidelines. Consequently, without the historical benefit of such experience, estimates of productivity paybacks from information technology are in many instances a "shot in the dark." This puts the problem squarely in the lap of U.S. managers, and in that regard the task is a formidable one. A first imperative is to develop an accurate system of measuring white-collar productivity. Only then can the cost of information technologies be evaluated more precisely against the work effort of those employees who rely on such equipment. One of the clear problems in "costing-out" information technology is the failure to recognize that a broad spectrum of workers is involved in its productive applications, ranging from the so-called knowledge workers who control the analytical applications to the vast cadre of management information system support staffs that are required to keep the machinery in operation. Thus, a revision of traditional accounting standards may be an important prerequisite to developing a white-collar productivity measurement system.6 Additionally, managers need to give more thought to a strategy of tech- nology acquisition. Fifteen years into a phenomenal technology-buying binge, corporate America appears hooked on each and every twist of the product cycle. The latest in computer gimmickry does not necessarily guarantee obsolescence of an existing installation of information technology. To avoid the temptation of open-ended technology budgets, the replacement cycle needs to be evaluated against more explicit productivity criteria of the user community. Another area of concern pertains to the "bows" and "whys" of the application of inflation technology to productive endeavors in the services sector. Two issues come to mind in this regard: product selection and soft- ware. From the standpoint of product selection the issue is really that of the general-purpose machine versus the more focused piece of equipment. Tech- nology acquisition needs to be directed more explicitly to the task at hand.

OCR for page 118
136 STEPHENS. ROAClI The automated teller machine of the banking industry is an example of a product-specific piece of technology equipment that has a highly focused application; mainframes and other general-purpose computers clearly would be redundant in such a capacity, which is a general lesson that could be relevant for other services sector industries as well. Software, needless to say, has always been critical to the productivity payback of information technology. To date, progress has been far more rapid in hardware than in programming and operating systems that link information technology explicitly to production. The manufacturing sector has thus far reaped most of the productivity benefits from software break- throughs especially in areas such as computer-aided design and manufac- ture, expert systems, and artificial intelligence. By contrast, the services sector clearly has been laggard in exploiting the efficiencies of technology . . app .lcatlons. This discussion has been aimed at developing a better understanding of technology's disappointing productivity payback in the services sector. Yet given the diversity of industries that provide services in the United States, it is quite possible that information technology just is not the answer in some areas. For transportation and lodging, it is obvious that greater efficiencies in computer-based reservation systems offer a real opportunity for greater booking volume per unit of worker input. An analogous conclusion can be drawn for the role that information technology plays in facilitating a greater volume of transactions in the financial and trade sectors. What about lawyers, consultants, investment counselors, educational in- stitutions, and other professionals who produce information as a final product? Is it really in the economy's best interest to equip these services providers with efficient technologies that have the potential for overwhelming the economy with information? Spreadsheets and word processing are remarkable tools, but if they are abused, the economy could find itself engulfed in an ocean of information. The potential is " information overload, " and questions in this regard obviously go well beyond the productivity challenges facing individual industries. It may well be that many services sector activities should also be scrutinized from the "macro" standpoint of how they affect the productivity of those who purchase such services an admittedly for- midable task. Whatever the answer for the services sector, it seems increasingly clear that the high-technology revolution of the information age has yet to deliver from the standpoint of national productivity enhancement. In this regard, it also seems increasingly clear that the link between technology and produc- tivity performance is central to the present debate on America's competitive dilemma. The lingering productivity shortfall in the United States is no longer a problem traceable mainly to manufacturing, since that sector has now moved back to its longer term productivity growth trend. However, the services

OCR for page 118
TECHNOLOGY AND THE SERVICES SECTOR 137 sector, which currently accounts for about two-thirds of private sector output and employment in the United States, has not been so fortunate, and it is safe to say that this now dominant segment of the economy has to bear greater responsibility for lagging national productivity growth. Unfortunately, for the economy as a whole, there may be small consolation in the competitive renewal taking place in manufacturing. If services fail to carry their weight, renewed factory sector vitality simply will not be enough to return the U.S. economy to its historical growth path. This is the essence of the United States' hidden competitive challenge. NOTES 1. Conversations with government officials indicate that the "other office machinery" category accounts for a very small portion of the total, in terms of both the stock and its rate of growth. 2. A similar conclusion can be supported by the analysis of Mark in his chaper (this volume). This analysis does suggest that measurement problems have worsened in commercial bank- ing in recent years; however, this industry accounted for only 11.4 percent of total finance sector output in 1986, hardly enough to alter the dramatic productivity disappointments for this segment as a whole. 3. This problem has plagued the productivity literature for years. See, for example, Baily (1981), Denison (1979), and Wolff (1985). 4. Estimates show that white-collar productivity has declined steadily over most of this decade to levels last seen in the late 1960s; this result is developed more fully in Morgan Stanley (1987). 5. Elsewhere it has been argued that today's computer revolution is comparable to the industrial revolution, the building of a nationwide rail system, and the advent of the factory assembly line all dramatic technological breakthroughs that sparked a process of structural change that paid off only after a considerable period of time; see Morgan Stanley (1986). 6. For an important start in this direction, see Strassman (1985). REFERENCES American Productivity Center. 1950-1985. Multiple Productivity Indexes. Baily, M. N. 1981. Productivity and the services of capital and labor. Brookings Papers on Economic Activities (1):1-65. Bowen, W. 1986. The puny payoff from office computers. Fortune 113(May 26):20. Denison, E. 1979. Explanations of declining productivity and growth. Survey of Current Business (August):l-24. Morgan Stanley. 1986. The productivity puzzle: Perils and hopes. Economic Perspectives (April 10). Morgan Stanley. 1987. America's technology dilemma: A profile of the information economy. Special Economic Study (April 22). Musgrave, J. C. 1986. Fixed reproducible tangible wealth in the United States: Revised estimates. Series in Survey of Current Business (January):51-75. Washington, D.C.: U.S. Department of Commerce. Neikirk, W. R. 1987. Technology in America, a five-part series. Chicago Tribune (July 5- 12). Schneider, K. 1987. Services hurt by technology. The New York Times (June 29):23.

OCR for page 118
138 STEPHEN S. ROACH Strassman, P. 1985. Information Payoff: The Transformation of Work in the Electronics Age. New York: Free Press. Wolff, E. 1985. The magnitude and causes of the recent productivity slowdown in the United States: A survey of recent studies. Pp. 29-57 in Productivity Growth and U.S. Competi- tiveness, W. J. Baumol and K. McLennan, eds. New York: Oxford University Press.