Click for next page ( 48


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 47
Growth of Services Employment in the United States RONALD E. KUTSCHER Any examination of the services sector in the U.S. economy immediately faces the definitional question What is to be included in "services"? There is no uniformly accepted definition. In this chapter the Standard Industrial Classification (SIC) system will be used to group industries. The SIC system broadly groups industries by whether their primary production is goods or services. The goods-producing industries most often include agriculture, min- ing, construction, and manufacturing. This naturally leads to the first and perhaps the broadest definition of services: i.e., all industries not producing goods, produce services. Such industries would include transportation, communications, public utilities, wholesale trade, retail trade, finance, insurance, real estate, services, and government. Private services-producing industries include all of the industries noted above except government. A more restricted definition of services includes only the industry group called services within the broader services-producing industry designation. This group includes personal, medical, business, and professional services. Each of these definitions is used in this chapter to examine specific aspects of employment growth in the U.S. services-producing sector, both past and prospective. Also, employment growth by occupation and by the educational requirements of these occupations is assessed. This chapter also considers why employment in the services-producing sector is growing so rapidly. Finally, in closing, the differing characteristics of the goods- versus services- producing portions of our economy are explored briefly. 47

OCR for page 47
48 RONALD E. KUTSCHER EMPLOYMENT GROWTH IN SERVICES 1972-2000 First, one must look at the key changes in U.S. employment patterns for each of the major services- and goods-producing industries from 1972 to 1986. Then, by using projections of employment to 2000, published recently by the Bureau of Labor Statistics (BLS), these industries can be examined for likely future changes. ~ Table 1 sets forth such data on both a historical and a projected basis. From these, Table 2 presents calculated past and projected shares of em- ployment growth. By using three different definitions, Table 2 demonstrates the significant role of the services-producing sector in employment growth in the United States. It is also interesting to compare the various periods in terms of the relative contributions of services- and goods-producing sectors. The 1979- 1986 period shows a declining employment in the goods-producing indus- tries, particularly manufacturing (see Table 11. Consequently, the services- producing sector provided more jobs than the entire net employment growth in the U.S. economy during that period. However, during the 1979-1986 period, the rate of employment growth in the overall economy was only 1.4 percent per year, about one-half the 2.6 percent annual rate from 1972 to 1979. There are two important factors about this slowdown to keep in mind: (1) the slowdown in growth of the overall labor force during the latter period and (2) the two recessions between 1979 and 1986, accompanied by pro- nounced foreign trade changes in the latter period. Projections to 2000, because of modest improvements expected in foreign trade and other factors, show no further declines in goods-producing employment (although an in- crease in construction employment offsets a projected decline in manufac- turing). Under each of the three alternative services sector definitions, the expected share of employment growth in services from 1986 to 2000 is lower than that from 1979 to 1986, but higher than that of the 1972-1979 period. The services industry group (the most narrowly defined designation) shows striking overall contributions to employment in both the 1979-1986 period and the 1986-2000 period. More disaggregated data for this specialized group indicate that business and medical services have led the rapid increases in this SIC group's employment growth. However, because of differing rates of measured productivity growth between the goods- and services-producing sectors, the output shift toward services is far less than the shift for em- ployment.2 Occupational Employment Changes, Past and Prospective The BLS projects that the U.S. economy will generate an increase of more than 21 million jobs between 1986 and 2000. Although this is a very high

OCR for page 47
49 8 of 1 - o ~ C) .= o ~ ~ o o m . Ct 3 to on e~ ~o 3 o oo - ~4 C~ Cr~ ~ _ oo s~ ~ ~ O ~ E~ o I_ CM o _ ~ _. o a~ 0 o _ oo U~ _ _ C~ C~ o o C~ _ _ _. CN ~ ~ _ _ _ C~ ~ ~o o _ _ - oo a~ o oo _ U~ o _ ~ _ ~ 1 ,_ ~ a~ 0 ~ - , _ ~ I=> ~ _ ~ _ 1 oo oo ~ ~ _ ~ 1 1 o^ C~ o ~ ~ ~ ~ ~ _ ~ _ V~ ~ o 1 ~ 1 1 1 cr ~ 0 ~ ~ ~ c~ crx o ~ ~ ~ o~ ~ o o o o oo ~ ~ o ~ ~ ~ - ~ ~ ~ ~ ~ ~ - - <~N oo ~ ~ r~ x ~ ~ ~ ~ ~ ~ ~ ~ c~ ~ ~ _ ~ ~ 00 ~ _1 r~ c~ ~ - ~ ~ ~ ~ o oo oo o ch ~ ~ oo - ~ ~ ~ - Oo ~ ~ c~ o M0 ~ ~D oo ~ c~ ~ o ~ ~ ~o ~ - ~ oo ~ ~ - oo oo cr kD ~ oo ~ ~ oo cM M ~ ~ ~ ~ ~ o <) ~ C~ 00 _ ~ 1 _ _ ~ _ _ ~ 1 - ') ] ~ co ~ ~ oo ~ o ~ ~ l ~ ~ ~ ~ ~ ~ l ~ _ ~ _ o _ 1 - t (~N oo O~ \~) ~t oo t - - oo t~ ~ ~ - ~ - oo ~ oo _ 1 1 N - ~ oo o ~ oo ~ oo - ) ~ - oo ~ ~ ~ t - - ~ o ~ c o - o ~ c~ 1 _ GN ~ o ~ 1 ~ _ ~ ~ ~ u~ ~ c~ ~ oo ~4 v~ ~ ~ ~ o oo - ~ ~ - ~ - ~ ~ ~ - ~ Oo ~) c~ oo - - oo o r~ ~ ~ - - ~n ~ c~ c~ o ~ - oo o ~ o o ~ ~ - t ~ ~ - ~ - oo r~ - - o - ~ ~ - ~ - - ~ - ;^ ; c) ~ x ~ ~ ce C) 0D ~ (L) ~ ~ ~ ~ ~ ~ = 0 ~ C ._ O - V- ~ ~ _ -~' z ~ ~ c) ~ - ~ ~: r~ o - a~ oo ~ ~ ~ - ~ ~ - oo ~ - c~ - ~o o ~o - M o ~ ~ ~ ~ - ~ - ~ ~ ~ - l ~ ~ ~o ~ - ~ ~ c~ o o ~ 3 s~ ce o ~ ~ z

OCR for page 47
so m ~- 1 o 3 ~ o C or O ~ ox o Doe' Up.=, Oo ~ -.- ~o 3 of Cal ~ Ct _ ~ oo o o C) Cal ._ _ X U: .= Ct Cal ~ == Cal o 3 rim ~ Two o o E~ Z O ~D ~ ~ ~ =\ . . . . . . . . o ~ o ~ ~ ~ ~ oo 0 ~ r - r ~ ~ C~ ~ O . . . . . . . O O O oo . . . . . . . O O ~ ~ oo oo O ~ ~ oo . . . . . . ~ 0 ~ ~ ~ r~ _ ~_ O . . . . . . _ O O . . . . . . O ~ ~D ~ _ oo O X ~ ~ ~ . . . . . . ~ ~ oo ~ ~ ~ _ ~ _ oo _ _ ~ ~ ~ . . . . . . ~ ~ C~ ~ ~ ~ ~ ~ _ oo ~ ~ ~ . . . . ~ oo O . . . . In oo ~ - ao . . . . o ~ u~ ~ oo ~ ~ - ~ ~ v~ ~ ~ - ~ o u~ o o ~ ~ oo oo ~ . . . . . . . . . . o o - o o o - o - - o ~ ~ c~ . . . . - o o - o 1 1 . . . . o o - o 1 1 u~ - o ~ - u~ ~ . . . . . . . o - - o - o 1 1 1 1 1 o . . . . . . . . . - ~ - - - o ~ 1 1 1 1 1 ~ ~ o ~ - c~ ~t . . . . . . . - ~ ~ - ~ o ~ -o -y-o ~ - oo . . ~ - . . . ~ - r~ oo . . . ~ oo ~ - - ~ u~ - . . . ~ ~ oo - - oo - ~ . . . o o 1 1 . . . . . o - ~ ~ ~ o ~D ~ oo . . . o o - 1 1 oo ~ ~ ~ ao ~ ~ ~ o - ~ ~ ~ ~ c~ o ~ ~ 1 1 3 ~ ~ 5 _ Ct ,3 O O =, ~ ~ _ ~ ~ ~ ~ ~ ~ ~ ~ O a~ V) C~ O o~ 3 _ o `s ~ ~ o ~ .) 50 s_4 o ~ ~ z cn m S: ._ C~ C~ - Ct Ct o ~_ rD Ct Ct o C~ - Ct x o a~ . ~ - Ct C~ au 5) C~ ._ ~n U, Ct - ~: o C~ C~ Q C) - - .= ~ . e~ V7 ~ ~V . _ U~ U) ~ Oo O .= C~ :u U, ~ ~S =i ii 0 z ~

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT IN THE U.S. TAl3LE 2 Share of Employment Growth Projected Sector 1972- 1979 1979- 1986 1986-2000 Services-producing sector 82.5 118.6 100.0 Private services-producing sector 66.0 110.6 ~ 92.0 Services industry group 29.1 60.3 49.8 Goods-producing sector 17.5 - 18.6 0.0 51 total number, it is only one-half the average annual rate of increase that occurred over the previous 14-year period, 1972-19863 (see Table 31. The projected change from 1986 to 2000 is more consistent with the slower growth rates of 1979-1986 than it is with those of the 1972-1979 period. Although overall growth in employment is slowing considerably, important changes are also occurring in the composition of employment by occupation. Faster than average growth has taken place in several occupational groups and is expected to continue in the 1986-2000 era. These include the follow- ing: (1) executive, administrative, and managerial workers; (2) professional workers; (3) technicians and related support workers; (4) sales workers; and (5) services workers. Employment declines have also occurred and are pro- jected to continue in two broad occupational groups: (1) private household workers and (2) farming, forestry, and fishing workers. Below-average growth is expected to continue in another occupational group: precision production, craft, and repair workers. Prospects for two occupational groups are expected to change in the future as compared to past periods. The operators, fabricators, and laborers occupational group declined from 1979 to 1986, after having increased from 1972 to 1979. This group is projected to increase again in 1986-2000, but by less than 3 percent. The occupational group of admin- istrative and support workers, including clerical, increased faster than the average from 1972 to 1979, then dropped to a below-average increase from 1979 to 1986. A slower than average increase is projected for 1986-2000 for this occupational group. Educational Content of Jobs Much has been written indicating that the changing occupational structure of U.S. employment will require a more highly educated work force. To see if the 1986-2000 occupational projections substantiate this view, the occu- pational clusters discussed in the previous section were divided into three groups defined by educational requirements. Group I included the occupa- tional clusters in which at least two-thirds of the workers in 1986 had one or more years of college. Group II included the occupational clusters in which the median years of school completed was greater than 12 and the

OCR for page 47
52 au a; o - ._ us: Ct on So So Blot C) o on 5 - c4o. us Ct o .~ Ct Cal o o sly m ;^ as O _ O lo- ~ an m ~ ~ - Coo - o . au in So o 00 00 Ch ~0 00 l Cal Cal 1 An _ _` Cal Ct V, o O So O _' _% Ct Cal o 00 cry . Ct . ~0 Ct C) C) o O ~ . ~ Sow ~ tO ~1 ~ ~ ~ t~ O C~ oo ~ ~ ~i ~i '~ ~ ~ ~ ~ 1 ~ 1 ~o ~ a~ ~ ~ ~ ~ C~ ~ ~ ~ ~ ~ ~ ~ ~ O t~ ~ ~ c~ ~ ~ 1 1 1 ~ ~ ~ ~ ~ ~ U~ O O oo ~ ~ ~ ~ ~ ~ ~ 0 ~ ~ ~ ~ ~ ~ ~ 1 1 1 ~ a~ oo ~ ~ ~ . . . . . . O ~ C~ ~ ~ C~ ~1 ~ ~ ~ - 1 ~1 O ~ ~ ~ ~ . . . . . oo U~ ~1 ~ 1 1 O ~ ~ ~ ~ ~ U~ ~ O C~ 1~ c~ O ~ ~C) ~ ~1 O ~ ~ ~ ~ ~ ~ C;- C~ ~ ~ U~ ~ ~ _ ~ ~ ~ _ ~ (~) ( - , oo ~ 0 ~ ~ ~ O C~ oo ~ ~ o U~ oo U~ ~ ~ ~ oo ~ U~ ~ ~ U~ O ~ ~ ~ ~ ~D ~ ~ c~ ~ ~ _4 _1 ~ ~ ~ _ - ~ e , r D ~ ~ ~ ~ ~ 3 ~ x, 0 e ,= E~ v: ~ ~ ~q ~ O ~ ~ ce O _ - ~ 0 C: ._ ~ u, ~ c, c~ o .= ~:5 - c-) .o ct ce ct ~ ;^ .- - ~ ;;> o = ~k 3= C~ ~ .O 3 C) .o o O ,D O O C~ ~: Cd C~ C~ ;^ C oo - '40 o C~ - Ct .s V, .. E~ O Z ~s ;^ Ct ~: Ct ~ ~ CQ o o C~ C.) Ct oo - E~ C) C~ C~ o C~ ._ C~ ._ Ct V) o - C~ C~ cr~ - o - ._ Ct ;> Ct o C) ; - 3

OCR for page 47
GROWTH OF SERVICES Ell~lPLOYMENT IN THE U.S. 53 proportion of workers with less than a high school education was relatively low. Group III included occupational clusters in which the proportion of workers having less than a high school education was relatively high. Since workers in any occupational cluster have a broad range of educational back- grounds, the three groups can only indicate educational levels of their pre- ponderant segment of workers; there are workers in each of the three groups at each of the educational levels. The distribution of past and projected employment by category within these clusters is shown in Table 4. These data indicate that the occupations requiring the most education, Group I, are likely to increase as a proportion TABLE 4 Employment in Broad Occupational Clusters by Level of Educational Attainment for 1986 and Projected 2000 Moderate Alternative Distribution (%) 1986 2000 - r ~ .OTAL ALL LEVELS 100.0 100.0 Level I OTAL 25. 1 27.3 Management and management related 9.5 10.2 Engineers, architects, and surveyors 1.4 1.5 Natural scientists and computer specialists .7 .8 Teachers, librarians, and counselors 4.4 4.3 Health diagnosing and treating specialists 2.3 2.8 Other professional specialists 3.5 3.7 ,~ . . . ~ecnn~c~ans 3.3 4.0 Level II TOTAL 40. 8 40.0 Sales workers 1 1.3 12.3 Administrative support, including clerical 17.8 16.7 Blue-collar-worker supervisors 1.6 1.5 Construction trades and extractive occupations 3.4 3.3 Mechanics and repairers 4.2 4.0 Precision production, and plant and systems workers 2.5 2.2 Level III TOTAL 34.0 32.7 Services workers 15.7 17.2 Agricultural workers 3.3 2.6 Machine setters and operators 4.5 3.6 Assemblers and other handworkers 2.4 1.9 Transportation and materials-moving workers 4.3 4.0 Helpers and laborers 3.8 3.4 NOTE: The data in Table 4 are the same as the 1986 and 2000 data in Table 3. However, the data in Table 4 are more disaggregated so comparison is not easy.

OCR for page 47
54 RONALD E. KUTSCHER of total employment from 25.1 percent in 1986 to 27.3 percent in 2000. The other two groups in which workers have less education are projected to decline as a proportion of total employment. The proportion of total employment is expected to decline most (from 34 percent in 1986 to 32.7 percent in 2000) in Group III, which requires the least education. The "services workers" group is in Group III, the only occupational cluster in this educational at- tainment group with a median of more than 12 school years completed. This group is increasing as a proportion of total employment, while all other occupational clusters within Group III are declining some by significant amounts. Conversely, in Group I all the clusters are increasing as a share of total employment, except one teachers, librarians, and counselors and there the decline is very small. Projections do not allow for future educational upgrading, such as that which has taken place in the past. Even with cor- rections for these factors, the year 2000's share of employment in the lower educational attainment group is expected to be larger than the share of em- ployment in the highest group. Thus, the shift in distribution of jobs by educational level could be char- acterized as (1) increasing the small but most highly educated job group, while (2) decreasing the larger but less educated job groups. Each of the three broad occupational clusters in Group I with the most highly educated workers is projected to grow more rapidly than the national average and thus increase its share of total employment. Collectively, the three broad occupational categories in Group I, which accounted for 25 percent of total employment in 1986, are expected to provide almost 40 percent of total job growth between 1986 and 2000. By contrast, factors such as office and factory automation, changes in consumer demand, and import substitutions are expected to lead to relatively slow growth, or a decline, for occupational groups requiring less education. Occupations such as administrative support, including clerical workers; farm- ing, forestry, and fishing workers; and "operators, fabricators, and laborers" are examples. "Services workers," a category that is expected to grow at a faster rate than total employment and account for more of the total growth in employment than any other broad occupational group, is an important exception to the general trend, since the group?s educational attainment is currently not among the highest. In broad terms, therefore, the economy is expected to continue generating jobs with higher educational requirements. This is clearly an important context in which to assess future job develop- ments. Many- but not all of the projected future shifts are consistent with past trends in the occupational structure of U.S. employment. Blacks, Hispanics, and Women Job opportunities for individuals or groups of workers are determined by a number of factors relating to the job market and the particular characteristics

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT Ilk! THE U.S. 55 of workers. Consequently, in developing projections of employment by in- dustry and occupation, the BLS does not develop projections of the demo- graphic composition of future jobs. However, data on the current demographic composition of jobs can be used in conjunction with projected changes in employment to determine the possible implications of those projections. For example, one can see how future patterns of job growth compare with the current pattern of jobs held by blacks and Hispanics. Blacks and Hispanics constituted approximately 10 and 7 percent, re- spectively, of employment in 1986. Although members of these two groups were employed in virtually every occupation, they were more heavily con- centrated in some occupational clusters. The occupational clusters in Table 5 are listed in decreasing order of their projected growth rates. In general, these data show that both blacks and Hispanics are found in a greater pro- portion of occupations projected to decline or grow more slowly than in those that are projected to increase rapidly. These occupational clusters generally require the least amount of education and training, whereas those projected to grow fastest require the most education and training. The only exception is the services workers cluster, which is growing rapidly. Normally, occupations having the fastest growth rates will offer better opportunities for employment. For blacks and Hispanics to improve their labor market situation, they must be able to take advantage of those oppor- tunities. Labor force projections developed by the BLS indicate that blacks and Hispanics will constitute approximately 17 and 29 percent, respectively, of the total growth in the labor force. Because the fastest growing occupations are those in which a high percentage of workers currently have postsecondary education, improvements in educational attainment are likely to be an im- portant factor in blacks and Hispanics being able to take advantage of these areas of rapidly growing opportunity. The relative proportion of women in the different occupational clusters varies considerably (see Tables 5 and 61. In general, however, women have relatively high proportions of employment in the faster growing occupations, with two exceptions. Women's share of employment as natural scientists and computer specialists is currently low. As engineers, architects, and surveyors, it is even lower (7 percent). Women tend to represent smaller proportions of the occupations projected to decline or grow slowly, with one major exception the "administrative support including clerical" category. This is a slowly growing occupational group in which women currently hold more than 80 percent of the jobs. EXPLANATIONS FOR THE RAPID GROWTH OF EMPLOYMENT IN THE SERVICES INDUSTRIES4 So far, this chapter has only described the services industries' rapid em- ployment growth patterns and characterized them by industry, occupation,

OCR for page 47
56 RONALD E. KUTSCHER TABLE 5 Projected 1986-2000 Growth Rate, and Percent of Total Employment in 1986 Accounted for by Blacks, Hispanics, and Women Percentage of Total Employment in 1986 Projected Change, Occupational Cluster 1986-2000 (%) Blacks Hispanics Women TOTAL, ALL OCCUPATIONS 1 9 1 0 7 44 Natural scientists and computer specialists 46 6 3 31 Health diagnosing and treating occupations 42 6 3 67 Technicians 38 8 4 47 Engineers, architects, and surveyors 32 4 3 7 Services workers 31 17 9 61 Marketing and sales workers 30 6 5 48 Managerial and management-related workers 29 5 4 43 Other professional workers 26 7 4 43 Construction trades and extractive workers 18 7 8 2 Teachers, librarians, and counselors 16 9 3 68 Mechanics and repairers 15 7 7 3 Administrative support, including clerical 11 11 6 80 Transportation and materials-moving workers 10 14 8 9 Helpers and laborers 6 17 11 16 Precision production, and plant and systems workers 4 9 9 23 Machine setters and operators 4 16 13 42 Assemblers and other handwork occupations 4 13 11 38 Agricultural forestry, and fishing workers 5 ~ it I b

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT IN THE U.S. TABLE 6 Employment by Occupation in 1986 and Projected 2000 Alternatives (percent distribution) Projected 2000 Alternatives 57 1986 Low 100.0 100.0 Moderate High 100.0 100.0 TOTAL, ALL OCCUPATIONS Managerial and management- related workers Engineers, architects, and surveyors Natural scientists and computer specialists Teachers, librarians, and counselors Health diagnosing and treating . . spectra fists Other professional specialists , ~ . . . . ec Cans Marketing and sales workers Administrative support, including clerical Services workers Agricultural, forestry, and fishery 9.5 10.2 10.2 10.3 1.4 1.5 1.5 1.6 0.7 0.8 0.8 0.8 4.4 2.3 3.3 3.3 11.3 17.8 15.7 workers Blue-collar-worker supervisors Construction trades and extractive workers Mechanics and repairers Precision production, and plant and systems workers 2.8 Machine setters and operators 4.5 Assemblers and other handworkers Transportation and materials- moving workers Helpers and laborers 3.2 1.6 3.6 4.2 2.4 4.3 3.8 4.4 2.8 3.6 3.8 12.3 16.6 17.2 2.6 1.5 3.4 4.0 2.1 3.5 1.9 4.0 3.4 4.3 2.8 3.7 4.0 12.3 16.7 17.2 2.6 1.5 3.3 4.0 2.2 3.6 1.9 4.0 3.4 4.3 2.8 3.5 4.0 12.2 16.6 17.1 2.5 1.5 3.4 4.0 2.2 3.6 2.0 4.0 3.4 and period. This section presents several possible explanations for this em- ployment growth. These explanations focus on the producer-services indus- tries a subset of the services industry group described in the first section.5 However, this assessment may provide a proxy for a much broader review, as one possible explanation of the overall employment shifts from the goods- producing to the services-producing industries. Factors in Growth of the Producer-Services Industries When using a framework from input-output analysis, several factors seem to contribute importantly to the rapid growth of producer-services industries.6

OCR for page 47
65 TABLE 9 Sources of Occupational Change in Manufacturing Employment, 1983-1986 (numbers in thousands) Employment Change Explained by Total Composition of Manufacturing Actual Employment Industry Staffing Occupation Change Growth Mix Patterns Othera Managers and administrative workers 131 39 Professional, paraprofessional, and technical 9 81 workers 239 74 36 1 18 1 Clerical and administrative support workers 49 79 17 44 -2 Services occupations - 23 12 -1 33 -1 Sales workers 70 20 5 37 2 Production and related workers 207 450 - 66 166 - 12 aResidual effects. reveals that the number of workers in professional, paraprofessional, and technical occupations would have grown by 74,000 rather than the actual growth of more than 239,000. Finally, the number of workers in "services occupations" within manufacturing would have increased by 12,000 if the only change were in the level of manufacturing employment. Employment in such occupations actually declined by 23,000. Again, another factor was causing employment of services workers to lag behind total employment growth in manufacturing. Industrial Composition, 1983-1986 Another potential source of employment growth among clerical occupa- tions could be the changing mix of manufacturing industries. Here, industry mix means the number of workers employed in a particular industry as a percentage of total manufacturing employment. For example, the motor ve- hicles industry accounted for 4.1 percent of the manufacturing workers in 1983 and 4.6 percent in 1986. The construction machinery industry accounted for 1.4 percent of all manufacturing workers in 1983 and 1.2 percent in 1986. What would the change in clerical employment between 1983 and 1986

OCR for page 47
66 RONALD E. KUTSCHER have been if industry mix alone had changed while both the level of man- ufacturing employment and the proportion of clerical workers within the individual manufacturing industries had not changed? The difference between this employment estimate and the number of clerical workers in total man- ufacturing in 1983 measures the impact of the changing mix of manufacturing industries. This analysis uses employment trends for 143 manufacturing in- dustries to measure the changes in industry mix. The industries are at the three-digit level of the SIC. The changes in industry mix themselves are the results of other forces including productivity trends among the various industries, changes in the composition of final demand, and changes in business practices. Both the level of total manufacturing employment and changes in staffing patterns among the individual industries are held constant in this step. Analysis on this basis shows that the changing mix of manufacturing industries alone caused the number of clerical workers in total manufacturing to increase 17,000 between 1983 and 1986. It also explains 9,000 of the actual 131,000 increase in the number of managers and 36,000 of the 239,000 increase in the number of professional, paraprofessional, and technical work- ers. Industry mix changes alone would have caused employment among services workers to decline by 1,000. Industry Staffing Patterns, 1983-1986 A final potential explanation for the employment growth of clerical workers in manufacturing from 1983 to 1986 could be the changed staffing patterns among the detailed manufacturing industries. Staffing patterns are the pro- portion of employment accounted for by clerical occupations within a par- ticular industry. What would the change in occupational employment have been if staffing patterns alone had changed while both total manufacturing employment and the composition of industries had not changed? Holding these two elements constant implicitly holds final demand composition, busi- ness practices, and industry productivity constant between 1983 and 1986. The difference between this estimated employment and the actual 1983 em- ployment isolates the effect of changing staffing patterns. If staffing patterns among the detailed manufacturing industries were the only change for this period, total employment of clerical workers within manufacturing would have declined by 44,000 (see Table 91. On this basis most manufacturing industries employed proportionately fewer clerical work- ers in 1986 than in 1983. However, for clerical occupations, the growth in total manufacturing employment more than offset changes in staffing patterns within the detailed industries resulting in the observed net increase of 49,000 clerical workers over the 1983- 1986 period. Similarly, if staffing patterns were the only change in 1983-1986, em-

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT IN TlIE U.S. 67 ployment in (1) managerial and (2) professional, paraprofessional, and tech- nical occupations would have grown by 81,000 and 118,000, respectively. However, the data show that manufacturing industries employed proportion- ately more workers in these occupations over the three years. Combining the Three Factors The three employment estimates can now be combined to help understand the changes occurring in clerical employment within manufacturing. The number of clerical workers would have increased by 79,000, based on total manufacturing employment growth alone. The number would have increased 17,000, based on changing industry mix alone. The number would have declined 44,000, based on changes in industry staffing patterns alone. As noted, the actual change was 49,000. The decline isolated by changing staffing patterns alone was the only estimate that indicates potential unbun- dling. Similar analyses of services workers employed in manufacturing indicate that some unbundling could be occurring. The number of services workers did decline by 23,000 between 1983 and 1986. The changes in staffing patterns among the detailed industries alone would have caused a decline of 33,000; however, this effect was reduced by the total employment change in manufacturing, whereas the effects of changing industry mix were slightly negative (1,000 jobs). Estimates for professional, paraprofessional, and technical occupations yield a different picture. The three effects were all positive. Changes in the level of manufacturing employment alone explained 44,000 of the actual 239,000 increase in the number of these workers; changes in industry mix alone explained 36,000; and changes in staffing patterns alone explained 118,000. From these three positive effects, one can conclude that unbundling of professional-type activities did not occur. Impact on Producer-Services Industries What does the decline of 44,000 in employment among clerical occupa- tions that can be explained by changing staffing patterns mean? One possi- bility is potential unbundling. Individual manufacturing industries might be employing proportionately fewer clerical workers because of unbundling. However, available data do not permit isolation of the causes of changing staffing patterns; the estimate represents the net effect of many other factors, such as technology and business cycle shifts, as well as potential unbundling. If unbundling were the sole explanation for the changes in staffing patterns, the estimate of 44,000 would be equivalent to about 2.8 percent of the total employment growth of producer services. The total number of workers in

OCR for page 47
68 RONALD E. KUTSCHER producer services increased 1,554,000 between 1983 and 1986. For the 44,000 estimate to be a result of unbundling would require all the producer- services activities related to these jobs to be simply transferred from man- ufacturing industries. Although we do not know whether the estimate of 44,000 was a direct transfer of activities from manufacturing to producer services, we do know that unbundling did not alter the staffing patterns in the producer-services industries. According to data from the Current Population Survey (CPS), the proportion of producer-services employment accounted for by clerical work- ers changed very little between 1983 and 1986.7 This demonstrates strongly that all activities within producer services grew, not just clerical-type activ- ities. The employment of services workers within manufacturing declined 33,000 because of changes in staffing patterns alone. If these workers were reab- sorbed in producer services, this estimate would account for 2.1 percent of the total employment growth. According to the above calculations, unbundling is not even a potential explanation for the trends in managerial, professional, paraprofessional, and technical occupations within the producer-services industries. Changes in staffing patterns alone caused employment in these occupations to increase within manufacturing industries. Conclusions About Unbundling From the above evidence it can be concluded that unbundling has been a very small factor in the employment growth of producer services. Occupa- tional employment trends within manufacturing show that unbundling is not occurring for the broad managerial, professional, and technical occupations within manufacturing. Relative employment in these occupations is actually increasing. Unbundling is potentially a factor that explains some trends af- fecting clerical and services occupations within manufacturing. However, employment shifts due to changing staffing patterns could account for only a small proportion of the total employment growth in producer services. Unbundling could, of course, be occurring in individual firms, because trends noted here were the net effects of all the individual fibs at the industry or sector level. Thus, considerable unbundling at one firm could have been offset by employment growth in the same occupations at other firms. WHY DO BUSINESSES DEMAND MORE PRODUCER SERVICES? Why have businesses demanded more producer-services' inputs to make their products? Why has contracting out increased? Changing business prac- tices explain a large proportion of the producer-services sector's output growth.

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT IN THE U.S. 69 Since these changes are not due to unbundling, the increased contracting must be for new or additional services. The remainder of this section lists some suggested explanations for increased contracting out, but the merits of each explanation are not reviewed. Economies of Scale in Information The growth of employment in producer services may be a response to increasing demands for information (Browne, 19861. Computer and data processing technologies allow services units to spread the high investment and development costs of their systems over many users, thus providing economies of scale. Similarly, management and busi- ness consulting services, engineering and architectural services, and other producer services have spread the costs of acquiring sophisticated technical knowledge about demographics, economics, marketing, engineering, and other fields over many users. Specialized Corporate Services The increasing number of large companies and conglomerates may have created a demand for specialized producer services (Cocheba et al., 1986; Stanbeck, 19791. According to this argument, a typical 1980s corporation is often diversified into many fields or industries, including manufacturing, retail, transportation, or personal services activi- ties. A 1960s corporation might have been involved in fewer fields or in- dustries. Thus, top managers in today's more complex enterprises must increasingly rely on experts in sophisticated producer-services units (such as business management and consulting firms) to ensure efficient operations. Government Regulations and Laws Some argue that there are more lawyers, accountants, and other technical experts today than in the 1950s and 1960s because of the number and complexity of laws passed by Congress, state legislatures, and city councils.8 New regulations and laws often present spe- cialized problems in dealing with banking, construction, environment, labor relations, safety, transportation, and other services issues. International Trade The growth of producer-services industries may be explained in part by the fact that such services have reached a scale and complexity where they can be exported themselves (U.S. Congress, Office of Technology Assessment, 19861. Resistance to Innovation One argument is that services-producing industries and workers resist innovations over time (Bailey, 1986; Baumol, 1967; Bau- mol et al., 1985; Thurow, 1981, 1985~. According to this explanation, the economy can be divided into two industries: stagnant and progressive. Stag- nant industries resist innovations; progressive industries rapidly incorporate change. If relative demands were unchanged over time, the stagnant industries

OCR for page 47
70 RONALD E. KUTSCHER would slowly absorb more and more of the economy's inputs. If this argument is valid, lawyers (and legal firms), janitors (and services to dwellings and other buildings), guards, accountants, and computer programmers should be performing their tasks the same way they did 10 to 20 years ago. Durability and Commonality One suggestion is that services are expanding because there is little opportunity to lower unit costs through repeating pro- duction. For example, one transaction often has little in common with the next in legal, automobile repair, or banking services.9 In contrast, the pro- duction of goods such as automobiles or wheat is more likely to be replicable in large quantity, thus allowing capital substitution and experience curve effects to reduce unit costs. Finally, there are serious problems of data definitions. How does one define output units and quality in services? Are goods really more durable and portable than services, thus creating greater long-term value in their customers' hands? Computer software programs are both durable and por- table. Some producer services such as temporary help or janitorial services are neither durable nor portable. Legal, engineering, or management con- sulting is highly portable by telecommunications and air travel. The benefit of a lawyer's or an engineer's work is as durable as the output of most manufacturing industries. WHY THE CONCERN ABOUT A SERVICES ECONOMY? Frequently one hears the argument that our economy cannot endure as just a services-producing society or that it cannot exist if people merely shine each other's shoes or take in each other's laundry. The reasoning for this concern, although never explicitly stated, seems to be as follows: (1) the production of goods is better because it generates more income than the production of services; (2) the production of services can take place only after the demands for goods have been satisfied; and (3) the country cannot maintain its wealth or indeed survive by producing only services. Why is the production of goods viewed as good, whereas the production of services is viewed as bad, poor, or at least less good? The argument that the production of goods is more income-generating than the production of services implies that the production of goods valued at one dollar generates more useful income than the production of services valued at the same price. Is the economy better off because a consumer spends $100 on skis rather than spending $100 on ski lift tickets the former being goods, the latter services? This seems doubtful. If consumers spend their own dollars to maximize marginal gains to themselves, the effect of the $100 depends not so much on what the consumer spends it for, but on how many times it is respent. On average in our economy, about two-thirds of spending ulti-

OCR for page 47
CROWTlI OF SERVICES EAiIPLOYMENT IN THE U.S. 71 mately translates into wages and benefits payments to wage earners. This average, however, varies considerably by industry. One basis for the argument that the production of goods is better than the production of services may rely on the assumption that goods last longer than services and, therefore, can be used again and again. Certainly, that is true of some goods, but other items currently classified as the products of a goods-producing industry (such as orchids, newspapers, or beer) are very short-lived. The contrary assumption is that a service is something we con- sume immediately which, therefore, is not long-lasting. That certainly is not true for the products of services-producing industries such as life insurance, computer programming, estate planning, or funeral services the latter being the most lasting of all. Problems of Classification It is not always easy to determine why an item is classified as it is in the industrial classification system currently in use. Many times the groupings are relatively arbitrary. Newspapers are classified in manufacturing, a goods- producing industry, whereas computer programming is often classified as a services-producing industry. If, in the production of goods, accountants or computer programmers are directly on the payroll of a manufacturing firm, their work is classified as contributing to the production of goods. However, if the accountants' or computer programmers' services were contracted for by the same manufacturing firm, they would be classified as services. The same holds for other activities that can be performed within a firm or con- tracted out, such as legal services, word processing, or building maintenance, to mention only a few. Perhaps the notion that goods production is more income-generating than services production may be linked to the presumption that investment an activity closely associated with the accumulation of some types of goods- may be more wealth-enhancing than current consumption. This would gen- erally be true of buildings, machine tools, or construction machinery in- vestment goods that can be more wealth-enhancing to the economy in the long run than can the production of many specific services. However, many other goods (e.g., processed foods, soft drinks, clothing, newspapers, and magazines) are current consumption items, not contributing much to wealth accumulation. By contrast, education the product of a services-producing industry enjoys perhaps the greatest long-term, wealth-producing potential of any single industry. The notion that goods production is of higher value than the production of services may also be based on a confusion over end products. Some suggest that services can only achieve permanent value by becoming inputs to goods

OCR for page 47
72 RONALD E. KUTSClIER production. Yet goods themselves can be either end products (such as re- frigerators and automobiles) or inputs to later stages of production (such as plywood and sheet steel). Similarly, services such as advertising can be inputs (to either goods production or the production of other services), or they can be final products used in the form produced (such as recreation a round of golf or an opera performance). Services A Prime Mover or a Dependent? Can a service be a prime economic mover and exist without prior goods production? Or must an economy have goods production first in order to support the services that follow? There is no reason why medical, legal, entertainment, education, banking, or many other services should require prior goods production to be effective. Such services activities require only that there be income for individuals to buy the services. The source of that income can be employment in either a goods- or a services-producing in- dustry, or other income such as rent, interest, savings, or dividends. Goods- producing industries (such as household furniture) likewise require only that individual buyers have the necessary income, regardless of source; hence, no differentiation appears. What about services-producing industries that primarily serve other busi- nesses, as advertising does? If the industry served is a goods-producing industry, the goods-producing industry is clearly the prime mover. However, many goods-producing industries manufacture items, such as logs, fasteners, steel, newsprint, or chemical feedstocks, that are inputs to other goods- or services-producing industries and thus depend equally on those other indus- tries. By contrast, hospitals, which require some inputs (such as medical instruments) from goods-producing industries, are also really economic prime movers, even though they are classified as a services-producing industry. Again it can be argued that no meaningful differentiation exists in an eco- nomic sense. The prime mover notion may be related to another argument: that goods production requires more primary inputs and produces more repair and main- tenance tasks later, whereas services do not create such secondary conse- quences. Closer examination of goods-producing and services-producing industries reveals that although some goods (such as automobiles) do require many inputs and future servicing activities, this is not true of many other goods (such as books or magazines). Some services (electricity generation or air transportation) are very complex and require many inputs as well as processing, monitoring, and later maintenance. Again, the differentiation between goods-producing and services-producing industries seems arbitrary. Can the country survive producing only services? There is no basis for the belief that goods production is more necessary to an economy, more

OCR for page 47
GROWTH OF SERVICES EMPLOYMENT IN TlIE U.S. 73 income-generating, or more wealth-enhancing than services. In the past it may have been thought that only goods could enter into foreign commerce, but today services do also. Although goods production is useful, desirable, and even vital from a national defense viewpoint, it does not follow that the economy exists only because of goods production. An economy could sur- vive, at least in theory, producing only services- if that economy's com- petitive advantage was in the production of services. This may be the current position of the United States with respect to a number of services and one of the factors contributing to the very rapid growth of employment in some of the services-producing industries in this country. Thus, the arguments typically advanced for considering services industries as inferior to goods-producing industries seem not to have much validity. An economy with a mixture of goods and services production is unques- tionably stronger. The current mix of goods and services in our economy is a response to what purchasers are demanding and not a cause for the alarm it seems to generate. There is no good argument for advocating a stronger shift to services-producing industries or for retarding that shift in some ar- bitrary way. SUMMARY Services have contributed significantly to job growth in the U.S. economy and are likely to continue to do so. The growth of the services economy should, however, not be cause for widespread concern. It is largely the growth of demand and employment in the services-producing industries and lesser demand and growth for U.S. goods-producing industries that has caused the large relative shift toward services. One element of this shift is the real growth in producer services. There are several explanations for the above-average growth of the pro- ducer-services industries. The most important explanation for this growth is a structural change in the way our economy produces goods and services. Based on the available evidence, the unbundling of services activities from manufacturing operations accounted for only a small portion of the total producer-services industries' above-average growth. However, unbundling in individual fibs could have been offset by employment growth for these same activities in other firms. Also, in individual unbundling situations, there may be displacement as opposed to transfer of individual workers. The impact of such unbundling on individuals must always be recognized. Finally, profound implications result from these and other trends in the type of employment and educational preparation required. Job requirements are moving over the long run toward those jobs with higher educational requirements. All factors point toward the continuing need for a better ed-

OCR for page 47
74 RONALD E. KUTSCHER ucated work force. Our data strongly indicate that those who are not under- taking such art education will be increasingly at risk in Me future job marketplace. NOTES 1. Projections of employment for 2000 are taken from Personick (1987). 2. This point is explored further in Kutscher and Personick (1986, pp. 3-13). 3. Occupational projections are taken from Silvestri and Lukasiewicz (1987). 4. This section draws very heavily on Tschetter (1987). 5. The producer-services industries are part of the services industry grouping (narrowly de- fined). This encompasses advertising, computer and data processing services, personnel supply services, management and business consulting services, protective and detective services, services to dwellings and other buildings, legal services, accounting and auditing services, and engineering and architectural services. This section of the chapter uses numerous data sources. When describing employment trends in the nonagricultural industries and producer services, the Current Employment Statistics (CES) are used. When describing occupational employment for manufacturing industries, the Occupational Employment Survey (OES) is used. An alternative industry data source is the BLS Current Population Survey (CPS), which is compiled from household interviews. There are important differences among the surveys. The CPS counts the number of persons who are employed; the CES and OES count jobs. Because of this difference, a person holding two or more jobs would be counted two or more times in the CES and OES, but only once in the CPS. Another difference is that the CPS includes estimates of self-employed workers, unpaid family workers, and wage and salary workers; the CES and OES include only wage and salary workers. 6. When viewed in a descriptive sense, as a system of data classification and accounting, input-output is generally acceptable to all economists. However, here input-output is used as a theory of production, with the assumption that the coefficients comprise a set of technological parameters in a linear homogeneous production function with fixed proportions among the various inputs. For another example of the analysis used in this chapter, see General Accounting Office (1987). The input-output tables used in this chapter are developed by the BLS and based on tables prepared by the U.S. Department of Commerce, Bureau of Economic Analysis. For a description of input-output tables, see U.S. Department of Commerce (1984). In the Department of Commerce's input-output tables, industrial purchases of producer services are usually based on occupational employment patterns: total receipts of a particular service are usually distributed to the purchasing industry based on the number of persons in a particular occupation in the purchasing industry. See U. S. Department of Commerce, Bureau of Economic Analysis (1980). 7. The analysis presented in this section was repeated with CPS data. The results from the CPS analysis were virtually identical, i.e., some unbundling might be occurring but it would explain very little of the employment growth in the producer-services industries. The analysis presented in this chapter is better because the OES is based on a substantially larger sample. 8. There is no specific proponent of this explanation, but it is reviewed in many studies of producer services. See McCrackin (1985). 9. For a discussion of the Producer Price Index and issues concerning price indices, see Bureau of Labor Statistics (1982).

OCR for page 47
GROWTH OF SERVICES EhIPLOYlilENT IN THE U.S. REFERENCES 75 Bailey, M. N. 1986. What has happened to productivity growth? Science (October):443-452. Baumol, W. J. 1967. Macroeconomics of unbalanced growth. American Economic Review (June):415-426. Baumol, W. J., S. A. B. Blackman, and E. N. Wolff. 1985. Unbalanced growth revisited: Asymptotic stagnancy and new evidence. American Economic Review (September):806- 817. Browne, L. E., 1986. Taking in each other's laundry: The service economy. New England Economic Review (July/August):20-31. Bureau of Labor Statistics. U.S. Dept. of Labor. 1982. BLS Handbook of Methods. (Bulletin 2134-1) Washington, D.C.: U.S. Government Printing Office. Cocheba, D. J., R. W. Gilmer, and R. S. Mack. 1986. Causes and consequences of slow growth in the Tennessee Valley's service sector. Growth and Change (January):51-65. General Accounting Office. 1987. Bureau of Labor Statistics employment projections: Detailed analysis of selected occupations and industries. Report GAO/OCE-85-1 (April 15). Henson, R. C. 1985. Coping with fluctuating work-force requirements. Employment Relations Today (Summer):149-156. Kutscher, R., and V. A. Personick. 1986. Deindustrialization and the shift to services. Monthly Labor Review (June):3-13. Mangum, G., D. Mayall, and K. Nelson. 1985. The temporary help market: A response to the dual internal labor market. Industrial and Labor Relations Review 38(4):599-611. McCrackin, B. H. 1985. Why are business and professional services growing so rapidly? Economic Review (Federal Reserve Bank of Atlanta) (August): 15-28. Personick, V. A. 1987. Industry output and employment through the end of the century. Monthly Labor Review (September):30-45. Piore, M. J. 1986. Perspectives on labor market flexibility. Industrial Relations XXV(2): 146- 166. Silvestri, G. T., and J. M. Lukasiewicz. 1987. Projections 2000: A look at occupational employment trends to the year 2000. Monthly Labor Review (September):46-63. Stanbeck, T. M., Jr. 1979. Understanding the Service Economy. Baltimore: Johns Hopkins University Press. Thurow, L. C. 1981. Strengthening the Economy. Washington, D.C.: Center for Democratic Policy. Thurow, L. C. 1985. Pruning our white-collar ranks: A key to productivity. Technology Review (November/December): 14- 15. Tschetter, J. 1987. Producer services industries: Why are they growing so rapidly. Monthly Labor Review (December):31-40. U.S. Congress, Office of Technology Assessment. 1986. Trade in Services: Exports and Foreign Revenues Special Report. OTA-ITE-316. Washington, D.C. (September). U.S. Department of Commerce. 1984. The input-output structure of the U.S. economy, 1977. Survey of Current Business (May):42-79. U.S. Department of Commerce, Bureau of Economic Analysis. 1980. Definitions and Con- ventions of the 1972 Input-Output Study (July 4). Washington, D.C.: U.S. Government Printing Office.