Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
1 Technological Charlge and Women Workers In He Office The employment effects of technological change have again become an issue of public concern. In particular, questions have been raised about future em- ployment opportunities for women workers because many of the new develop- ments especially in telecommunications and microprocessing have already dramatically altered work involving information processing, an area of work dominated by women. Recent research suggests that automation in the clerical sector is altering the numbers and types of jobs available, the nature of jobs and their geographic location, working conditions, and career opportunities. A ma- jor quantitative study recently predicted that the number of clerical jobs in the economy will decline by 1995, not only relatively but also absolutely (Leontief and Duchin, 1984), although this prediction is controversial. Earlier waves of automation in office work also raised fears of unemployment, yet clerical, pro- fessional, technical, and managerial employees have increased their share of the labor force in every decade since at least 1940 (Hunt and Hunt, 1985a). Is the present situation different? For the past several decades, as employ- ment in the U. S. economy has continued to shift from agriculture and manufac- turing to services, the latter sector has absorbed large numbers of new workers, particularly women and young people. Technological developments have con- tributed to the introduction of new products and services that provide new kinds of jobs. Sectoral shifts in demand, accompanied by economic growth, histori- cally have produced better opportunities and wages for some workers. Two striking examples for women workers are the nineteenth-century shift of young women workers from agriculture to manufacturing and the recent shift of black women from agriculture and domestic service to clerical positions. Economy- wide, technological change is credited with contributing to productivity 1
2 COMPUTER CHIPS AND PAPER CLIPS growth, standard of living increases, and generally higher wages. In the last decade, service industries have adopted new strategies, reorganized work, and introduced new machines to increase productivity. If productivity gains in- crease substantially in the service sector, will it continue to generate enough jobs to absorb all the workers seeking employment? What kinds of jobs will they be? Will they offer safe and healthy working conditions, job stability, good wages, and opportunities for advancement? Both future levels of employment and the "quality" of work are closely related to the introduction and use of new technologies. This report identifies and analyzes the effects of technological change on both the quantity and quality of women's paid employment. It focuses particu- larly on innovations in information processing and telecommunications and their applications in offices past, present, and likely future. The interest of the Panel on Technology and Women's Employment is in determining whether women may be differentially affected by these innovations relative to men and, if so, how; to what extent women of different ages, educational backgrounds, and race and ethnic groups may be affected differently; and what factors may be shaping these effects. The "machine" aspects of technological change in office automation have captured the public imagination, but they are only part of the picture. They have drawn the attention of scholars, workers, employers, and policy makers to a much more pervasive set of changes in the organization of work, its geographic location, and the characteristics of workers. Hence, when the report examines employment effects resulting from technological change in microprocessing and telecommunications (information technology or "telematics"), it includes the wider changes in work organization and composition of the labor force as well as the more immediate effects of the new machines themselves. In seeking to determine the effects of technological change on women's paid employment, the report focuses on new technologies in clerical occupations, both because the technical developments in these occupations are dramatic and appear to have been widely implemented and because so many women work in these jobs. Clerical occupations are a diverse group, ranging from the "typical" office ones of secretaries, typists, and file clerks to bill collectors, interviewers, telephone operators, dispatchers, mail carriers, insurance adjusters, bank tell- ers, and proofreaders. Table 1-1 displays the number of workers in clerical There are many definitions of clerical occupations; conceptually, clerical work is not a clear category, and the boundaries are difficult to delineate. This report generally uses the 1980 census classification of "administrative support workers" as its definition of clerical workers, but notes significant changes that have occurred in the government classifications. The most typical clerical occupations, those that would be regarded as clerical workers in any classification system secretaries, stenographers, typists, file clerks, bookkeepers, accounting and financial clerks, and general of flee clerksaccount for 8.9 million, or more than half of the 16.8 million workers identi- fied as administrative support workers in the 1980 census.
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 3 occupations in the 1980 census (16.9 million), with the number and percent female in each. The five clerical occupations with the largest number of work- ers were secretaries (3.9 million), bookkeepers and accounting clerks (1.8 mil- lion), general office clerks (1.6 million), typists (0.7 million), and general office supervisors (0.6 million). For women clerical workers, the first four largest occupations were the same, but general office supervisors (0.4 million) were in seventh place, below receptionists (0.5 million) and bank tellers (0.5 million). (Cashiers, another large occupation- 1.7 million and about 85 per- cent female often thought of as clerical workers, were reclassified in the 1980 census as sales workers.) The 13.0 million women clerical workers represented more than three-fourths of all clerical workers and more than one-third of all employed women workers in 1980. In reviewing data and research on employment trends and on the develop- ment and implementation of new information technologies in several clerical occupations and sectors, the report seeks to identify both the positive and nega- tive aspects of this wave of technological change and to reduce the uncertainty surrounding estimates of the likely size, incidence, nature, and timing of the effects. The report also considers how employers' decisions to adopt and imple- ment new technologies and workers' participation in these processes can affect outcomes. Finally, the panel identifies and recommends private and public policies that can alleviate negative effects, promote improvements in women's employment opportunities, and produce a more humane work environment for all. The remainder of this chapter first defines technological change and de- scribes some recent and anticipated changes in information technologies. It then discusses the measurement and characteristics of technological change and explores in a schematic way the economic and social context of technological change and the employment effects that can result. It next describes women's employment situations and explains why the panel expects differential effects from technological change for women workers. Chapter 2 discusses selected examples of technological change in communi- cations and information processing and considers especially its effects on wom- en's employment levels, absolutely and relative to men, and on the content and quality of work. The effects discussed include indirect, unforeseen, and uneven effects. These topics are examined in several critical types of women's employ- ment: communications work and clerical work in information and data process- ing, with a comparative look at retailing and nursing. The chapter also con- siders briefly how workers have responded to technological change in these cases and how managers' and workers' interests have differed and coincided. Chapter 3 analyzes the current and future effects of technological change on the levels of employment and the structure of occupations and considers changes in both the supply and demand of workers. It reviews projections made by the Bureau of Labor Statistics and presents the panel's estimates of the most
4 COMPUTER CHIPS AND PAPER CLIPS TABLE 1-1 Employment of Administrative Support Occupations in 1980 Total Percent Female Occupation Employment Female Employment Administrative support occupations 16,851,398 77.1 12,997,076 Supervisors, administrative support occupations 1,056,710 47.1 497,668 Supervisors, general office 631,337 56.1 354,410 Supervisors, computer equipment operators 42,142 29.4 12,392 Supervisors, financial records processors 157,409 49.0 77,172 Chief communications operators 66,765 34.3 22,898 Supervisors, distribution, scheduling, and adjustment clerks 159,057 19.4 30,796 Computer equipment operators 408,475 59.0 241,155 Computer operators 384,392 58.9 226,354 Peripheral equipment operators 24,083 61.5 14,801 Secretaries, stenographers, and typists 4,656,955 98.3 4,579,938 Secretaries 3,870,582 98.8 3,823,248 Stenographers 85,785 90.7 77,841 Typists 700,588 96.9 678,849 Information clerks 894,178 85.4 763,561 Interviewers 134,002 78.0 104,582 Hotel clerks 61,217 68.2 41,756 Transportation ticket and reservation agents 99,449 57.5 57,161 Receptionists 516,498 95.8 494,800 Information clerks, n.e.c. 83,012 78.6 65,262 Nonfinancial records processing 965,107 77.2 745,372 Classified-ad clerks 13,552 77.6 10,521 Correspondence clerks 19,309 81.5 15,741 Order clerks 311,321 67.4 209,871 Personnel clerks 75,235 87.4 65,759 Libras clerks 140,731 81.2 114,294 File clerks 277,592 79.7 221,350 Records clerks 127,367 84.7 107,836 Financial records processing 2,254,084 88.4 1,991,619 Bookkeepers and accounting clerks 1,827,890 89.7 1,640,233 Payroll clerks 159,292 83.3 132,622 Billing clerks 129,380 88.9 115,020 Cost and rate clerks 85,855 68.4 58,731 Billing, posting, calculating machine operators 51,667 87.1 45,013 Duplicating, mail, office machine operators 58,671 65.6 38,462 Duplicating machine operators 18,822 61.0 11,484 Mail and paper handling machine operators 7,052 62.3 4,390 Office machine operators, n.e.c. 32,797 68.9 22,588 Communications equipment operators 308,690 89.5 276,148 Telephone operators 292,165 91.0 265,938 Telegraphers 7,604 35.7 2,711 Communications equipment operators, n.e.c. 8,921 84.1 7,499
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS TABLE 1 - l (Continued) 5 Total Percent Female Occupation Employment Female Employment Mail and message distribution clerks 773,826 29.6 229,096 Postal clerks 267,035 35.8 95,511 Mail cancers, postal service 256,593 12.9 33,179 Other mail clerks 167,973 47.3 79,425 Messengers 82,225 25.5 20,981 Material recording, scheduling, and distributing 1,662,256 34.4 571,300 Dispatchers 94,830 31.2 29,568 Production coordinators 254,625 44.2 112,539 Traffic, shipping, and receiving clerks 481,958 23.6 113,554 Stock and inventory clerks 570,906 34.7 198,345 Meter readers 41,407 10.2 4,239 Weighers, measurers, and checkers 72,040 36.6 26,348 Samplers 2,542 45.5 1,157 Expediters 106,146 53.9 57,242 Material recording, n.e.c. 37,802 74.9 28,308 Adjusters and investigators 515,666 62.3 321,234 Insurance adjusters, examiners, investigators 163,586 60.2 98,407 Noninsurance investigators and examiners 243,616 62.4 151,951 Eligibility clerks, social welfare 24,128 81.8 19,744 Bill and account collectors 84,336 60.6 51,132 Miscellaneous administrative support occupations 3,296,780 83.2 2,741,523 General office clerks 1,648,934 82.1 1,353,251 Bank tellers 494,851 91.2 451,465 Proofreaders 27,321 79.1 21,610 Data-ent~keyers 378,094 92.4 349,477 Statistical clerks 139,174 75.0 104,345 Teachers' aides 206,695 92.7 191,564 Administrative support, n.e.c. 401,711 67.2 269,811 NOTE: n.e.c., not elsewhere classified. SOURCE: Data on total employment from Hunt and Hunt (1985a:Table 2.1(a)); data on female employment and percent female from Hunt and Hunt (1985a:Table 2.4); based on 1980 decennial census data. plausible worst case. The chapter also examines recent and projected shifts in the demand for various occupations. Chapter 4 analyzes the current and likely future effects of technological change on employment quality and examines how job content changes in terms of autonomy, responsibility, and knowledge. The chapter considers workers' attitudes and job satisfaction, computer-based monitoring and pacing, telecom- muting, and other aspects of job quality. It also explores the roles of managers and workers in implementing new technologies and improving employment quality.
6 COMPUTER CHIPS AND PAPER CLIPS Chapter 5 identifies some remaining areas of uncertainty, particularly with regard to the panel's analysis of likely future change. The panel offers recom- mendations to facilitate the mutual adaptation of technology and employment. The panel concludes that the problems posed by the new technologies (particu- larly for women workers) are capable of solution (1) through public and private policies that increase opportunities for women workers to benefit from change by means of education, training, and equal employment opportu- nity and that provide assistance to those who become unemployed as a result of change; (2) through the development of models for collaborative decision making and their dissemination; and (3) through research on the effects of technology and alternative ways to implement it. TECHNOLOGICAL CHANGE Much recent technological change involves the systematic use of new scien- tific knowledge in the production of goods and services. Earlier technological change, which could be called inventive technology, was episodic and unpre- dictable; its experiential, noncumulative character limited its applications. Starting around the turn of this century, however, scientific research began to substitute for random inventions and to produce solutions to socially defined problems, and methodical scientific research now feeds technology with poten- tial solutions on a continuous, regular basis. This relatively recent approach has become the dominant mechanism of technological change, although discontin- uous invention by imaginative individuals has by no means disappeared: for example, that quintessential twentieth-century invention, the Xerox photocopy method, was developed in 1938 by a creative patent attorney working indepen- dently with an unemployed physicist; they could not sell their idea to any large office machine supplier (Washington Post, August 21, 1985:B1-B21. Built on a scientific base in addition to inventors' efforts, applications have become more flexible; contemporary technological change both provides more choice and generates continuing change. Today, technological change is some- thing humans decide to do, and it can be directed in ways to produce positive results. Scientific knowledge and inventive technology make change feasible; social decisions, shaped by cultural attitudes as well as economic consider- ations, determine where change occurs and what is produced. Technological change alters the goods and services that are produced or the methods by which they are produced. It is presumed that such alterations in- crease productivity or competitive advantage. These alterations produce in- creased output with the same inputs (or reduce or make less costly the inputs needed to produce the same output), improve the quality of the outputs, or create entirely new products or services. Technological change is often con- ceived simply as the application of innovation in science and engineering to production for example, the development of rubber, the telephone, penicillin,
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 7 ceramics, or computers. However, it also encompasses new organization of production or distribution processes the assembly line, typing pool, or self- service. Innovation and the reorganization of production also occur amidst gen- eral changes in the uses of capital and labor and the location of both these factors. Technological change and its concomitants, by this very broad defini- tion, have the potential to affect all areas of social life, including art and litera- ture. This report confines its attention to technological change that occurs in the workplace and has an impact on paid employment, particularly on women workers in clerical jobs. Other non-workplace-based technological changes- for example, birth control and household technologies have had enormous effects on women's paid employment in this century, but they are not examined here. INFORMATION TECHNOLOGIES The most important technological developments affecting women's work are those involved in the miniaturization of computers and the dramatic improve- ments in telecommunications. Both developments affect the speed and ease of manipulating, analyzing, and communicating information, tasks in which the vast majority of women workers are employed. With the advent of the micro- processor, small computers with as much memory as the early large mainframe computers (which in their earliest vacuum-tube design filled entire rooms) be- came feasible. Sufficient circuits for their central processing units, including work space and operating system, can be placed on one silicon chip as small as one-quarter square inch. With the development of sophisticated software that requires little expertise in computer languages, small computers have become ubiquitous. They are incorporated in home appliances, automobiles, and hand- held calculators, as well as in stand-alone word processors and home and per- sonal computers. The miniaturization of processing is proceeding apace. The number of components that can be placed on a silicon chip has doubled every 12 to 18 months for the past 20 years. Experts see the near-exponential growth in chip capacity continuing; denser chips perform increasingly complex logical tasks faster. The cost of chips has decreased a thousandfold in 20 years (Ross, 19851. Advances in telecommunications have also occurred rapidly. If devel- opments in aircraft design had occurred on the same scale as those made in chips, "planes carrying 500,000 passengers each would be flying between New York and London for a fare of 25 cents" (Ross, 1985 :351. Computers Some observers of the very rapid increases in processing speed and memory size of computers suggest that the cost of computing hardware will continue to fall. They also expect a very rapid diffusion of computers even to small busi-
8 COMPUTER CHIPS AND PAPER CLIPS nesses, so that by 1990 perhaps one-half of all offices will have some form of computer available foroff~ce tasks. They point to annual increases in sales of 13 to 16 percent and to increasing capital/labor ratios in much of the service sector: in financial services, for example, capitalization is at $8,000 to $10,000 per worker, compared with approximately $25,000 to $30,000 in manufacturing, but the gap is closing (Office of Technology Assessment, 19851. Other observ- ers point out hidden costs (such as software development, software mainte- nance, system maintenance) and remaining technical difficulties (such as estab- lishing local area networks that link personal computers to each other and to a mainframe) and suggest that diffusion will be much slower. These observers also point to very recent decreases in sales of new computer equipment and consequent layoffs among vendors, which they take as an indication of sources of resistance to computerization in the office (Hunt and Hunt, 1985a). Still other observers point to established social arrangements in offices that may make the full adoption of office automation less feasible (Murphree, 1985; Iacono and Kling, 1986; see also Chapter 2 of this report). Finally, the rate of adoption will of course depend on the capacities and prices of the equipment offered, which will be affected by the rate of technical change within the com- puter-manufacturing industry and in telecommunications. Recent develop- ments in these areas are considered next. Data-Entry Technologies A keyboard is currently the dominant mode of data entry, but other methods are being developed. For example, an optical character reader (OCR) is available in many workplaces for specific uses and selected type styles. The OCR currently available is 40 to 50 times faster than keyboard entry. Although OCR capacities are increasing quite rapidly, it will probably be some time before handwriting can be interpreted reliably by ma- chine. Speech recognition and voice-activated "typewriters" are another po- tential input methodology, but technical development here is proceeding more slowly; most machines are limited to selected speakers and small vocabularies. Nevertheless, some experts expect that by the year 2000, 35 percent of data entry will be by OCR or speech recognition. Storage and Processing For microcomputers, magnetic discs are the domi- nant form of storage, and their storage capacities have increased several times since their first appearance. Optical discs, currently being developed, can store up to 25 times more than magnetic discs at one-thirtieth the cost; 250,000 pages can be stored on one optical disc (Office of Technology Assessment, 19851. Improvements are also being made in image processing; the development of software (because of increasing reliance on already-developed subroutines); the creation of expert systems, such as medical diagnostic systems, which are intended to simulate experts in analyzing information; and electronic mail,
TECHNOLOGICAL CHANGE AND wOMEN OFFICE WORKERS 9 including software for "automatic calendaring" the process of keeping track of appointments and arranging mutually convenient meetings. Output and Display Technologies Rapid change is occurring in improving display capabilities: better-quality printers, such as the laser printer, the liquid crystal screen, and more flexible and powerful plotters, all of which will be- come available in multiple colors. Speech synthesis, a nonvisual output, is also developing rapidly. It is already used in several large-scale applications, such as in telephone number information services. System Compatibility and Interconnection System interaction is clearly crucial if services such as electronic mail are to become widely used and if microcomputer users are to have access to a variety of information. It appears, however, to be a very difficult area. Although a report by the Office of Technol- ogy Assessment (1985) indicates that the use of local area networks (LANs) is expanding rapidly, some businesses are currently having trouble connecting personal computers and mainframes within their individual organizations; sub- stantial programming is usually involved, which slows the process. This prob- lem could be at least partially solved by standardization, which is not now occurring. In the absence of significant standardization, peripheral conversion equipment is being developed. Telecommunications Telecommunications has also improved dramatically in the past two decades and is increasingly integral to computing capability. The recent report of the Office of Technology Assessment (1985) suggests that computers and tele- phones are becoming increasingly alike. The interrelation between communi- cations and computing is easily seen: for example, a single large (but standard) switch in the Atlantic Telephone & Telegraph system has 1.6 million lines of programming (Ross, 1985~. The major components of telephony are a cus- tomer terminal, transmission, switching, and signaling. Over the years new functions and capabilities have been added to the tele- phone handset, the most common customer terminal today; other types of ter- minals in use include teletypewriters, data terminals, and computer terminals. Each of these terminals can incorporate microprocessors, giving them "logic" and "memory," making them "intelligent" and capable of more functions. The handset now provides call waiting, call forwarding, three-way calling, and so on, and sometimes serves as a terminal for data entry and information retrieval (for example, at-home banking). An earlier technical advance was the ability to transmit the digital information used by computers over the analog lines of the telephone system with the use of the modem (named for its function, modula-
10 COMPUTER CHIPS AND PAPER CLIPS tor-demodulator). Increasingly, telephone services are installing digital lines, which will speed transmission of data and video material as well as voice. The technical problems in transmission involve distance and volume. Trans- mission today uses copper wire, coaxial cable (concentric copper cables), radio (including microwave and satellite), or optical fiber cable. A very large coaxial cable can carry 132,000 simultaneous conversations. Optical fiber is small and lightweight, with vein high capacity (approximately 1,400 simultaneous con- versations on a single fiber), relatively immune to electrical interference and resistant to signal loss, and may eventually replace all other cable. A trans- oceanic optical fiber cable is being developed for use in the late 1980s. Optical fiber cables transmit information in digital form, which is more flexible than analog form, since voice, video, or data can be transmitted in digital form. Telephone switching has undergone enormous technical change. Today's automatic switching devices handle more than 600,000 calls per hour. Signal- ing takes place over a parallel network that provides the means of signaling and controlling transmission, switching, and customer terminals. The signaling system allows long-distance calls to be switched completely from their origin to their destination in less than two seconds. Among the developments that are foreseen over the next decade in comput- ing and telecommunications are improvements in the rapidity of computer re- sponse, greater responsiveness to the user in the simultaneous accommodation of several computer languages, increased use of telecommunications to connect computers and to distribute data and computation where needed, development of local area networks, increased ease of human communication with com- puters through voice recognition and more natural computer languages, and sophisticated and highly portable software (Bucy, 19851. All of these changes have affected and will continue to affect information-processing jobs by de- creasing the cost of that processing and increasing its ease. Coupled with in- creased transportation capabilities, these innovations also increase the flexibil- ity with which office work can be performed. For example, the increased miniaturization and decreased cost of computers makes work at home on small computers feasible. Large "back offices" in remote locations can serve the national needs of a company with retail branches or executive offices located elsewhere. And, within offices, innovations have contributed, and will con- tinue to contribute, to reorganizing work and changing its nature. SOCIAE CONTEXT OF TECHNOEOGICAE CHANGE Technological change, narrowly defined as the progressive adoption of a particularinnovation, generally proceeds unevenly; an innovation takes time to diffuse throughout the economy until its use becomes the new norm. Histori- cally, the time required from invention to first commercial application of an
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 11 innovation has usually ranged from 5 to 20 years, and the period from first use to widespread use, from 1 to 15 years; diffusions of major innovations, such as electricity, have taken up to 50 years (Katz and Lazarsfeld, 1955; Mansfield, 19661. Electronic computing has been used in offices since the late 1950s, when large mainframe computers first became available. Operated by specialized data-processing staffs and custom programmed for the particular applications needed, they were first used to assist in large-scale and repetitive tasks: payroll, statistical analysis of large data sets like the national census, and so on. Interac- tive systems, like that for airline reservations, were developed in the 1960s. Now microcomputers with direct telephone hookup promise a new office form: the integrated workstation, where one person can quickly and simultaneously have access to information from several data bases. Although technological change may be readily conceptualized as the adop- tion of particular innovations, it is in fact very difficult to measure technologi- cal change embodied in innovations and to assign specific results to particular innovations. Two examples- the assembly line and word-processing equip- ment suggest some of the measurement problems. The assembly line, a major technological innovation, contributed to reducing labor input per automobile and to lower sales prices for automobiles. In the same period, personal incomes increased, movies spread new tastes and behavior, and suburbanization created housing spatially removed from central cities. The automobile became a mass- consumption product leading to a substantial increase in output and employ- ment in automobile manufacturing. The decline in labor input per automobile resulted in an aggregate employment gain, but how much of the gain can be directly attributed to the assembly line? Word-processing equipment may well contribute to increases in the ease and speed of word entry and to reductions in the cost per page entered, but it may also contribute to increased or more consequential error because of the auto- mated capacities, for example, a deleted page or a "lost" document (one that was not properly stored electronically). Furthermore, as many people have observed, word processors may lead to greater numbers of drafts of a given manuscript and so to more time spent in word processing. Again, an innovation that might be expected to save labor and lead to reductions in employment might actually increase employment. And, as word processors give way to personal computers and workstations, more far-reaching effects may occur: the lines between secretarial and professional/managerial work may blur as secre- taries increase their access and ability to manipulate information, allowing them to generate analytic reports; the location of work may become more flexi- ble with stand-alone equipment and telephone access to centralized data bases; and a reduced need for paper record-keeping may alter work organization dra- matically. These examples illustrate the ways in which the uses and effects of technol-
2 COMPUTER CHIPS AND PAPER CLIPS ogy are interrelated with many other factors, including the social organization of work, the location of housing, and the norms of a community. Innovations do not by themselves increase productivity or alter employment levels; the ways they are implemented, coupled with changes in other factors, cause them to have these or other effects. The context in which new technology is introduced influences its effects on productivity, the quantity of employment, job quality, and the work environment. One consequence of the twentieth-century pattern of science-based techno- logical change has been an increase in choices about alternative paths and appli- cations by decision makers, although the degree of choice is not always appar- ent. For any given technology, there is usually a range of choices in design and implementation. Equipment may constrain, but it does not determine, work organization or job quality. The substantive scientific base of current techno- logical developments in telematics increases flexibility between technical capa- bilities and job outcomes. Microprocessors can be programmed and used in a variety of ways: for example, software can be more or less customized, more or less interactive. There can be considerable freedom in how managers and work- ers use information-processing equipment, but the degree of choice they exer- cise will depend on their knowledge and authority in the organization. These factors vary among workers and managers as well as between them. Although the same piece of equipment can be used in different ways to achieve any given goal, goals are not always clearly articulated and, like knowledge and author- ity, also differ among the parties involved. The political and social organization of the firm as well as its economic situation will thus affect the choices made about the implementation of technology. Although it is socially embedded, technology is at the same time and in itself a source of dynamism in society simply because it is likely to alter the continu- ing feasibility of change and shape its direction. It is difficult to predict the uses of new technologies: some promising ones fall by the wayside; others lie fallow for many years. The majority of innovations introduced to the market fail, but others far surpass anything imagined when they were initially developed: the use of the phonograph and radio for entertainment was not envisioned by their inventors; when computers were first developed it was thought that 8 to 10 computers would be sufficient to meet the needs of the entire United States because only a few organizations were thought to be of sufficient size to use them efficiently. The unpredictability of the long-term consequences of techno- logical change has itself become commonplace in our history and our expecta- tions. Having experienced the changes brought about by industrialization, as- sembly lines, automobiles, airplanes, telephones, televisions, charge cards, computers, gene splicing, and nuclear reactors, people in a modern society no longer imagine life without substantial technological changes even if no one knows what they will be.
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS OUTPUT AND EMPLOYMENT: TRENDS AND INTERPRETATIONS 13 In the economy as a whole, both output and employment have grown sub- stantially since World War II. Nonfarm employment grew nearly 70 percent between 1948 and 1978, while output increased nearly 165 percent (Mark, 1979~. However, employment, output, and productivity growth have slowed since 1974. Past experience has given rise to several contradictory interpreta- tions of this slowdown. One is that "too much" technological change has caused the decline of employment growth (Sadler, 1981; Noble, 1984~. Advo- cates of this interpretation note that a long-term tendency for unemployment to increase seems to be developing, despite productivity improvement in the re- cent recovery. They fear that economic growth based on rapid technological change may not encourage employment growththat the widespread use of microcomputers may contribute to the employment problem, not solve it. The nature of the new computer-based technologies has also led to questions about whether the educationally disadvantaged will become even more relatively dis- advantaged if they do not have appropriate training. A second interpretation holds that greater investments in technology are nec- essary to enhance productivity and to increase output, to keep the United 'States internationally competitive, and to maintain a healthy economy in which em- ployment will grow (Schmitt, 1983; Adler, 1984b). Indeed, several U.S. indus- tries, such as steel, textiles, and shoe manufacturing, that failed to innovate and restructure have become less and less internationally competitive. Exponents of this interpretation point to the increase in the number of persons employed, including many women, in such industries as banking and insurance, which have been technologically innovative industries. Although the rate of growth has slowed in some occupations and a few have nearly disappeared, service sector employment has grown overall because demand for its products and services has burgeoned. New microprocessing and telecommunication technol- ogy have both reduced costs of old services and permitted the introduction of new products and services in banking, insurance, and other industries (Appel- baum, 1984~; the innovations have thus contributed to increases in demand. Because these changes also involve new forms of organization, recruitment, and training, some of these observers predict that open opportunity and access to training may lead to a more democratic workplace rather than to further disadvantaging the disadvantaged. A third, and the most futuristic, interpretation of the effect of technological change on employment suggests that in the foreseeable future, productivity may be so enhanced that employment may become a rarity for everyone. People will need to learn how to use vastly increased leisure time well; new social mechanisms, other than employment, will have to be developed to distribute
14 COMPUTER CHIPS AND PAPER CLIPS income (Leontief, 1983; Bell, 1984~. As these conflicting interpretations illus- trate, new technology poses uncertainties, problems, and possibilities. Output Measures Most measures of technological change are economywide and indirect. The most commonly used indicator of technological change is the increase in output per hour of labor input, or labor productivity. What has been the change in gross national product (GNP) relative to the increase in hours worked? The labor productivity measure attributes all changes in output, including those caused by increased input of capital, to increases in the productivity of labor. A more accurate indicator of technological change, total productivity, relates output to total inputs, or at least to both capital and labor. But measurement and interpre- tation problems persist with both labor input and total input measures: for ex- ample, both methods measure changes in quantity better than changes in qual- ity. Furthermore, measures of the output of all services, but particularly of government services, are based almost totally on inputs (see National Research Council, 19791. Total productivity measures, which estimate the average rate of technologi- cal change at approximately 2 percent per year, attribute to technological change all output growth other than that due to changes in the quantity of the inputs. It is estimated that as much as 90 percent of the growth in output per capita since 1900 has been due to technological change and macroeconomic factors other then changes in the quantity of capital and labor (Mansfield, 1966; Hunt and Hunt, 1985b). Foremost among these other factors associated with greater output is improvement in the quality of labor, particularly through in- creases in average educational attainment. Other, somewhat more direct measures of technological change attempt to measure scientific and technical effort: for example, changes in the amount of money spent on research and development, in the number of patents issued, and in the number and proportion of scientific and technical personnel. Case studies of specific innovations in selected industries have also been carried out. They are generally able to assess how rapidly a particular innovation was adopted throughout an industry or several industries and how effective it was in improv- ing productivity. Such direct measures, however, are difficult to interpret inde- pendently of effects in the economy as a whole. Employment Effects Identifying and measuring the employment effects of technological change are even more difficult than measuring the output effects of technological change. The employment effects may not occur until long after the introduction
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 15 of an innovation; they may occur in entirely different firms, industries, or loca- tions or entirely different occupations from those actually using the innovation; or they may be entirely unanticipated. Furthermore, as with output changes, employment effects are qualitative as well as quantitative, and qualitative change is always much more difficult to measure. Finally, as noted, it is diff~- cult to isolate the effects of technology from the effects of other economic and social changes occurring simultaneously with the introduction of new technolo- gies; this inseparability is the fundamental problem in the interpretation of em- ployment effects. Employment Levels In the production of a particular product or service, it is axiomatic that if technological change increases the productivity of labor while demand for output does not increase, employment must fall if not immedi- ately, eventually; if not in the particular firm that innovates, in another that does not; if not in the number of workers, in the number of hours worked; if not in laying-off present workers, in failing to hire new workers. But productivity growth can also increase demand for output by lowering its price, increasing its quality, or creating new products that tap new demands, so that employment can even increase substantially. History provides examples of each ofthese effects. In U.S. agriculture, rapid technological change linked with modest growth in demand for output led to substantial displacement of workers from the agricultural sector. In 1900 nearly 12 million workers, or40 percent of the labor force, were employed in agricul- ture; today, agricultural employment is just over 3 million, or 3 percent of the labor force. In automobile manufacturing, rapid technological change led to substantial reduction of automobile prices and to increased demand and greater employment in the early to mid-twentieth century. More recently, in telephone communications, high productivity growth from rapid technological change between 1960 and 1975 was coupled with substantial increases in output and employment; the number of people employed declined in some occupations, such as telephone operator, while it increased in others, such as customer serv- ice representative. Overall technological change and shifts in consumer demand have resulted in an employment shift to services over the last several decades. In 1950 about 48 percent of the civilian employed labor force (59 million workers) worked in the goods-producing sector (agriculture, mining, construction, and manufactur- ing), while 52 percent worked in the service-producing sector (transportation and public utilities; wholesale and retail trade; finance, insurance, and real estate; business and personal services; and government). By 1980 only 31 per- cent of the labor force (99 million workers) worked in the goods-producing sector, while 69 percent worked in the se~vice-producing sector (Monthly L`z- bor Review, January 1985; calculated from Tables 1 and 91. The percentage
16 COMPUTER CHIPS AND PAPER CLIPS increase in employment in the service-producing sector was 143 percent, com- pared with 39 percent in the goods-producing sector. Among the industries within the service sector with the fastest-growing employment over the 30-year period were government (169 percent); finance, insurance, and real estate (173 percent); and business and personal services (234 percent). Technological change coupled with shifts in consumer demand for specific products is, of course, only one source of change in the level of employment. Changes in aggregate demand are equally if not more important. Unusual changes in labor supply also contribute to changes in employment and unem- ployment. The baby-boom cohorts born after World War II were associated with a dramatic increase in young entrants to the labor market that exacerbated youth unemployment during the 1970s and 1980s. These decades will be fol- lowed by several in which the relatively small cohorts of the 1970s and 1980s will enter the labor market. According to a recent study in the Monthly Labor Review (Podgursky, 1984), unemployment has increased substantially over the past several business cy- cles. Between 1969 and 1982, unemployment at the troughs increased by 4.8 percentage points, and unemployment at the peaks increased by 3.8 percentage points. One explanation for this long-term increase in unemployment is the rapid growth in the labor force, especially among youth and women; unem- ployment from this source could be viewed as frictional, associated with the normal difficulties new entrants have in finding work. Podgursky finds this source an important factor in the early 1970s, but by the late 1970s and early 1980s, prime-age men were the group whose unemployment contributed most to the increase. This finding is consistent with his finding that blue-collar work- ers in manufacturing contributed disproportionately to the unemployment in- crease over the period. This result in turn suggests that structural shifts in de- mand were important, although Podgursky argues that the data suggest a possible overall slackening in aggregate demand as well. Concerns about levels of employment were the major motivation for an im- portant earlier examination of the relation between employment and technolog- ical change. Twenty years ago the U. S. National Commission on Technology, Automation, and Economic Progress (1966) issued its report and six volumes of supporting studies. With the general increase in employment that has oc- curred since the late 1960s, the questions that motivated it are relevant again today: "Had the pace of technological change accelerated until the economy could no longer make adequate adjustments? Was technological change a major cause of persistently high general levels of unemployment? Was the new tech- nology so twisting the demand for labor that the undereducated and unskilled were becoming unemployable while the demands for highly trained manpower were insatiable?" (Bowen and Mangum, 1966: 1~. The commission found that fears of massive unemployment because of technological change were unwar-
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 17 ranted: the pace of technological change had not accelerated much if at all, and even the skill mismatch was exaggerated. However, the commission did predict that the situation for minority workers and for youth would worsen substantially if appropriate training programs were not instituted, a prediction that has proved to be correct. Despite the fact that women's rapid entrance into the labor force had already begun, it is interesting to note that not a word was said about any special needs or different experiences that women might have in the labor market. Employment Quality The commission's report was also relatively silent on another aspect of employment the quality of work. Today, concern about changes in employment quality is voiced along with concern about quantity. At a 1982 international conference on office work and the new technology, orga- nized by the Working Women Education Fund, speaker after speaker warned of the dangers to job quality posed by the new office technologies (Marschall and Gregory, 19831. Machung (1983), for example, warned of secretarial work becoming deskilled, repetitive, and monotonous as it is transformed into word- processing work. In its 1983 report, The Future of Work, the AFL-CIO Com- mittee on the Evolution of Work warned that automation would lead to down- grading of many jobs in both factories and offices. Some features of new technology suggest that it could contribute to a reduc- tion in the quality of the work environment in some jobs. Early experience with the application of microprocessing and telecommunications in the clerical sec- tor have led to reports of discomfort or possibly more serious physical problems that accompanied the use of video display terminals. Some workers perceived their autonomy and opportunity for career development to have been reduced. The new technologies have the capability of monitoring the worker's output and behavior more closely, possibly increasing stress in the job (Feldberg and Glenn, 1983~. For example, in 1983 and again in 1984, an information operator was threatened with dismissal for exceeding the C&P Telephone Company's 30-second average work time per call by 3 seconds; the operator insisted she was morally obligated to assist customers in need of extra help and to provide a higher standard of service (CWA News, March 1983, May 1984~. But the new technologies also have the potential to improve work quality as they reduce drudgery and promote more integrated work processes. Many workers have welcomed the challenge of learning new skills and mastering complex systems. The important point is that whatever the reason for adopting new equipment and organization of work, that very adoption opens choices that have consequences for the quality as well as the quantity of employment. The choices of technology made by designers, producers, purchasers, and imple- menters (whether these seek profit, competitive advantage, a higher-quality product, or other goals) and the way that they organize the work process in
18 COMPUTER CHIPS AND PAPER CLIPS relation to equipment may produce positive or negative effects, or both, on the quality of employment. WOMEN'S EMPLOYMENT OVERVIEW Women's employment has risen rapidly since World War II and especially in the past 25 years. The increase in women workers accounted for 60 percent of the growth in the labor force in the past decade and is expected to account for 70 percent of the growth in the next decade. The range of occupations in which women worked also grew, as occupations held by both men and women grew faster in the last decade than the traditionally female occupations, and women entered many predominantly male occupations formerly closed to them. Table 1-2 shows changes in the distribution of women workers across occupations for the past three decades. Women's wages still remain low relative to men's, however. The ratio of women's to men's wages for full-time, year-round work has averaged around 60 percent for several decades. In 1983 women who worked full time, year round averaged $13,468, 64.8 percent of men's average of $20,000 (Women's Bureau, U.S. Department of Labor, 19851. Some researchers predict that this TABLE 1-2 Major Occupation Groups of Employed Women, 1950-1980 (percent) Occupation 1950 1960 1970 1980 Total women 100.0 100.0 100.0 100.0 White-collar workers 52.5 56.3 61.3 63.5 Professional 12.2 13.3 15.5 15.9 Managers 4.3 3.8 3.6 6.8 Clerical 27.4 30.9 34.8 33.8 Sales 8.6 8.3 7.4 7.0 Blue-collar workers 43.9 41.8 37.9 35.5 Crafts 1.5 1.3 1.8 1.8 Operatives 20.0 17.2 14.8 10.7 Laborers 0.9 0.6 1.0 1.3 Private household 8.9 8.4 3.9 3.0 Other services 12.6 14.4 16.3 18.8 Farm workers 3.7 1.9 0.8 1.0 Managers 0.7 0.6 0.2 0.3 Laborers 2.9 1.3 0.6 0.7 SOURCE: Bianchi and Spain (1984:Table 3).
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 19 gap will close as women workers' years of experience in the labor force come to approach more closely those of men [(Smith and Ward, 1984; Goldin, 1985~; see, however, Treiman and Terrell (1975) and Corcoran and Duncan (1979), who show a strong relationship between gender and earnings even when years of labor force experience are similar for women and men]. Others believe that reductions in discrimination will bring about a smaller wage gap (Blau and Ferber, 1986~. In general, the percent female of an occupation is strongly corre- lated with its average earnings: the more women in an occupation, the less it pays. At present, job segregation by sex in the labor market is still substantial and continues to affect women's earnings and career mobility. And significant evidence suggests that promotional opportunities and access to on-thejob training for women, which are at least partially determined by employer actions, are restricted relative to those for men (Reskin and Hartmann, 1986~. Women continue to devote more of their time off the job to home, family, and child care than do men (Hartmann, 1981), and these family responsibilities, which are especially burdensome for minority women, may constrain their edu- cational and labor market opportunities. Hispanic women tend to have larger families, and black women are more likely than white women to be single parents with the total responsibility for raising children and financially support- ing them. Because women workers tend to be concentrated in a limited set of occupa- tions, because they are sometimes less geographically mobile, and because their access to education, training, or promotion within and across firms may be more limited than men's, the panel expects technological change to affect women and men differently. Because women earn less than men, they may also have fewer resources with which to respond to technological change. If techno- logical change affects women's employment opportunities more negatively, or less positively, than it affects men's, it will contribute to maintaining, or even worsening women's relatively disadvantaged status. Conversely, technological change, if managed toward that goal, could provide a means to equalize the status of women and men. WHY TECHNOLOGY MAY AFFECT WOMEN DIFFERENTIALLY Job Segregation Job segregation by sex has been large and relatively stable in the United States, although the past two decades have seen some decline. Women work largely in different occupations than do men, and they are in occupations that are predominantly female. More than 36 percent of all employed women work in just 10 occupations, and 9 of these are female dominated: secretaries, ele- mentary school teachers, bookkeepers, cashiers, office clerks, "managers-
20 COMPUTER CHIPS AND PAPER CLIPS not elsewhere classified," waitresses and waiters, salesworkers, registered nurses, and nursing aides. Only 1 of the 10 largest occupations for women workers, "managers not elsewhere classified," is among the 10 largest occu- pations for men (see Table 1-3~. Job segregation by sex decreased in the past decade both because female dominated occupations grew less rapidly than in the past and because a substantial number of occupations became more inte- grated, particularly in the professions and management. The clerical occupa- tions, however, for the most part became more female dominated (Reskin and Hartmann, 19861. TABLE 1-3 Employment in the 10 Largest Occupations for Men and Women, 1980 Ten Largest Occupations for Men Detailed 1980 Census Occupational Title and Code 1. Managers, n.e.c. (019) 2. Truckdrivers, heavy (804) Number of Men 3,824,609 1,852,443 3. Janitors and cleaners (453) 1,631,534 4. Supervisors, production(633) 1,605,489 5. Carpenters (567) 1,275,666 6. Supervisor, sales (243) 1,137,045 7. Laborers (889) 1,128,789 8. Sales representatives (259) 1,070,206 9. Farmers (473) 1,032,759 10. Auto mechanics (505) 948,358 Percentage Female 1980 26.9 2.3 23.4 15.0 1.6 28.2 19.4 14.9 9.8 1.3 1970-1980 Change in Percentage 1970 Female 11.6 0.8 13.1 10.3 9.9 5.1 1.1 0.5 17.0 11.2 16.5 2.9 7.0 7.9 4.7 5.1 1.4 -0.1 1970- 1980 Ten Largest Occupations for Women Change In Detailed 1980 Census Number Percentage Female Percentage Occupational Title and Code of Women 1980 1970 Female 1. Secretaries (313) 3,949,973 98.8 97.8 1.0 2. Teachers, elementary school (156) 1,749,547 75.4 83.9 - 8.5 3. Bookkeepers (337) 1,700,843 89.7 80.9 8.8 4. Cashiers (276) 1,565,502 83.5 84.2 - 0.7 5. Office clerks (379) 1,425,083 82.1 75.3 6.8 6. Managers, n.e.c. (019) 1,407,898 26.9 15.3 11.6 7. Waitresses and waiters (435) 1,325,928 88.0 90.8 - 2.8 8. Salesworkers (274) 1,234,929 72.7 70.4 2.3 9. Registered nurses (095) 1,232,544 95.9 97.3 - 1.4 10. Nursing aides (447) 1,209,757 87.8 87.0 0.8 NOTE: n.e.c., not elsewhere classified. SOURCE: Rytina and Bianchi (1984).
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS Clerical Occupations 21 The growth in clerical employment during the post-World War II period undoubtedly facilitated the increase in female labor force participation that oc- curred over the same time period. Between 1940 and 1980, the proportion of all workers employed as clerical workers doubled from about 10 to 20 percent, while female participation in the labor force nearly doubled. The proportion of women working as clerical workers also increased substantially. Clerical occu- pations remain the mainstay of women's employment, and it is these occupa- tions that are currently undergoing substantial technological and organizational changes integral to broader structural changes in the service sector. The service sector disproportionately employs clerical workers and women. Changes in either the level of employment or the quality of work in that sector will directly, and disproportionately, affect women. In 1981 only 6.3 percent of men worked in clerical occupations, compared with 34.7 percent of women. Clerical occupations are diverse along several dimensions. Some are better paid and require more skill than others. Some are held disproportionately by minority women; others disproportionately by majority women. Earlier waves of automation were expected to decrease the number of clerical workers, but the number continued to increase. Earlier automation not only had displacement effects in individual occupations but also substantial employment-increasing effects across occupations. Recently, innovations in computation, dictation, and record-keeping have affected various occupations in different ways: since 1950, the number of stenographers has declined dramatically; the number of typists, file clerks, and postal clerks has decreased since 1970; and the number of computer operators increased substantially between 1950 and 1980 (Hunt and Hunt, 1985a). Scholars disagree about how to interpret emerging trends and to project the future effects of technology on levels of clerical employment. In their well- known study, Leontief and Duchin (1984) offer disaggregated forecasts for the next 20 years which suggest that, relative to other occupations, clerical work will be more affected by displacement due to computer-based automation. In contrast, the Bureau of Labor Statistics (Silvestri et al., 1983; Silvestri and Lukasiewicz, 1985) predicts substantial growth of clerical workers through 1995. The contradictory forecasts stem from differences in assumptions about the rate of innovation and diffusion of various new technologies, their produc- tivity effects, and the size of changes in final demand over time for the products and services clerical workers help to produce. Whichever prediction turns out to be more accurate, however, the concentration of women in clerical occupa- tions and the concentration of technological change in the same occupations suggest that women are likely to be differentially affected relative to men. Of
22 COMPUTER CHIPS AND PAPER CLIPS course, if women had equal opportunity in the labor market, they would almost certainly be less concentrated in the clerical occupations and technological change there would have less differential effects on women. Differential Status and Access to on-the-Job Training Even when they are in the same occupations, men and women may face different opportunities because they work in different firms with different pay levels (Blau, 1977; Strober and Arnold, 19851. Even within the same firms, women and men are likely to have different access to on-thejob training and promotion (Duncan and Hoffman, 19791. Employers' attitudes toward workers and their treatment of them are often conditioned by gender (Reskin and Hart- mann, 1986~. These stratification patterns may signal potentially negative ef- fects of technological change for women in general; in addition, minority women experience ethnic and racial prejudice as well as sex-based discrimina- tion, which constrains their opportunities further (Malveaux, 1982~. For exam- ple, minority women may be placed in back-off~ce jobs in which access to promotion and training opportunities is limited. Women may also be differen- tially affected relative to men because as workers, men and women do not enjoy the same status and power or access to organizational and collective resources (such as labor unions). Differential Responsibility for Family Care Even if men and women shared equal opportunity to benefit (or suffer) from workplace technologies, women might be prevented from taking advantage of opportunities because of their greater responsibility for family care and house- work. They may also be more geographically restricted, less able to participate in educational programs, and more constrained in their job choice. One aspect of the capability of the new technologies may be especially relevant to differ- ences between women and men. The microprocessor may make home-based production economically feasible for employers, and women's household re- sponsibilities and the lack of affordable child care may make homework espe- cially attractive to women. Part-time homework, either on a salaried or self- employed basis, may be preferred by some women. The new technologies may also lead, however, to increases in involuntary part-time and temporary work, if they have significant employment-displacing effects. Because women have traditionally provided more care for family members than men have and because men and many women too have seen this ar- rangement as right and proper, some managers have developed assumptions about women workers that are increasingly inaccurate today. One such historic assumption is that women workers have higher turnover rates and less commit-
TECHNOLOGICAL CHANGE AND WOMEN OFFICE WORKERS 23 ment to their jobs than men because of their family obligations (Feldberg and Glenn, 1979~. This assumption may lead to less concern about eliminating women's jobs or to reorganizing their work in potentially inequitable ways, such as in the creation of temporary or part-time work with limited opportuni- ties. But changing technology and reorganization can also be an opportunity for opening up new occupations and increasing mobility for women if the changing characteristics of women workers, in particular their increasing attachment to the labor force, are taken into account. Conclusion The panel expects that differential effects of the new technologies with re- spect to sex are likely. And for some of the same reasons differences in choices, opportunities, treatment we expect that some effects will also differ among subgroups based on ethnic, racial, age, marital, educational, or geo- graphic characteristics of women. In many areas relevant to the new technolo- gies, however, the effects are essentially the same for all workers and all sub- groups of women. The new technologies have monitoring capabilities, for example, whether either women or men work with them, or both men and women do. The work environment is equally important for all workers. This report therefore discusses issues that apply to all workers, although they are particularly relevant for women. It focuses primarily on technological change in clerical occupations.