The Information Society


More Information, Better Jobs? Occupational Stratification and Labor-Market Segmentation in the United States' Information Labor Force
By Rob Kling
Vol. 7(2) 1990

This paper examines the mix of good jobs and bad jobs in the restructuring of United States labor markets for information work between 1900 and 1980. Is the information sector still growing relative to other occupational sectors? What is the relative proportion of good jobs and bad jobs in the information sector today? Is the mix of good jobs and bad jobs within the information sector changing over time? To answer these questions, we examine changes in the relative size of the information sector's labor markets and changes in five occupational strata within it -- professional, semi-professional, supervisory and upper level sales personnel, clerks, and blue collar workers. 

The information occupations mushroomed in size from 17% of the United States work force in 1900 to over 50% in 1980. Information sector jobs vary widely in quality. Few information sector jobs are fully professional, and clerical jobs form the largest single occupational stratum. When we examined the growth of the various strata between 1900 and 1980, we found that clerical jobs became more dominant, not less dominant. But this distribution has been masked by the steady growth of information sector jobs in the highly professional and semi-professional strata, as well as clerical jobs. The occupational stratum between clerks and semi-professionals -- the supervisory and upper level sales workers -- has steadily declined in relative size. 

Two lower strata -- clerks and sales and supervisory workers -- account for 55% of the jobs in the information sector. Our data suggest that information labor markets are divided into relatively impermeable segments. As the information sector expanded, it took on many characteristics of the overall economy. It includes a mix of jobs that are diverse in their pay, status, and power. Its internal divisions reflect patterns of segmentation that have developed elsewhere in the society -- a dual labor market. Overall, the information sector has become sufficiently large that is not an alternative to the dominant social order -- it simply reproduces many of its features. 

Table of Contents
  • Work in an Information Economy 
  • The Concept of an Information Economy 
  • Changes in the Distribution of Information Occupations 
  • Stratification of Jobs in an Information Economy 
  • Scale, Growth and Structure of the Information Work force 
  • The Segmentati­on of Information Labor Markets 
  • Conclusions 
  • References 
  • APPENDIX A -- Measuring the Size of the Information Work force
  • APPENDIX B -- Notes on Sources of Data and Methods
  • List of Figures and Tables
  • Figure 1 -- Theoretical distributions of Jobs across 4 occupational strata
  • Table 1 -- Four-sector aggregation of the United States Labor Force strata
  • Table 2 -- Occupations in the Information Economy
  • Table 3 -- Federal Employment in the Information Sector: by Occupat­ion 1980 
  • Table 4 -- Distribution of Occupational Strata Within the Information Sector: National Economy: 1900-1980
  • Table 5 -- Distribution of Occupational Strata Within the Information Sector
  • Work in an Information Economy
    Is the United States' economy producing a large number of relatively good jobs, relatively poor jobs, or some mix? Are there increasing opportunities for occupational mobility across occupations and labor submarkets, or are occupations and labor markets sharply segmented? This paper answers these questions by examining the changing patterns of employment in the United States 

    The information sector is a relatively new construct. It is composed of those jobs in which people record, process or communicate information as a large fraction of their work. These are diverse occupations, including managers, lawyers, accountants, realtors, stock brokers, and clerks of all kinds. People process information in important ways as part of any job -- including truck drivers, trapeze artists, and machinists. The label "information work" is a concise way to characterize jobs where information is a key product of the job, or where the person is likely to spend a large fraction of each workweek communicating, reading, search for information, or handling paperwork in its various forms, including electronic transactions. 

    The information work force differs significantly from the hi-tech work force. The hi-tech workforce is composed of people who work in a variety of jobs within hi-tech manufacturing firms. These firms have over half of their work force in hi-tech occupations, such as engineers and biologists. But 30-40% of their work force can be composed of people in other occupations, such as secretaries, accountants, assemblers and truck drivers. 

    In contrast with the hi-tech work force, the information work force is composed of a specific set of occupations. The information work force gives us a complementary, but larger window, through which to examine social change.

    Information processing jobs are playing a major role in today's economy. Several analysts have argued that information handling is not simply a feature of existing jobs, nor even a central element in a few jobs. Rather, they see it as a key dimension for characterizing labor markets and urban economies. Castells (1984), for example, simply declares that information handling is a defining activity in new metropolitan formations. He dubs leading edge urban developments as "an informational city." Fishman (1987) labels post-suburban regions as "tech­noburbs." Thus he indicates that their dependence on transportation and communication technologies undergirds their social and spatial forms. Knight (1986) follows Daniel Bell's characterization of post-industrial societies, and argues that knowledge work is a core activity in a transformed urban economy. 

    These provocative theses are worthy of investigation. They provide an important alternative entry point to the study of the work force in advanc­ed economic areas than do the traditional trichotomy of agriculture, manufacturing, and services. Unfortunately, analysts who write about the primacy of information work usually treat it as a "social fact." They rarely examine the meanings of information work in new social formations. In contrast, Lyons (1988) argues persuasively that information work should be examined as a new problematic, and not be taken for granted. Zuboff (1988) goes further than Lyons. She argues that organizations with highly "informated" jobs require specially skilled workers who can challenge traditional managerial styles and whose skill is a novel resource for mangers to cherish. 

    The "information sector" is a relatively new analytical category which we believe is worth exploring. The United States work force is composed of people in hundreds of occupations. Every simple category scheme mixes diverse occupations and working conditions. The traditional economic sector scheme mixes field hands and agribusiness accountants (agriculture); welders, stock clerks, and industrial engineers (manufacturing); cab drivers, waitresses, and lawyers (services). Dist­inctions between union and non-union cut across occupations, and depend upon local circumstances. Similarly, the crude distinctions between information work and other kinds of work mixes diverse occupations, like postal clerks and lawyers. But this paper goes beyond the usual accounts of information work by grouping information workers into five occupational strata which differ in the typical quality of their jobs. 

     This paper examines the composition of the information workforce with special attention to simple differences between better and worse jobs (see below). We primarily use demographic data to examine the changing composition of dif­ferent strata in the information work force. 

    The Concept of an Information Economy
    Social analysts characterize the major economic transformations of this era with two different rubrics:a posti­ndustrial society (Bell, 1973; Ginzberg, et. al., 1986) and an information society (Porat, 1973; Bell, 1981; Huppes, 1987). Some analysts loosely mix these portraits (Naisbitt, 1984; Huppes, 1987), but they have different connotations. The service sector dominates in Bell's (1973) characterization of post­industrial economies. He argues that post-industrial societies also depend critically upon credentialed experts, especially "knowledge producers" such as scientists and engineers. But scientific and engineering occupations form only a small fraction of the jobs in even the most technologically advanced societies. The service sector is dominated by a variety of industries, from transportation to restaurants, from insurance and banking to utilities (Ginz­berg, et. al., 1986). These service industries are composed of two key kinds of jobs:

  • (a) jobs where people provide direct service -- such as bank tellers, waitresses, stock brokers, lawyers, security guards, bus drivers, insurance agents, etc. 
  • (b) jobs in the administrative core of these organizations -- such clerks, accountants and office managers of various kinds as well as specialists in marketing, computers etc. (These jobs are found in all industries, not just service industries).
  • In contrast, the imagery of an information economy focuses upon occupations in which the processing of information is a central and time consuming activity. All jobs require that people process some information, even if only sensory information to know where they stand and where they are going. But some workers also provide information as a central element of the services they provide. These jobs include certain service jobs such as teaching, practicing law, research, etc. They also include core admin­istrative jobs such as clerks, accountants, and computer programmers, wherever they may be found -- in agriculture, manufacturing or services. For example, Scott (1988:178) notes that over 35% of the United States' manufacturing work force is composed of white collar workers. And most of these are information workers such as engineers, inspectors, clerks, and accountants.

    Timothy Luke and Stephen White (1985:31) argue that capitalism has entered a new phase, informational capitalism in which "data-intensive techniques, cybernetic knowledge, and electronic technologies" are the new strategic resources for corporate production. They point out that core information services have expanded (banking, insurance, telecommunications, mass media, advertising, education) In addition, managers of service, industrial, and agricultural firms have invested heavily in strategies to base production on information: banks invest in data processing and electronic funds transfer, manufacturing plants invest in automated inventory control and robotics; agricultural firms invest in computer-based farm management programs; and managers in all sectors invest in information systems to support basic accounting and cash flow management. Behind these strategies sit a variety of workers -- from the specialists who design them to the wide variety of people who use them. 

    Marc Porat (1977) estimated that the information sector accounted for more employment than manufacturing or services by 1950, and that approximately 46% of the work force was employed in the information sector in 1970 (Table #1). 
    TABLE #1


    Experienced Civilian Workforce

    Information Sector
    Agriculture Sector
    Industry Sector
    Service Sector
    · Data from Table 9.2 Daniel Bell (1981:522). Based on Bell's median definition of the information economy, which differs somewhat from our own definition reflected in other tables. 


    Most of the occupations in the information sector are white collar jobs. The terms "white collar" work force and "information work force" overlap substantially. But they are not identical. Much of the growth of the information sector in this century was driven by the same forces which drove the growth of white collar employment: the massive gains in productivity in agriculture and manufacturing; the rise of services; and (especially) the rise of large bureaucracies -- public and private (Mills, 1951:68-69). 

    Even though the terms "information economy" and "information work" are becoming commonplace, they often are used loosely. We see three problems with these casual usages:

  • (a) Many authors talk about an "information age" or "information economy" as a social fact that can be taken for granted (e.g., Naisbitt, 1984; Huppes, 1987; Strassman, 1985; Luke and White, 1985). These accounts usually draw upon Porat's (1977) pioneering analysis which segmented the United States' economy into four sectors. The national data used by Porat, and often repeated in other publications, aggregates workers across regions with widely disparate industrial mixes and misses any distinguishing regional characteristics. Employment in rural regions such as California's San Juaquin Valley, centers of smoke­stack industry such as Gary Indiana, and service centers such as Hartford Connecticut, are all combined. But some metropolitan economies are much more information intensive than others (Kling and Turner, in press). 
  • (b) Many analysts who examine the information economy focus on the best jobs - particularly professional, technical, and managerial jobs. Some of them ignore the poorer jobs, such as clerical work in its various forms (Strass­man, 1985). Or they treat poorer jobs as transitional occupations which may soon disappear (Giuliano, 1982). Or they argue that office automation will eliminate a large portion of the least skilled "entry level" jobs (Rosenberg, 1986; p.234) Are good jobs or bad jobs most prevalent in an information economy? Is the mix of jobs tilting toward a larger number of better jobs?
  • (c) Some authors assume there is a natural evolutionary sequence from agricultural to manufacturing to service economies, and that information economies are simply the fourth step in the sequences (Huppes, 1987). Daniel Bell argued that agricultural economies naturally evolve into post-industrial (service) economies, and sees the information sector as part of the infrastructure for a post-industrial society (Bell, 1981). We do not assume there is a natural four stage evolutionary sequence from agricultural societies to information societies. Rather, we treat the information sector as an important economic sector which crosscuts the traditional three sectors and whose occupational structure may shed interesting light on work in modern society.
  • Many commentators confuse the social meanings of information work by confounding it with the use of some form of information technology, especially advanced computer systems. For example, Giuliano (1982) advanced a typical interpretation of "information work" and "information-age" offices when he examined the shift of office technologies from pen and paper through typewriters and mechanical devices through interactive computer-based systems available on every desk.

    He argues that the social organization of office work is evolving through three stages (a) an informal "preindustrial" office; (b) a highly regimented "industrial" office; and (c) a flexible "information age" office in which computerized information systems are available on every desktop. There are major technological differences in his illustrations of these archetypical offices. His preindustrial office relies on telephones, paper, and organized files. His industrial office relies on batchrun computerized information systems, as well as paper and telephones. His information-age office relies on universal desktop computing linked to interactive databases. He characterizes the information-age office as one which 

    "exploits new technology to preserve the best aspects of the preindustrial office and avoid their failings. At its best, it combines terminal-based work stations, a continuously updated data base, and communications, to attain high efficiency along with a return to people centered work rather than machine-centered work. In the information age office the machine is paced to the needs and abilities of the person who works with it...­The mechanization of office work is an essential element in the transformation of American society to one in which information work is the chief economic activity."
    Giuliano's article illustrates the typical confusion of information work and specific technologies. All of his illustrations refer to offices which are exclusively devoted to information handling in some form. There is no substantive rationale for distinguishing any one of these office forms as "information age" offices. The label glamorizes the kind of office technology which Giuliano would like to see widespread -- interactive computer systems linked to integrated data bases. As Porat's study of the information work force shows, the information work force was 17% of the national workforce by 1900, and it grew to 30% by 1950 -- long before electronic computer systems were installed in offic­es.

    We believe that these computer technologies can be very interesting and become indispensable. But the ways in which they transform work is still open to question and investigation. As Iacono and Kling (1987) observe, "Despite the dramatic improvements in office technologies over the past 100 years, career opportunities and working conditions for clerks have not similarly improved. Although clerical tasks today require more skills in using a complex array of technologies, these skills are not reflected in status or pay. A new generation of integrated computer-based office systems will not automatically alter the pay, status, and careers of clerks without explicit attention (p.75)." While these issues of pay, status, and career lines are peripheral to understanding how the labor process and phenomenology of information work differs from other kinds of work, such as craft work, we return to them in the next section when we examine the occupational structure of the information work­ force.

    Shoshana Zuboff's In the Age of the Smart Machine is the most daunting and serious recent study which examines the labor processes and phenomenology of work with computer-based systems. She provides vivid and often brilliant descriptions of the phenomenology of work with special computer systems in specific work settings. Superficially, her account differs substantially from Giuliano's. For example her illustrations portray work­places like his industrial age office as much less pleasant work­places than information age offices. But she makes similar errors to Giuliano by not carefully distinguishing between the phenomenology and labor process of computer-based work and informa­tion work more generally. Like Giuliano, Zuboff coins a special term ("informate") with broad informational connotations to describe some special aspects of work with computers (pp. 9-10).

    Zuboff identifies abstraction as a special feature of computerization rather than as a generic feature of symbol systems, whether represented by cunei form on papyrus, quill pen marks on parchment, pencil marks on paper, or data displays on an elect­ronic screen (pp. 69, 79, 83-84). Moreover important kinds of information work do not always entail the use of abstract symbol systems. 

    A good deal of information work also entails interpersonal communication, often face to face, but sometimes mediated by telephone -- between students and teachers, judges and defendants, clerks and organizational clients. Although it is difficult to find occupations in which participants do not rely on any symbol systems, workers vary in the extent to which their world is mediated by symbol systems rather than communicated by personal experience. Hotel concierges and realtors, for example, often work with a good deal of key information based on their personal knowledge of localities and provide it to clients verbally. In contrast, other information workers, such as stock brokers, are much more wrapped in a complex world of abstract symbols and systems of relationships between them (e.g., stock prices, trading volumes, market averages, interest rates, transaction costs). Through the 1970s and even early 1980s, most stock brokers relied upon paper systems, telephones, and specialized information services such as Quotron. By the 1980s, most stock brokers gained access to more complex information systems which provide a wider array of data faster and which allow easier comparisons between data. They can also routinely monitor and trade in a wider variety of financial instruments and international markets, ranging from Eurodollars to the Japanese stock market. Information technologies have helped reshape the job of stock brokers. Their work may have been much simpler in the 1960s than in the 1990s. But their knowledge was anchored in complex symbolic systems, although it was based on paper. The routine dynamics of stock markets and their relationships to other markets, such as money and bond markets, have not substantially changed because of computer systems -- except for the special phenomenon of program trading. 

    Zuboff views computerized work as a major transformation in labor processes which diminishes the importance of the physical body as an acting and knowing agent. But Zuboff glosses key similarities between information work and computer based work. She focuses on a small set of cases in which relatively low level paper plant operators and clerks were beginning to use computer-based systems that were thrust upon them. Her sensitive observations about the phenomenology of computerized work in these special settings can often apply just as well to many forms of information work.

    Giuliano's and Zuboff's accounts of information work, information technology and work illustrate common arguments about the kinds of labor processes which undergird information work. We find both of these accounts suggestive, but also misleading because they confuse information work with special kinds of work which uses advanced computer-systems.

    We will not examine the phenomenology of information work in more detail here. While a more careful investigation would be very useful, it would apply to information work wherever it is done and in whatever historical period: London England in the 1600s, and Irvine California in the 1990s. But it will not teach us about recent changes in the work and labor markets within the United States without important additional information about the distribution of jobs, occupations, and technologies in the region. 

    Changes in the Distribution of Information Occupations
     Studies based on different units of analysis could shed special light on changes in a regional labor market -- a study of changes in specific occupations within the region, a study of changes in information work at the firm level, and studies of the occupational mix in a specific labor market. We will focus on changes in the distribution of jobs within the information sector of the United States as a way to better understand the changing mix of good and bad jobs. This choice helps shed light on the structure of information labor markets by using the same kinds of data that protagonists of the information economy use. 

    Table 2



    lawyers and judges
    life scientists
    operations researchers
    physicians and related practitioners
    physical scientists
    social scientists
    teachers, college and univ.


    bank tellers
    billing clerks
    clerical workers (nec)
    collectors, bill and account
    counter clerks, except food
    dispatchers and starters,vehicle
    enumerators and interviewers
    estimators and investigators, n.e.c
    expediters and production controllers
    file clerks
    hucksters and peddlers
    library attendants and assistants
    mail carriers, post office
    mail handlers, except post office
    messengers and office helpers
    newspaper carriers and vendors
    office machine operators
    payroll and timekeeping clerks
    postal clerks
    sales clerks, retail trade
    sales workers, except clerks
    shipping and receiving clerks
    statistical clerks
    teacher aides, except school monitors
    telephone operators
    ticket, station, and express agents
    welfare service aides 


    bank officers 
    computer specialists 
    financial managers 
    foresters and conservationists 
    health administrators 
    librarians, archivists, and curators 
    managers and administrators (nec) 
    nurses, dietitians, and therapists 
    office managers 
    officials and administrators (public) 
    personnel and labor relations workers 
    research worker 
    school administrator 
    social and recreation workers 
    teachers, except college 
    technical workers (nec) 
    vocational and educational counselors 
    writers, artists, and entertainers 

    Sales and Supervisory

    advertising agents and sales workers 
    blue-collar worker supervisors, n.e.c 
    buyers and purchasing agents 
    clerical supervisors, n.e.c 
    credit and collection managers 
    engineering and science technicians 
    health technologists and technicians 
    insurance adjusters & examiners 
    insurance investigators 
    insurance agents & brokers 
    insurance underwriters 
    officials of lodges & societies 
    real estate agents and brokers 
    sales managers, including retail trade 
    sales representatives, manufacturing 
    sales representatives, wholesale 
    stock and bond sales agents 
    union officials 

    Blue Collar Information Workers

    checkers, examiners - manufacturing 
    data processing machine repairers
    inspectors - manufacturing
    office machine repairers
    photographic process workers
    printing craft workers
    radio and television repairers
    telephone line installers
    telephone repairers

    Our basic strategy in this paper is very simple. 
  • (a) We have adapted Porat's (1977) characterization of information workers and his list of information workers (Table #2) which he used for estimating the size of the information work force in the United States. We use a similar list of occupations to estimate the overall size of the informa­tion work force in the United States in 1980. These are aggregate estimates of employment. 
  • (b) We are particularly concerned with the mix of good jobs and bad jobs in the information work force. Many analysts treat the information work force as relatively homogeneous. We have divided information jobs into five status strata -- professional, semi-professional, supervisory and upper level sales personnel, clerks, and blue collar information workers (see below). We examine the relative number of jobs in each of these strata. For example, are professional jobs, such as accountants and lawyers, the largest stratum of jobs in the information sector? 
  • (c) We are also concerned with changes in the information sector over time. Is the information sector still growing in relative size? Is the mix of good jobs and bad jobs within the information sector changing over time? To answer these questions, we examine the information sector in the United States national economy level from 1900 to 1980. 
  • No one has asked these questions about the character of jobs in the information economy so directly. Some answers appear implicit in key writing about the information labor force: 
  • (a) it is relatively large, and is the dominant sector of the United States labor force (Porat, 1977; Bell, 1981; Huppes, 1987; Strassman, 1985); 
  • (b) it is continuing to grow in size; 
  • (c) Good jobs are most prevalent (Parker, 1981:73; Giul­iano, 1982; Strassman, 1985).
  • We will critically examine these ideas in this paper. 

    Stratification of Jobs in an Information Economy
    Optimistic themes of universal progress undergird most accounts of the information economy. However, few authors carefully explain how good jobs replace bad ones. Giuliano (1982) implies that jobs will be upskilled. In three diagrams that illustrate his article, he indicates that highly specialized accounting clerks will become "account managers" when work is electronically integrated through the use of advanced information systems. Giuliano's claim goes beyond the arguments about upskilling and deskilling since he implies that job holders take on new jobs -- moving from narrow clerical jobs like "posting clerks" to much broader "account managers". In bureaucratic terms, these information oriented jobs have simply been reclassified.

    We were originally critical of accounts of information work like Giuliano's which casually assumed that semi-professional jobs would replace clerical jobs. During the twentieth century, there has been a white collar revolution in the United States. White collar workers shifted from about 18% of the work force in 1900 to about 48% in 1974. Manual work remained at 35-40% of the work force during these 75 years. And farm work declined from 38% of the work force in 1900 to 3% by 1974.

    We were also critical of accounts which focussed on professionals and ignored clerks, since clerks are a major occupational group in the US. Clerks grew from 3% of the national work force in 1900 to 18% in 1974 - a 6 fold increase. In contrast, professionals grew from 4% of the work force in 1900 to 14% in 1974 - a growth factor of about of 3.5. Managers grew from 6% of the work force in 1900 to 10% in 1974 - a factor of about 1.6. In 1900, managers outnumbered clerks by two to one; in 1974, clerks outnumbered managers by 1.6 to 1. Clerical jobs formed about 25% of the white collar jobs by 1974. Information jobs are not the same as white collar jobs. But they are similar enough to suggest that clerks work could form 20-30% of the information work force. 

    There are at least five kinds of criteria for ranking jobs from better to worse. Economic criteria focus on pay, benefts, security, and career opportunities. Psychological criteria focus on feelings, such as challenge, autonomy, boredom, and filiation with coworkers. Sociological criteria focus on prestige and power. Health and safety criteria focus on routine exposure to toxic materials or dangerous situations. Situational criteria focus on idiosyncratic features of jobs such as fit with one's personal life, flexibility of schedules, and the time spent commut­ing to work. These characteristics are not fixed for all people in a specific occupation. Ranking of jobs by these criteria can depend on the specific practices of particular employers and the specific preferences of particular employees. Particular jobs may be ranked at one extreme on one set of criteria, yet be mediocre or poor on other criteria. Some industries such as transportation and aerospace pay their workers more for comparable jobs than do banking and insurance. While we recognize these complexities, we sought a simple strategy for characterizing the quality of jobs and comparing them over time. Any classification scheme that places some occupations into a "better work" category and other occupations into a "worse work" category simplifies complex criteria by which people assess specific jobs.

    We considered two different ways to characterize the quality of jobs. We reviewed economic criteria such as income, and social criteria such as status, autonomy and related working conditions. We could not locate adequately detailed income data for each of the occupations in the information work force for 1900 to 1980 at the Federal level. Therefore, we turned to social criteria for comparing the quality of jobs in the information sector.

    We followed Porat's (1977), list of jobs in the infor­mation labor force. We went beyond his seminal work by dividing the information sector occupations into five broad strata. (See Appendix A for details). We constructed four white collar occupational strata and one blue collar stratum. The white collar strata range from fully fledged profess­ions at one extreme to clerks at the other. They do not include all white collar jobs (e.g., dentists), since we are examining the information sector rather than white collar work. We used standard sociological categories for professions and semi-professions. In addition, we identified one occupational stratum between clerks and semi-professionals.

  • (a) Professional occupations includes the 8 most highly professionalized jobs in the United States in the information work force, including accountants, architects, lawyers, and physicians. The most highly developed professions have legal monopolies over legitimate practice and credentialing requirements. These are usually the most prestigious occupations, and many of these occupations pay relatively well -- on average. Jobs within these occupations vary considerably along other criteria, such as stress.
  • (b) Semi-professional occupations includes 19 groups which have some professional standing, and are not fully fledged professions. They include computer specialists, engineers, managers, school administrators, social workers, and teachers. Semi-professional occupations are usually less prestigious and less well-paid, on average, than the full fledged professionals. But they are usually much more autonomous and prestigious than the occupations in the next lower stratum.
  • (c) Supervisory and upper level sales occupations constitute a category whose status lies between that of semi­professionals and that of clerks. It includes advertising agents, health technologists, insurance agents, office managers, purchasing agents, real estate agents, and stock brokers. This is a complex stratum which lies between the moderately presti­gious semi-professional occupations and the less prestigious, less autonomous and lower paid clerical occupations.
  • (d) Clerical occupations includes clerical jobs of all kinds (including cashiers and sales clerks­) . We view clerical jobs as problematic because they usually pay poorly compared with other information jobs. Clerical jobs vary considerably in autonomy -- from telephone operators to executive secretaries. But they are often fairly regimented. And they usually provide few opportunities for moving to much more autonomous or substantially better paying jobs.
  • (e) Blue collar occupations includes technicians who install or repair communications, printing, and other information processing equipment.
  • We classified each occupation in the information sector into one of these five strata in Table #2 and report their employment levels in Table #3. 
    TABLE #3
    1980 Employment in the United States Information Sector: by Occupation 
    Fully Professional
    physicians, dentists, and related practiti
    teachers, college and university 
    lawyers and judges 
    life and physical scientists 
    social scientists 
    operations and systems researchers and ana
    all other managers and administrators 
    teachers, except college and university 
    nurses, dietitians, and therapists 
    writers, artists, and entertainers 
    bank officers and financial managers 
    computer specialists 
    social and recreation workers 
    school administrator, college and elementa 
    official and administrators, public admini 
    health administrators 
    librarians, archivists, and curators 
    vocational and educational counselors 
    research worker 
    foresters and conservationists 
    all other professional and technical worke 
    Upper Level Sales and Supervisory
    blue-collar worker supervisors, n.e.c
    engineering and science technicians
    sales representatives, wholesale trade
    sales managers, including retail trade
    real estate agents and brokers
    health technologists and technicians
    insurance agents, brokers, and underwriter
    personnel and labor relations workers
    buyers and purchasing agents
    sales representatives, manufacturing indus
    clerical supervisors, n.e.c
    insurance adjusters, examiners, and invest
    stock and bond sales agents
    advertising agents and sales workers
    inspectors, except construction and public
    officials of lodges, societies, and unions
    credit and collection managers
    sales workers, n.o.m. 
    all other clerical workers
    office machine operators
    estimators and investigators, n.e.c
    bank tellers
    shipping and receiving clerks
    statistical clerks
    teacher aides, except school monitors
    counter clerks, except food
    file clerks
    telephone operators
    postal clerks
    mail carriers, post office
    expediters and production controllers
    payroll and timekeeping clerks
    hucksters and peddlers
    mail handlers, except post office
    billing clerks
    library attendants and assistants
    ticket, station, and express agents
    newspaper carriers and vendors
    dispatchers and starters, vehicle
    messengers and office helpers
    welfare service aides
    enumerators and interviewers
    collectors, bill and account
    Blue Collar
    checkers, examiners, and inspectors - manu
    printing craft workers
    telephone and line installers and repairer
    radio and television repairers 
    photographic process workers
    data processing machine repairers
    office machine repairers
    * The data for this table come from Table B20, Bureau of Labor Statistics (1982).

    Figure 1

    Optimistic and pessimistic theories about the quality of jobs in the information economy can be evaluated by examining how good jobs and bad jobs are actually distributed across the five strata of the information work force. For example, the most optimistic stories suggest that the proportion of professional jobs ( good jobs) should be much larger than the proportion of jobs in any other stratum. Further, the relative number of poor quality jobs should decline from one stratum to the next. This story, which we call "professional dominance" can be illustrated graphically (see Figure #1). We can compare the shape of the curve which characterizes the actual empirical distribution of the relative size of the strata with the profes­sional dominance curve. We refer to the graphs associated with a particular story about the relative size of the strata as a "theoretical distribution." 

    These theoretical distributions are static. Most stories of occupational change in the information labor force are dynamic; the mix of jobs changes over time. Consequently, we will examine the fit of theoretical distributions to empirical distributions over time. For example, no one argues that professional jobs dominated the information workforce in 1900. But many authors imply that they do today. Thus we can examine whether the empirical distribution of jobs in the information workforce has been moving towards professional dominance between 1900 and 1980.

    In Figure #1 we graph three alternative theoretical distributions for the three stories which are most commonly discussed in the literature about information work: 

  • (a) professional jobs dominate the information work force: Most information workers hold highly professionalized jobs like accountants, scientists and lawyers; a smaller group holds semi-professional jobs such as engineers and school teachers.
  • (b) middle level jobs dominate the information work force: Most information workers hold middle level jobs. We have portrayed the alternative where the most dominant middle level jobs are the sales and supervisory jobs. The other information occupations are less numerous than these occupations.
  • (c) lower level jobs dominate the information work force: Most information workers hold lower level clerical jobs. The higher level occupational strata employ relatively fewer people, inversely proportional to their status.
  • Thus, these theoretical distributions range from optimistic portraits of an information sector which is characterized by generally good (professional) jobs to a pessimistic portrait of the same sector characterized by predominantly poorer (clerical) jobs. These distributions are static, but some of the theories describe information work as the end point of important "trends" -- that the current distribution of information occupations is moving toward one of these three distributions. In the next section we examine how well the data about employment fit these three theories. 

    Scale, Grow­th and Structure of the Information Work force
     The information occupations mushroomed in size from 17% of the United States work­ force in 1900 to over 50% in 1980 (Table #4b). In this period, the United States moved from a dominantly agricultural economy to a services economy. 

    TABLE  4a
            Distribution of Occupational Strata
            Within the United States Information Sector: 
            Employment levels - By Sector
               Full   Semi-  Sales &   Clerks   Blue |   Total     Total
               Prof    Prof  Superv.          Collar |Workforce     Info
       1980 ¦  3328   16997     8047    21583   1249 |   99303     50108
       1970 ¦  2152    9995     6796    16600   1192 |   80603     36735
       1960 ¦  1338    7553     5803    12286    913 |   67990     27893
       1950 ¦  1065    5473     4900     9508    786 |   59230     21732
       1940 ¦   687    4195     3773     6992    506 |   51742     16153
       1930 ¦   595    3972     3469     5952    469 |   48686     14457
       1920 ¦   437    2803     2586     4438    344 |   42206     10607
       1910 ¦   338    2296     2126     2933    288 |   37291      7981
       1900 ¦   280    1395     1557     1604    209 |   29030      5045
    Table 4b
            Occupational Strata as Per Cent
            of Information Workforce
               Full   Semi-  Sales &   Clerks   Blue             
               Prof    Prof  Superv.          Collar            
       1980     6.5%   33.2%    15.7%    42.4%   2.4%                    
       1970     5.9%   27.2%    18.5%    45.2%   3.2%           
       1960     4.8%   27.1%    20.8%    44.0%   3.3%          
       1950     4.9%   25.2%    22.5%    43.8%   3.6%            
       1940     4.3%   26.0%    23.4%    43.3%   3.1%           
       1930     4.1%   27.5%    24.0%    41.2%   3.2%         
       1920     4.1%   26.4%    24.4%    41.8%   3.2%          
       1910     4.2%   28.8%    26.6%    36.8%   3.6%            
       1900     5.6%   27.7%    30.9%    31.8%   4.1%          
    *       1900-1970 data in this table come from Series D-182 through D-682, Bureau of the 
    CensusBureau of the Census (1976:139-145).
     1980 data come from Table B20, Bureau of Labor Statistics (1982:664-667) and Table 
    276 Bureau of the Census (1984)and selected to match the smaller number of 
    occupational categories from this time series. The 
    resulting 1980 data does not not match other tables, 
    but is comparable with the data in this table.
            TABLE  5 
            Distribution of Occupational Strata 
            Within the United States' Information Sector: 
            Occupational Strata as Per Cent
            of Total United States' Workforce
                                                       ¦Information as 
               Full   Semi-  Sales &   Clerks   Blue   ¦% of Total
               Prof    Prof  Superv.          Collar   ¦Work Force
       1980     3.4%   17.1%     8.1%    21.7%   1.3%  ¦ 50.5        
       1970     2.7%   12.4%     8.4%    20.6%   1.5%  ¦ 45.6
       1960     2.0%   11.1%     8.5%    18.1%   1.3%  ¦ 41.0
       1950     1.8%    9.2%     8.3%    16.1%   1.3%  ¦ 36.7
       1940     1.3%    8.1%     7.3%    13.5%   1.0%  ¦ 31.2
       1930     1.2%    8.2%     7.1%    12.2%   1.0%  ¦ 29.7
       1920     1.0%    6.6%     6.1%    10.5%   0.8%  ¦ 25.1
       1910     0.9%    6.2%     5.7%     7.9%   0.8%  ¦ 21.4
       1900     1.0%    4.8%     5.4%     5.5%   0.7%  ¦ 17.4
    *       1900-1970 data in this table come from Series D-182 through D-682, Bureau of the 
    CensusBureau of the Census (1976:139-145).
     1980 data come from Table B20, Bureau of Labor Statistics (1982:664-667) and Table 
    276 Bureau of the Census (1984)and selected to match the smaller number of 
    occupational categories from this time series. The 
    resulting 1980 data does not not match other tables, 
    but is comparable with the data in this table.
    Table 6
                            Size  Occupational Strata 
            Within the United States Information Sector, 1980                       
                                        Number        Per cent of
                                         (000)       Information
         Occupational Stratum
         Professional                    3,501           6.5 
         Semi-professional              17,762          33.2 
         Sales and Supervisory           9,044          16.9 
         Clerical                       21,130          39.5 
         Blue Collar                     2,085           3.9 
        Information Workforce:          53,522*          53.9 
            Total Workforce             99,303
    *The information work force total is larger than that reported in Table 1. 
    See Appendix A for details.
    Table #4b reports the percentage of workers in each stratum of the information work force at the Federal level between 1900 and 1980. Professional and semi-professional workers have composed a remarkably stable proportion of the information work force between 1900 and 1970: 4-6% and 25-29% respectively. Together, these professional strata formed a minority of the information workforce -- between 30% and 33%. Blue collar information workers slowly declined from about 4% in 1900 to 3% in 1960. 

    Since 1900, clerks have been the largest strata of information workers. They rose from 32% of the information work force in 1900 to about 42% in 1920. Clerical employment grew at a slightly faster rate than overall employment in the information sector between 1920 and 1970, when it peaked at about 45%. Simultaneously, the higher level sales and supervisory stratum shrank from 31% to 19% of the information work force. These two lower white collar strata formed the majority of information workers, and the relative number of mid level jobs has declined significantly since 1900 while clerical jobs have risen.

    There are signs of a different pattern of occupational growth and decline between 1970 and 1980. As in the previous 70 years, the number of information workers continued to increase in all five strata. However, the relative size of some strata shifted in small but noticeable ways (Table #4b). The proportion of highly professionalized and semi-professional workers rose to about 40% of the information workforce. Clerks declined somewhat in relative size to 42%, although they continued to grow as a proportion of the total work force. In addition, the strata of supervisors and higher level sales personnel continued to decline. And the blue collar strata continued to decline in relative size.

    Overall, the information work force mushroomed in the last 80 years so by 1980 it composed over 50% of the work force. Its lower level white collar workers outnumber professional workers. But this distribution has been masked by the steady growth of information sector jobs in the highly professional and professional strata, as well as clerical jobs. The occupational stratum between clerks and semi-professionals -- the supervisory and upper level sales workers -- has steadily declined in relative size. In 1900 it was twice as large, in relative size, as it was in 1980. 

    Two lower strata -- clerks and sales and supervisory workers -- account for 55% of the jobs in the information sector. Semi-professionals also account for a major fraction of the jobs -- about 33%. But there is a large depression in the distribution of jobs across the three strata. The number of sales and supervisory jobs is about half the size of the semi-professional stratum and less than half the size of the clerical stratum. 

    The relative size of the five occupational strata does not fit any of the three theoretical models which we discussed above -- professional, middle or lower level jobs dominating in a monotonic pattern. This distribution comes as a surprise. We expected that either middle or clerical strata dominated the information work force. We did not have strong expectations about trends in the relative size of strata. We were astounded by the stability of the relative size of these strata within the inform­ation sector between 1900 and 1970. And we were particularly surprised by the steady and precipitous decline in relative size of the sales and supervisory stratum. The kinked occupational distributions provide interesting evidence for segmentation in information labor markets (Berger and Piore, 1980). We examine the segmented character of information labor markets in the next section. 

    The Segmentati­on of Information Labor Markets
     In this section we argue that information labor markets are divided into four relatively impermeable segments: (a) clerical work; (b) supervisory and higher level sales jobs, and (c) the two strata of professional jobs; (d) blue collar jobs. Our thinking has been strongly influenced by dual labor market theorists, even though their emphasis has often been very different. Dual labor market theorists have primarily focussed on the segmentation between jobs in the primary and secondary labor markets. Most information sector jobs are primary sector jobs. Dual labor market theorists often mention that jobs in the primary labor markets are also segmented, but they rarely examine segmentation in the primary sector. The key element of segmentation pat­terns in labor markets is that significant structural barriers inhibit people's mobility from one labor market to another - or, in our case, one occupational stratum to another (Berger and Piore, 1980). 

    Our arguments that the information labor markets are segmented are very simple. Most occupations in the professional and semi-professional strata are segmented primarily by special education and licensing requirements. All of the occupations which we have listed as professions (Table #2) require specialized college or post-graduate degrees. Some, professions, such as law and medicine, impose stringent professional licensing requirements. 

    Many of the semi-professions have similar barriers that inhibit people from "moving up" into them, although some of them do not require formal training or licensing. "Writers, artists, and entertainers" are perhaps the most intriguing occupations since theoretically anyone can write, paint or play. But managers and administrators account for the majority of semi-professional jobs in the United States' economy. 

    Our data reveal a less apparent structural barrier that seals many women into clerical careers. Women who wish to rise from clerical jobs to "something better" often lack the special education and credentials required for fully professional and semi­professional jobs. Management is the primary semi-professional occupation which does not require college credentialling. Historically managerial jobs have been male dominated, while many clerical specialties became female dominated by 1900. While managerial jobs have become more open to women in the last decade, they have remained male dominated (Taeuber and Valdisera, 1986). Some structural barriers still limit women's mobility from clerical jobs to managerial jobs (Kanter, 1977).

    The supervisory and higher level sales jobs are within reach for a larger number of clerks since they do not have significant educational and credentialing barriers. Some pay more than semi-professional jobs. But they have steadily declined in relative proportion from approximately equal to the number of clerical jobs in 1900 to be approximately one third the number of clerical jobs in 1980. Our argument is based on the relative number of slots for clerks in the next stratum: clerks who want to "move up" in the information sector will see a significantly smaller num­ber of jobs they can qualify for.

    Several questions might weaken this numerical argument. We do not believe that the these counter-arguments have substantial force. But they are worth noting. We have no specific data about the occupational mobility of workers between these strata in 1900 or 1980 or at times in between. An argument that two occupational strata are segmented depends upon mobility data. Indirect evidence indicates that clerks often do not move into other kinds of careers (Kanter, 1977). However, occupational mobility data would not reveal the num­ber of clerks who sought jobs in higher level information sector strata, but did not find them or were not recruited into them. Nor do we know the extent to which clerks would find the various jobs in the stratum attractive and would have sought them if they thought they were more readily available.

    While sex discrimination in hiring has continued to impede women's access to certain professional and semi-professional jobs, their conceptions of attractive jobs also influences where they work. The higher strata jobs were probably not equally desirable to clerks at all times between 1900 and 1980. Clerical perceptions have almost certainly changed between 1900 and 1980. In the early 20th century, most clerks were men and they may have viewed sales jobs and supervisory jobs which require more initiative than many clerical jobs as acceptable, if not attractive. Traditional women, who dominated the clerical work force by the 1950s, may have found these higher level sales and supervi­sory jobs less attractive. However, the women's movement has influenced women's conceptions of acceptable careers in the last 15 years. As a consequence, we suspect that more women clerks would find these jobs attractive today than in 1960, if they could move into them. However, these "better jobs" have been declining precisely during the time that they could become a move up for many clerks.

    While the information work force became more segmented, access to many information jobs has become more difficult. The educational and credentialing requirements for jobs at all strata in the information sector have generally tightened during this century. College degrees were once the prerequisites for only the most specialized and technical or most professionalized occupations. Since World War II there has been a form of credential inflation; bachelors (and sometimes graduate) degrees have become commonplace requirements for many semi-professional jobs. Some employers are beginning to selectively hire people with college degrees into clerical jobs. Clerical work is probably the primary occupational opportunity for the majority of college educated women with degrees in the liberal arts who do not acquire professional or graduate degrees. 

    We found that the information sector continues to provide a majority of jobs in the United States work force. Information sector jobs vary widely in quality. We have characterized the quality of jobs by one dimension: location in the status hierarchy of occupations. This simplified conception captures important aspects of pay, status, autonomy and other working conditions. Relatively few information sector jobs are fully professional, and clerical jobs form the largest occupational stratum. When we examined the growth of the various strata between 1900 and 1980, we found that clerical jobs became more dominant, not less dominant.

    We do not have direct data about actual occupational mobility. But our data suggest that the mobility of clerks is significantly limited by structural features of the information sector. The information sector is internally segmented, not just differentiated. The information sector will grow somewhat in overall importance, and clerical jobs will continue to grow in absolute and relative size. 

    It is easy to represent the information worker as a male professional such as an accountant, urban planner or engineer. This representation misleads. A female clerk is more accurate, although no stratum is so dominant that the other strata can be ignored. Contrary to the argument that the information work force is becoming professionalized through the use of information technologies (Giuliano, 1982), we have observed a steady growth in the relative size of the clerical stratum between 1900 and 1970. The growth of the semi-professional stratum has been less marked. It was relatively stable beween 1900 and 1970, compared with the other strata in the information sector, and it swelled dispropor­tionately between 1970 and 1980. 

    It is ironic that clerical jobs are still expanding when the educational level of women, as measured by the number of college degrees awarded, is at an all time high. Our own observations suggests that certain clerical jobs, like secretaries and bookkeepers, have usually become more varied as a byproduct of computer technologies and higher education. In fact, some managers report quite happily that they are now recruiting college educated women into clerical jobs which formerly drew only high school or junior college graduates. However, other clerical jobs, such as cashiers and counter clerks may remain relatively routinized, even when they have automated support through specialized point of sales terminals.

    We had not expected to find the information sector structured like a dual labor market for such a long period of time. Because we focus on the information sector, our data do not tell us about key aspects of dual labor markets, such as the employment of skilled craftsmen or even of mobility between other sectors and the information sector. Moreover, we do not have the kind of income data to definitely answer key questions about "the declining middle class (Lerman and Salzman, 1987)." ­These questions were outside the scope of this study. Even so, our data lend support for the declining middle thesis. Moreover, our data cast doubt on arguments that office workers are becoming more para­professionalized and professionalized (Noyelle, 1987). Based on our own field studies of computerization in white collar work, we believe that office work is becoming more skilled (Iacono and Kling, 1987). Our data doesn't show substantially more job openings for office workers to enter semi-professional and profes­sional occupa­tions (Table #4). These jobs normally require college degrees. Even the high skilled information jobs above the clerical stratum which do not require college degrees are in relatively short supply.

    In a recent article, Robert Reich (1989) identifies three major segments of the United States labor force which parallel our own categories. His three segments account for about 75% of United States' employment: symbolic analysts (lawyers, investment bankers, management consultants, research scientists, academics, etc.); (b) providers of routine production services (clerks); and (c) providers of routine personal services (barbers, retail sales personnel, cab drivers). He argues that the United States' location in international markets makes the jobs of symbolic analysts relatively valuable and well paid. In contrast, providers of routine production services are often competing with low wage labor elsewhere in the world. However, providers of personal services have captive local markets, and can fare better than clerks. After all, one doesn't fly to Seoul just for a haircut or a cab ride, even if it is cheap! But a publisher may well have a book typeset in South East Asia, and thus displace clerical jobs in the United States. Reich's symbolic analysts parallels our professional and semi-professional information workers. His loosely argued thesis suggests an important dyna­mism which may reinforce the dual structure of information labor markets. It certainly merits further investigation.

    We have not addressed the question whether massive technological change, particularly computerization and the use of advanced telecommunications systems, will alter the structure of the information occupations. Some economists predict major declines in the size of the clerical work force. But these projections rest on studies of efficiency gains on narrowly defined tasks and simple assumptions about the substitution of capital for labor.

    As the information sector expanded, it took on many characteristics of the overall economy: jobs that vary significantly in terms of pay, status, and power. Moreover, its internal divisions reflect patterns of segmentation that have developed elsewhere in the society. Overall, the information sector has become sufficiently large that is not an alterna­tive to the dominant social order -- it simply reproduces many of its features. 

    Bell, Daniel. 1973. The Coming of Post In­dustrial Society: A Venture in Social Forecasting, New York: Basic Books.

    Bell, Daniel. 1981. "The Social Framework of the Information Society", The Microelectronics Revolution: The Complete Guide to the New Technology and Its Impact on Society, ed. by Tom Forester, 500-550. Cambridge, Ma. MIT Press. 

    Berger, Suzanne and Piore, Michael. 1980. Dualism and Discontinuity in Industrial Societies, Cambridge University Press, Cambridge, London, New York.

    Bureau of Labor Statistics. 1982. Labor Force Statistics Derived From the Current Population Survey: A Databook, Volume 1; Bulletin 2096. Washington. 

    Bureau of the Census. 1976. Bicentennial Edition, Historical Statistics of the United States, Colonial Times to 1970, Part 1. U.S. Department of Commerce. Washington. 

    Bureau of the Census. 1984 1980 Census of the Population, De­tailed Population Characteristics, United States Summary, Section A. U.S. Washington.

    Bureau of the Census, 1985. Statistical Abstract of the United States 1986, 106th Ed­ition. Washing­ton.

    Burris, Beverly. 1983. No Room at the Top, New York: Praeger.

    Castells, Manuel. 1984. "Towards the Informational City? High Technology, Economic Change, and Spatial Structure : Some Exploratory Hypotheses Institute of Urban and Regional Development, University of California. Berkeley, Ca.

    Castells, Manuel. 1985. "High Technology, Economic Restructuring, and the Urban-Regional Process in the Unit­ed States" in High Technology, Space, and Society. Manuel Castells (ed.). Beverley Hills, CA: Sage. 

    Castells, Manuel. 1989. The Informational City: Information Technology, Economic Restructuring, and the Urban-regional Process. Cambridge, MA : B. Blackwell.

    Cooper, Michael D. 1983. "The Structure of the Information Economy", Information Processing and Management, 19:9-26.

    Dunlop Charles and Rob Kling. (in press). "Computerization and the Transfor­mation of Work" in Computerization and Controversy: Value Conflicts and Social Choices. Charles Dunlop and Rob Kling (Ed.) San Diego, Academic Press. 

    Fishman, Robert 1987. Bourgeois Utopias: The Rise and Fall of Suburbia. New York: Basic Books.

    Forester, Tom. 1980. The Microelectronics Revolution: The Complete Guide to the New Technology and Its Impact on Society, Cambridge, MA. MIT press.

    Forester, Tom. 1987. High Tech Society: The Story of the Information Technology Revolu­tion , Cambridge, MA. MIT Press. 
    "Gender wage gap narrows, but women still shortchanged", 1987. Orange County Regist­er, page 1, Friday September 4. 

    Ginzberg, Eli, Thierry Noyelle, and Thomas Stanback. 1986 Technology and Employment. Boulder Colo., Westview Press.

    Giuliano, Vincent. 1982. "The Mechanization of Office Work", Scientific American, 247(3)(September):148-164.

    Goldthorpe, John H. and Keith Hope. 1974. The Social Grading of Occupations: A New Approach and Scale. Oxford, Clarendon Press.

    Gordon, David M., Richard Edwards, and Michael Reich. 1982. Segmented Work, Divided Workers: The Historical Transformation of Labor in the United States. Cambridge [Cam­bridgeshire]. New York: Cambridge University Press.

    Gregory, Judith and Karen Nussbaum. 1982. "Race Against Time: An Analysis of the Trends in Office Automation and the Impact on the Office Work Force" Office: Technology and People, 1:197-236.

    Hartmann, Heidi, Robert E. Kraut and Louise A. Tilly, ed. 1986. Computer Chips and Paper Clips: Technology and Women's Employment. Volume I. Washington, D.C., National Academy Press.

    Howard, Robert. 1985. Brave New Workplace. New York: Viking.

    Hunt, H. Allen and Timothy L. Hunt. 1987. "The Impact of Technological Change" in Computer Chips and Paper Clips: Technology and Women's Employment: Volume II -- Case Studies and Policy Perspectives. Heidi L. Hartmann (Ed.) Washington, D.C., National Academy Press.

    Huppes, Tjerk. 1987. The Western Edge: Work and Management in the Information Age. Boston, Kluwer Academic Publishers 1987.

    Iacono, Suzanne and Rob Kling. 1987. "Changing Office Technolo­gies and Transformations of Clerical Jobs." in Technology and the Transformation of White Collar Work. ed. Robert Kraut. Hillsdale, NJ: Erlbaum.

    Kanter, Rosabeth Moss. 1977. Men and Women of the Corporation. New York: Basic Books.

    Katz, Raul L.. 1986. "Measurement and Cross-National Comparisons of the Information Work­ Force", The Information Society, 4(4):231-277.

    Kling, Rob, Suzanne Iacono and Joey George (1990). "Occupa­tional Power, Patterns of Desktop Computer Use, and Quality of Worklife" in Desktop Information Technology: Organizational Worklife in the 1990s. Kate M. Kaiser and Hans J. Oppeland (eds.) Elsevier/­North Holland. Amsterdam, 1990.

    Kling Rob and Clark Turner. (in press). "The Structure of the Information Econo­my." in Post Suburban California: The Transformation of Postwar Orange County, R. Kling, M. Poster and S. Olin (Ed.) University of California Press. 

    Knight, Richard V. 1986. "The Advanced Industrial Metropolis: A new Type of World City." in The Future of the Metropolis: Economic Aspects. eds. Hans-Jurgen Ewers, John B. Goddard, and Horst Matzerath. New York: Walter de Gruyter. 

    Kumar, Krishan. 1978. Prophecy and Progress: The Sociology of In­dust­rial and Post-Industrial Society. Allen Lane, Penguin Books Ltd. 

    Kuttner, Bob. 1983. "The Declining Middle", At­lantic Monthly, 252(1)(July):60-72.

    Leontief, Wassily and Faye Duchin. 1983. The Impacts of Automation on Employment, 1963-2000. New York: Institute for Economic Analysis, New York University. 

    Leontief, Wassily W., and Faye Duchin. 1986. The Future Impact of Automation on Workers. New York: Oxford University Press.

    Lerman, Robert and Harold Salzman 1987. Deskilling and Declass­ing: Whither the Middle Stratum? Boston, MA: Center for Applied Social Science, Boston University.

    Luke, Timothy and Stephen K. White. 1985. "Critical Theory, the Informational Revolution, and an Ecological Path to Modernity" in Critical Theory and Public Life, John Forster (ed.). Cambridge, MA. MIT Press.

    Lyon, David. 1986. "From 'Post-Industrialism' to 'Information Soc­iety': A New Social Transfor­mation?", Sociology, 20(4)­(November): 577-588

    Lyon, David. 1988. The Information Society: Issues and Illusions. New York: Basil Blackw­ell.

    Markusen, Anne Roell and Robin Bloch. 1985. "Defensive Cities: Military Spending, High Technology, and Human settlements." in High Technology, Space, and Society. ed. Manuel Castells Beverley Hills, CA: Sage. 

    Mills, C. Wright. 1951. White Collar: The American Middle Classes. New York: Oxford University Press.

    Montagna, David. 1977. Occupations and Society: Towards a Sociol­ogy of the Labor Market. New York: John Wiley.

    Mowshowitz, Abbe. 1986. "The Social Dimensions of Office Automation." In Advances in Computers 25, ed. Marshall Yovits, New York: Academic Press.

    Naisbitt, John. 1984. Megatrends: Ten Directions Transforming Our Lives. New York: Warner Books.

    Noyelle, Thierry J. 1987. Beyond Industrial Dualism: Market and Job Segmentation in the New Economy. Boulder, CO: Westview Press.

    Parker, Edwin. 1981. "Information Services and Economic Growth." The Informa­tion Society, 1(1):71-78.

    Perrolle, Judith A. 1987. Computers and Social Change: Information, Property, and Power. Belmont, CA: Wadsworth Pub. Co.

    Porat, Marc Uri. 1977. The Information Economy: Definition and Measurement United States Office of Technology Special Publication 77-12(1), Washington: Department of Commerce, Office of Telecommunications.

    Reich, Robert. 1989. "As the World Turns: United States Inequality Keeps on Rising." New Republic 3876, May 1:23, 26-28.

    Ritzer, George. 1977. Working: Conflict and Change. (2nd ed.). Englewood Cliffs, NJ.: Prentice Hall.

    Rosenberg, Richard. 1986. Computers and the Information Society. New York: John Wiley.

    Scott, Allen. 1988. Metropolis: From the Division of Labor to Urban Form. Berkeley: Univer­sity of California Press.

    Stonier, Tom. 1983. The Wealth of Information: A Profile of the Post-Industrial Economy, London: Thames Methuen.

    Strassman, Paul. 1985. Information Payoff: The Transformation of Work in the Electronic Age, New York: Free Press. 

    Taeuber, Cynthia and Valdisera, Victor. 1986. Women in the Amer­ican Economy, Current Population Reports, Special Studies Series P-23, No. 146, November. Washington: Department of Commerce.

    Zuboff, Shoshana. 1988. In the Age of the Smart Machine: The Future of Work and Power. New York: Basic Books.



    Measuring the Scope of the Information Workforce
    In his landmark study which has provided the most widely used basis for defining and measuring the size of the information economy, Marc Porat realized that, "[s]tating precisely who is an information worker and who is not is a risky proposition (Porat, 1977:105)." He identified information workers by asking, "[w]hich occupations are primarily engaged in the production, processing, or distribution of information as the output, and which occupations perform information processing tasks as activities ancillary to the primary function (Porat, 1977:105)?" We agree that an answer to this question yields a meaningful list of information jobs.

    We based our typology of information workers on Porat's. We began with his list of information occupations to produce a list which is substantially similar to his. Porat identifies 188 information occupations, while we appear to identify 83. This is only an indication of the finer occupational distinctions in Porat's list as many of our single occupational titles are composed of some combination of separate titles from his list. We match Porat's list very closely.

    Porat's "Knowledge Producers" consists of "Scientific and Technical workers" and "Private Information Services". We do not match Porat's occupations "mathematical scientists" or its sub category "actuaries," "farm management advisors" and "home management advisors" Porat's "Knowledge Distributors" are all matched to occupations on our list. "School administrators" are listed but no numbers are provided for us. We match all of Porat's "Information Processors" with the exceptions of "health record technicians" and "railroad conductors" which were not listed in our data. "Motion picture projectionists" is included in our data. 

    We match "information Workers" except for "sign painters" which were not listed in our data, and "data processing machine repairers" which were listed. Note that "radio operators" is included in our count but in the non-information sector under "technicians, except health, science and engineering".

    Porat (1977) uses the alternative "restrictive" and "inclusive" definitions of information workers. Basically, he believes that 28 of his 188 listed occupations are "mixed" in nature and he is uncomfortable with including them as wholly information occupations. He thus allocates the "ambiguous" occupations proportionately to separate sectors. For example, he allocates "Physicians" 50% to information and 50% to service under his "inclusive" definition - and 100% to service under his "restrictive" definition.

    Porat reports his data on the growth of the information workforce exclusively in his [often reproduced] graph of the four sector aggregation of the U.S. workforce 1860-1980 "using median estimates". These "median estimates" mean that the data points on the graph are the median of the "restrictive" and "inclusive" figures for the given year.

    Our own approach to counting the number of information workers is more straightforward. We include Porat's 28 "ambiguous" occupations as information occupations without ambiguity. We believe it is reasonable to include them, for example, "physicians" and "Registered Nurses" appear to be primarily information workers. We also believe that "sales clerks" increasingly perform roles as information workers, for instance, as operators of "point of sale" terminals. Similarly, "miscellaneous clerical" workers (such as general office clerks and medical insurance clerks) are generally employed to enter, file, or move information for their employers. We perceive the managerial occupations to be highly information oriented and allocate them as information occupations. In the blue collar stratum, we identify "inspectors" and "examiners" to be primarily information occupations as the titles imply.



    Notes on Sources of Data and Methods
    We drew our data from the following sources: 

  • Bicentennial Edition, Historical Statistics of the United States, Colonial Times to 1970, Part 1; U.S. Department of Commerce, Bureau of the Census. Table: "Detailed Occupations of the Economically Active Population: 1900 to 1970," series D 233-682 (pps. 140-145). [Federal Time series by occupation 1900-1970. We use it to carry through our information occupations by level 1900-1970] 
  • 1980 Census of the Population, Detailed Population Characteristics, United States Summary, Section A: U.S.; Table: 276 - "Detailed Occupation of the Experienced Civilian Labor Force and Employed Persons by Sex: 1980 and 1970," pps. 1-166 to 1-175.[Federal Employment Data for 1970 and 1980]
  • Labor Force Statistics Derived From the Current Population Survey: A Databook, Volume 1; U.S. Department of Labor, Bureau of Labor Statistics, Bulletin 2096, September 1982, Table: B-20, "Employed Persons by detailed occupation, sex, and race, 1972-81", pps. 664-667.[More Federal occupational data for 1980.]
  • Statistical Abstract of the United States 1986, 106th Edition; U.S. Department of Commerce, Bureau of the Census, December 1985.[Data at the Federal level on employment for 1970 and 1980]
  • Occupational Comparisons: 1900-1980
    In assembling our historical series of Federal data for 1900-1980 displaying the 5 strata of the information workforce, we relied primarily upon the Historical Statistics of the United States (Series D-182 through D-682), but were forced to use data from other sources to fill in some unexplainable gaps in the series and to add data for 1980. (For example, teachers and nurses are counted for all years except for 1970!) We also dealt with some apparent reclassifications of occupations across categories for given years. 

    Since our main focus is to track the proportions of workers within a given information occupation stratum, our job was simplified. Even if the job was reclassified with others, as long as they all remain within the same occupational stratum, the crucial total is correct. We explain key examples of our calculations below. We used Historical Statistics of the United States data for 1900-1970 unless otherwise noted. We used Labor Force Statistics as our source for 1980 data, unless otherwise noted.

    In the professional stratum, there are no estimates for "teachers, college and university" in 1970. We estimate employment for this year from Statistical Abstract for the United States. The estimates for "physicians" from Historical Statistics of the United States include "osteopaths" before 1960. They include "chiropractors" and "therapists and healers" for 1900. "Life and physical scientists" are a combination of the Historical Statistics of the United States categories of "chemists" and "natural scientists, n.e.c." for 1900-1970. The category "operations and systems researchers" is not listed in Historical Statistics of the United States, but we obtained data from the Census of Population for 1970 and 1980.

    In the semi-professional stratum, the number of "all other managers and administrators" had to be calculated from Historical Statistics of the United States by subtracting the number of all managerial occupations already included (10 in all) in our list from the total. "Teachers, except college and university" were constructed analogously to "teachers, college and university" above in the professional stratum. The category "all other professional and technical" has no entry for 1970 and the data gave us no basis for estimate the size of this occupational category in 1970. We left it as 0. But we estimate the size of this occupational category for 1980 from Labor Force Statistics which includes not only "all other ..." but also the semi-professional categories "research workers" and "vocational and educational counselors" which were not counted separately in the Historical Statistics of the United States data. 

    The "writers, artists, and entertainers" category is a composite for both the Historical Statistics of the United States and Labor Force Statistics estimates. From the Historical Statistics in the United States tables, it consists of: "actors and actresses", "dancers and dancing teachers", "entertainers, n.e.c.", "artists and art teachers", "authors", "editors and reporters", "designers", "musicians and music teachers", and "photographers". From the Labor Force Statistics table we derive the figure from the difference of the "writers, artists, and entertainers" and the excluded sub-category "athletes and kindred". The category "computer specialists" is not included in our Historical Statistics in the United States data. We estimate 1970 and 1980 employment from Statistical Abstract for the United States consisting of the sum of the estimates for "computer systems analysts and scientists" and "computer programmers".

    The "sales representatives" and "sales workers, n.o.m." (clerks and low level sales) are matched in the Historical Statistics in the United States data with "salesmen and sales clerks (n.e.c.)"; manufacturing and wholesale, and "salesmen and sales clerks (n.e.c.)"; retail, respectively. Sales representatives are upper level sales workers and sales workers are clerical workers, yet the occupations are aggregated into one occupation as "salesmen and sales clerks, n.e.c." for 1900-1940. We estimate their proportions by the ratio from 1950 and enter the data separately.

    The "all other clerical" category comes from "clerical and kindred workers (n.e.c.)" in the Historical Statistics of the United States data. We carefully calculated an Labor Force Statistics figure by subtracting the sum of all other clerical counts(17 of them) that were included from the total clerical figure.

    Estimates of the Number of Clerks
    Our "clerical" stratum of information occupations differs from the usual list of clerical occupations. In trying to capture the lower end of the white collar information jobs, we include the low level sales workers, such as sales clerks (who are usually counted in other grosser categories, such as "sales workers"). This category is large in size relative to the others. For example, the Bureau of Labor Statistics (1980) reported about 2.4 million sales clerks in 1980. This produces a significant difference in the reported size of the "clerical" workforce. 

    Another explanation for possible differences between our count and other counts is the differing data sources. We were surprised (and unfortunately enlightened) to find disparities in different data sources for single occupational titles of some large magnitude. 


    Home | Table of Contents | Editorial Letters