Published by EH.NET (May 2007)
JoAnne Yates, Structuring the Information Age: Life Insurance and Technology in the Twentieth Century. Baltimore: Johns Hopkins University Press, 2005. x + 351 pp. $50 (hardcover), ISBN: 0-8018-8086-6.
Reviewed for EH.NET by Thomas Haigh, School of Information Studies, University of Wisconsin ? Milwaukee.
In Structuring the Information Age, JoAnne Yates tells the story of the use of computers and tabulating machines within the life insurance industry. Beginning with a snapshot of the already-mature life insurance industry during the first decade of the last century, Yates guides us through the introduction of successive waves of tabulating and punched card equipment, the interaction of life insurance firms with early producers of tabulating and computer technology, and the use of three generations of electronic computers large and small from the 1950s to the 1970s.
This timely and important work is the first scholarly history devoted to the use of information technology within a single American industry. Much less historical attention has been devoted to the use of computers than to the history of computer technologies and of computer manufacturing firms. But the economic and social importance of information technology stems more from its contributions to almost every industry, social activity and scientific discipline than from the relatively tiny industry devoted to designing and building computers. The internal dynamics of hardware and software firms provide fascinating stories but are rarely more than tangentially related to the experiences of computer users. Recent work by scholars such as James Cortada (2003), Martin Campbell-Kelly (1994), Jon Agar (2003) and me (Haigh 2001, 2001) has begun to explore the role of organizations and individuals as users of information technology. Structuring the Information Age makes a major contribution to this literature, establishing itself as the most important work to date on the use of information technology within business.
Yates tells the story of life insurance in parallel with the story of information technology within it. History has been defined as the study of change over time, but any historical work must also explain continuities: what didn’t change and why. By focusing on a single industry, Yates shows successive generations of information technology shaping, and being shaped by, the industry’s existing business processes, cultures and practices. Stories told from the viewpoint of the computer and its creators inevitably tend to produce tales of disruption and sudden upheaval. But, told from the viewpoint of a computer-using industry, this story is for the most part one of gradual and evolutionary change.
The first chapter, a sketch of the life insurance industry in the early 1900s and its origins, sets the stage for what follows. As well as ensuring the book is accessible to those without previous knowledge of the industry, it outlines a number of distinctive features of the life insurance business. This material supports later arguments that the specific course of technological development within the industry had a great deal to do with its history and culture. This was a conservative industry, heavily regulated and concerned with stability and efficiency rather than profit or rapid growth. Yet processing paperwork was its main business process, rather than a sideline as for a manufacturing industry, meaning that improvements to administrative efficiency promised a real boost to overall performance.
The use of tabulating machines by insurance companies from the 1890s onward is the subject of the next three chapters. Although a considerable amount has been written on the business and technical history of the tabulating machine industry (Campbell-Kelly 1989; Heide 1997), almost nothing had been published on their administrative application in America between their celebrated debut at the US Census of 1890 and their adoption on a massive scale by the newly established Social Security Administration circa 1935. Yates’ work here is the deepest analysis to date of their actual use in business. She is the first author to analyze what I consider a crucial transition: from application of punched card technology purely within the niche field of statistical tabulation for which it was originally designed (in this case, actuarial work within life insurance) to its general use for a much broader range of routine administrative tasks such as payroll, billing, and account balance calculation. Yates identifies early operational use of the machines by Phoenix Mutual as early as 1910 (pp. 46-47), long before manufacturers had begun to target applications of this kind by adding hardware features to support it.
When punched card machines were modified to better support administrative work, Yates places users, rather than supplier firms, in the vanguard. As in her award winning Business History Review paper (Yates 1993) she presents technological improvements to tabulating machines during the 1920s and 30s as a “co-evolution” between technology and use. The two most important advances during this era were the addition of printing capabilities and the extension of tabulating equipment to process letters as well as numbers. Both of these advances were pioneered by insurance firms, working closely with established suppliers and entrepreneurial innovators, sponsoring the development of new machines and even, in one case, purchasing a supplier firm.
Beginning with the 1920s, Yates pays considerable attention to the role of trade associations in the spread of technological practices within the life insurance industry. This too is an important contribution. While Phil Scranton and others have drawn attention to the historical development of industries as clusters of firms (Scranton 1997), business historians more often write as if each firm acted independently when introducing managerial, technological, and organizational innovations. Trade associations, in this case the Life Office Management Association, play a vital role in sharing experiences, discussing new ideas, and legitimating particular applications of technology. They also provide a bridge between managers within firms, equipment suppliers, and independent experts. LOMA set up formal working groups and committees to evaluate new technologies and formalize best practices for their use.
The second half of the book is about the computerization of the life insurance industry. Two chapters deal with early developments during the 1940s and 1950s, another with the third generation computer systems introduced in the 1960s and used into the 1970s, and a final narrative chapter presents two detailed case studies. The topics of these chapters mirror those already established: interactions between the users and producers of technology, incremental and evolutionary change, and the importance of trade associations in shaping technological adoption. The symmetry Yates achieves between her analyses of the punched card and computer eras captures one of her main insights: that user organizations experienced computer technology as an extension of well-established tabulating technologies rather than as a revolutionary break with the past. Indeed, this perception itself played an important role in shaping the ways in which computers were used and may therefore have been something of a self-fulfilling expectation. (While the continuities between the tabulating industry and the business computing industry have been well documented, less has been written about these continuities in use (Haigh 2001)).
Yates offers an extended discussion of Prudential Insurance’s Edmund Berkeley, one of the most colorful figures in the early history of computing. Berkeley was exposed to early computing efforts during the war, and returned to the world of insurance keen to apply the power of these new electronic marvels to insurance work. He worked closely with the designers of the Univac I, and pushed Prudential into becoming the first firm to order a commercial computer. (The order was later cancelled, but by then Berkeley had already embarked on an idiosyncratic career as author of the first popular guide to computing, publisher of the first computer journal, and founding secretary of what grew into the main association for computer science.) Berkeley, it must be said, is something of an exception in this book, as one of the few individuals granted much personal agency or a life story. His story has been told before, not least by Yates herself (Yates 1997), but it benefits here from its context as part of the longer history of technological innovation in the life insurance history.
Subsequent discussion returns the focus to trade groups, particularly the Society of Actuaries, in studying and standardizing the use of computers. Yates discusses the strategies taken by a number of firms large and small, illustrating differences in the kinds of computers ordered, the tasks to which they were applied, and the aggressiveness with which companies attempted to impose new business processes along with the new hardware.
These chapters also include some brief discussion of the organizational changes around the technology: feasibility studies, personnel issues, and worries about technological unemployment. The early-1960s transition from vacuum tube computers to transistorized machines such as the IBM 1401 appears here as another evolutionary development, as does the arrival of IBM’s hugely successful and technologically disruptive range of 360 series machines in the mid-1960s. While Yates follows some aspects of the story into the early 1970s, the narrative fizzles out without reaching any clearly demarcated stopping point or milestone event. Instead she shifts to case-study mode, recapitulating events of computer era through developments in two companies: New England Mutual Life and Aetna Life. These case studies, sourced in part from interviews with managers involved in events, help to bring some texture to the story and demonstrate the interplay of industry-wide trends with personal and cultural factors specific to individual firms.
Yates does an excellent job in meeting a key challenge facing anyone writing on the historical use of information technology: how to convincingly analyze technological capabilities without getting the book stuck in a swamp of model numbers and electronic widgets. Throughout the book, she provides just enough detail on the technological capabilities of different generations of tabulating machines and early computers to support her analysis of their business applications. As a result it should prove accessible to readers unstepped in the existing history of computing literature.
Stories familiar from the viewpoint of producer histories, such as the late-1950s success of IBM’s small but flexible 650 computer, take on a new aspect when seen from the viewpoint of user organizations. Yates provides a particular service by showing how identical machines were used in different ways by different firms. Users play a particularly important role in determining the social and organizational impact of computer technologies because of the huge flexibility of the underlying hardware and the consequent importance of software. Insurance companies wrote their own application programs and applied the machines to different areas of the business. Some aimed to consolidate processes, while others mimicked the existing paper-based procedures. While no individual user could exert the same influence over computer hardware in the 1960s that Berkeley enjoyed in the 1940s, the importance of software meant that individual firms continued to exercise an enormous amount of control over the functioning and organizational impact of their own computers. The success of the popular ’62 CFO package introduced by IBM for smaller firms showed that even companies using the same application software package could achieve very different results. In what may be the first historical analysis of the user community of a single mainframe software package, Yates finds that firms used the package in different ways, with some rewriting it to create their own versions. She demonstrates that software packages played an important role within the life insurance industry a good decade before the independent packaged software industry was established in the early 1970s. (For more on this issue, including discussion of an earlier paper by Yates on the topic (Yates 1995), see Haigh (2002).) It remains unclear whether the insurance industry was unique in this respect, or if an equally close examination of other industries will reveal similar examples of early software standards.
Yates’ methodologies and sensibilities are unmistakably those of business history rather than economic history. One fascinating aspect of the story summarized here is the willingness, indeed the eagerness, of hundreds of companies to order computers long before their economic viability had been established. Yates places less emphasis than I do on the symbolic value of computerization as an emblem of modernity, which I believe rapidly led to a situation in which no self-respecting firm could afford not to place an order for a computer. Did the sometimes frenzied investments made in computer technology during the 1950s, 60s and 70s ever pay off? Yates doesn’t know for sure, but neither does anyone else ? including the people who ordered the computers. But unlike other historians who have tackled the business use of computer technology, she did make a concerted effort to find some solid data on the topic. Numbers are crunched, industry averages calculated, and estimates made of the percentage of insurance premiums consumed by administrative overhead. The result: no clear link between computerization and efficiency within individual firms or within the industry as a whole. (This mirrors the findings of Paul Strassman (1997) and other observers of the so-called “productivity paradox” during the 1980s and early 1990s.) Believers in the strategic importance of computing might object that this focus on clerical cost reduction obscures what were often called the “intangible” benefits of flexibility, better managerial decision making, new products or services made possible by automation, and so on. But these benefits appear only rarely in the book, and Yates shows in several of the case studies that firms attempting to make more aggressive use of computer technology to integrate their operations or provide real-time interactive access to information tended to run into insurmountable technical obstacles during the 1950s and 60s.
Like most historians, Yates devotes little explicit discussion to methodological or theoretical questions. This is an observation of disciplinary norms rather than a criticism, and her narrative moves smoothly and convincingly throughout. She relies largely on archival sources, meaning that her evidence is rich but must face the same question as any business history based largely on case studies: how representative are these firms of the general population of institutions of the same kind? As mentioned previously, Yates goes a long way toward answering this question by exploring sources from a second kind of institution: trade associations. This gives a glimpse at the kinds of questions preoccupying administrative managers within life insurance firms at different points in time and of the stories they were telling each other about the power of technologies and systematization.
The one overt appearance of theory in the book is her invocation (pp. 4-5) of Anthony Giddens’ structuration theory as an underpinning justification for her approach and as the source of the book’s title. (Wanda Orlikowski, a collaborator of Yates on earlier work, is well known for applying structuration theory to the use of technologies within organizations.) Structuration theory aims to reconcile the freedom of individual action with the power and persistence of social structure. Its central insight is that social structures of all kinds constrain and guide individual actions, but are themselves constantly reproduced and minutely shifted by those actions. Though the potential relevance of this metatheory to the mutual shaping of institutions and practices is clear, Yates never explicitly comes back to it in the body of the work or in its conclusions to show more directly how it shaped her analysis and findings. As a theory of everything, structuration runs the risk being applicable to any situation without necessarily providing the observer with specific guidance on how to analyze its particulars.
Yates might have found a more directly related body of theory within the literature on “The New Institutionalism in Organizational Analysis” associated with Paul DiMaggio and Walter Powell (1991). As they argue in the introduction to this volume, the New Institutionalist approach has many parallels with structuration theory, but has the benefit of being directly concerned with the evolution of organizational form. Certainly Yates’ findings seem to support their central concept of institutional isomorphism: the idea that organizations within a field (in this case the life insurance industry) tend to grow alike in terms of organizational structure, culture, and practices. Powell and DiMaggio offer a useful set of mechanisms to explain this process: regulation and other external pressure, imitation, and most relevantly here, creation of shared cultural norms within professional communities.
Yates has discussed her methodological goals for the project more thoroughly in a companion essay, published recently in Enterprise and Society (2006). She argues that, while historians of technology have been talking about the social construction of technology and the importance of users for more than two decades, their conception of the user has almost always been that of an individual consumer. Yates insists instead on the importance of thinking of organizations as users of technology, an idea she refers to there as “broadening the demand-side turn.” She reports some resistance to this idea from historians of technology, captured in her recorded surprise at being asked by such an historian why there were no users in her narrative. Having been immersed in the business-school environment in which the idea of user-firms is well established, she wonders in turn why the unnamed historian insisted on focusing on individual people involved with computer technology (whom she would characterize as computer operators) when the relevant decisions were made by “an organization made up of many individuals with different roles and interactions” (Yates 2006, p. 434).
This observation crystallizes the strengths and limitations of Structuring the Information Age more clearly than any single statement within the book itself. Yates’ primary concern is with organizations as users of technology, and therefore she has written a story in which firms are the main actors: firms learn, firms make decisions, firms investigate technologies, and firms create and administer computer applications. But to ascribe such actions to a firm is as much of a simplification as to ascribe it to an individual. This is especially true in the case of software. Most computer application systems during the period Yates discusses were specified, designed and coded by computer staff within “user” firms, using hardware and software tools provided by vendors such as IBM. From the point of view of an IBM sales representative, both the insurance firm and its computer department are users and consumers of technology. But from the point of view of the programmers inside the firm, and their internal customers, the data processing department was a producer of IT systems. Whether a firm appears to be a user or producer of technology depends on where one stands.
Since firms are made of people, Yates inevitably addresses the actions of individuals, such as Berkeley. Issues of organizational politics and personal conflict appear at various points within the case studies, but are not really taken up in her broader analysis. The “roles and interactions” within the firms, which Yates rightly draws attention to, surface sometimes within her case studies but are not generalized into her overall conclusions. “Personnel issues” are dealt with only briefly, and presented (for example on pp. 161-67, 181-82) primarily to discuss the challenges of reassuring workers the computers would not eliminate their jobs and relocating any displaced clerks.
Yates thus downplays something that looms large in my own interpretation of these events: the emergence of new groups of technical and professional specialists within corporate society, each with their own identity and interests (Haigh 2003). Though Yates has a great deal to say about the evolution of the life insurance firm as an institution and its relationship to technology, she has almost nothing to say about the organization, management or culture of the new departments of data processing that accreted, deep inside life insurance firms, around the new machines. Machine operators, systems analysts, computer programmers, corporate accountants and middle managers matter a great deal as classes and communities within the corporate world, even if their individual agency is tightly constrained. (After all, the structuration theory Yates references in her title aims to show how ordinary people can shift social structures even as they reenact them.)
Studying occupational subcultures within the corporation provides an intermediate level of analysis, between firms and individuals, in which to explore the creation of new institutions within firms, such as the data processing department, and the emergence of new occupational groups such as the systems analyst, computer operator, or computer programmer. From Berkeley onward, actuaries, clerks and managers who joined projects to study or implement computer systems often found their careers taking unexpected detours into the new world of data processing. Indeed, the very study groups and industry associations she chronicles must have played an important role in establishing new identities and communities around the intersection of computers and life insurance. A culture of computer enthusiasm created a heterogeneous alliance devoted to the installation and expansion of computer systems. Yates’ predisposition to view technological adoption as the result of rational decision making by the firm as a whole may obscure the importance of these communities of computer enthusiasts within and between firms in pushing the new technology forward.
Today, American businesses employ more than ten million IT workers, and senior managers rely on computer specialists to create and oversee the systems that run every aspect of their firms’ core operations. Only teams of highly specialized technical staff can understand, still less modify, the behavior of these systems. We take this for granted today, but in the 1950s and 60s it represented an important shift in managerial society. As Yates showed in her earlier book, Control through Communication (1989), progressive American managers from the 1890s onward had embraced the personal mastery of new technologies such as graphs, written procedures, organizational charts and filing systems as symbols of modern management expertise. With computerization, however, detailed knowledge of administrative technologies was inevitably separated from general managers to create a new class of specialists. Yates mentions that some managers distrusted magnetic tape because, unlike punched cards, it contained no visible mark of the data within. But she does not further address the cultural or structural consequences of this shift of power within the managerial ranks.
Many computer personnel came to identify more closely with the “data processing profession” than with banking or insurance. The new ranks of technology managers and specialists, mediating between the practices and cultures of business and those of the computer room, faced a set of conflicts and challenges familiar to anyone who has leafed through a book of Dilbert cartoons. Managers complained about the perceived lack of loyalty of the computer staff, whose primary identity often formed around machines or skills rather than a particular organization or even industry. Identities and practices were knitted together across a range of industries by organizations such as the Data Processing Management Association and trade publications such as Datamation and Business Automation. This highlights one of the inevitable limitations of a single industry study. While the organization, identity and practices of data processing departments appear to have been quite stable across industries, Yates tells the story of information technology use within the life insurance industry as a self-contained narrative. Until similar studies are produced of other industries we will not know which characteristics of computer use here were exceptional and which merely mirrored broader trends. James W. Cortada (2003, 2006) has already published the first two in a projected three volume series of books surveying computer use in a series of industries, providing a complementary perspective.
These comments reflect a subject area so rich than no single study could begin to exhaust its potential. As historians come to grips with the business history of the end of the last century and the beginning of this one there will be few industry historians who can avoid the topic. Likewise, historians of technology dealing with the past half century will find rich pickings in the history of business administration and its systems. A flood of books describing information technology use in different industries will sooner or later appear, and their authors will find Structuring the Information Age an invaluable guide and model. The book is a significant landmark within the history of computing literature. I hope Yates succeeds in her stated aim of convincing historians that businesses can be creative users of technologies. We would all benefit if it can also serve what must have been an implicit aim: to remind business school faculty that history explains a great deal about how technology does and doesn’t work when applied within an industry.
Agar, Jon. 2003. The Government Machine. Cambridge, MA: MIT Press.
Campbell-Kelly, Martin. 1989. ICL: A Technical and Business History. New York: Oxford University Press.
Campbell-Kelly, Martin. 1994. The Railway Clearing House and Victorian Data Processing. In Information Acumen: The Understanding and Use of Knowledge in Modern Business, edited by L. Bud-Frierman. London: Routledge.
Cortada, James W. 2003. The Digital Hand: How Computers Changed the Work of American Manufacturing, Transportation, and Retail Industries. Oxford: Oxford University Press.
Cortada, James W. 2006. The Digital Hand, Volume 2: How Computers Changed the Work of American Financial, Telecommunications, Media, and Entertainment Industries. Oxford: Oxford University Press.
Haigh, Thomas. 2001. “The Chromium-Plated Tabulator: Institutionalizing an Electronic Revolution, 1954-1958.” IEEE Annals of the History of Computing 23 (4): 75-104.
Haigh, Thomas. 2001. “Inventing Information Systems: The Systems Men and the Computer, 1950-1968.” Business History Review 75 (1): 15-61.
Haigh, Thomas. 2002. “Software in the 1960s as Concept, Service, and Product.” IEEE Annals of the History of Computing 24 (1): 5-13.
Haigh, Thomas. 2003. “Technology, Information and Power: Managerial Technicians in Corporate America.” Ph.D. dissertation, History and Sociology of Science, University of Pennsylvania, Philadelphia.
Heide, Lars. 1997. “Shaping a Technology: American Punched Card Systems 1800-1914.” IEEE Annals of the History of Computing 19 (4): 28-41.
Powell, Walter W., and Paul J. DiMaggio, eds. 1991. The New Institutionalism in Organizational Analysis. Chicago: University of Chicago Press.
Scranton, Philip. 1997. Endless Novelty: Specialty Production and American Industrialization, 1865-1925. Princeton, NJ: Princeton University Press.
Strassmann, Paul. 1997. The Squandered Computer. New Canaan, CT: Information Economics Press.
Yates, JoAnne. 1989. Control through Communication: The Rise of System in American Management. Baltimore: Johns Hopkins University Press.
Yates, JoAnne. 1993. “Co-evolution of Information-processing Technology and Use: Interaction between the Life Insurance and Tabulating Industries.” Business History Review 67 (1): 1-51.
Yates, JoAnne. 1995. “Application Software for Insurance in the 1960s and Early 1970s.” Business and Economic History 24 (1): 123-134.
Yates, JoAnne. 1997. “Early Interactions between the Life Insurance and Computer Industries: The Prudential’s Edmund C. Berkeley.” IEEE Annals of the History of Computing 19 (3).
Yates, JoAnne. 2006. “How Business Enterprises Use Technology: Extending the Demand-Side Turn.” Enterprise and Society 7 (3): 422-455.