Hostname: page-component-cd9895bd7-lnqnp Total loading time: 0 Render date: 2024-12-24T17:56:02.293Z Has data issue: false hasContentIssue false
  • English
  • Français

Co-evolution of Information-Processing Technology and Use: Interaction between the Life Insurance and Tabulating Industries

Published online by Cambridge University Press:  13 December 2010

JoAnne Yates
Get access Rights & Permissions [Opens in a new window]

Abstract

Punched-card tabulating equipment, an important commercial predecessor of the computer, was used for processing large amounts of data in many business firms during die first half of the twentieth century. Life insurance was an information-intensive business dependent on firms' abilities to manage large quantities of data. This article examines both the role that tabulating machinery played in shaping insurance firms' business processes and the simultaneous role that Ufe insurance as a user industry played in shaping the development of tabulating technology between 1890 and 1950. The ongoing interaction between the Ufe insurance and tabulating industries shaped both in significant ways, setting the stage for continued interaction between the two industries during the transition to computers beginning at mid-century.

Type
Articles
Information
Business History Review , Volume 67 , Issue 1 , Spring 1993 , pp. 1 - 51
Copyright
Copyright © The President and Fellows of Harvard College 1993

Access options

Get access to the full version of this content by using one of the access options below. (Log in options will check for institutional or personal access. Content may require purchase if you do not have access.)

References

1 New York Tribune, 25 April 1890, as quoted by Austrian, Geoffrey D., Herman Hollerith: Forgotten Giant of Information Processing (New York, 1982), 83.Google Scholar

2 Although I will refer to both tabulating and the life insurance industry as industries, they are both clearly parts of larger industries. Cortada, James W., in Before the Computer: IBM, NCR, Burroughs, & Remington Rand & the Industry They Created, 1865–1956 (Princeton, N.J., 1993)Google Scholar, sets the tabulating business within the office machiner. industry, and that industry definition could be broadened further to include the office appliance and office supply businesses. Indeed, Cortada argues that all the forms of office and data-handling equipment evolved in interaction with the information needs of businesses in general, and I have made a similar argument with regard to earlier office techniques and technologies in Yates, JoAnne, Control through Communication: The Rise of System in American Management (Baltimore, Md., 1989).Google Scholar Nevertheless, here I am making a much narrower and more specific argument, and for these purposes the tabulating business itself was fairly well defined, with a very limited set of vendors in the United States and Britain. Thus it is useful to discuss it as an industry by itself. Similarly, although life insurance is clearly part of a larger insurance industry, it functioned quite separately during this period, for regulatory and other reasons; even its trade associations were usually solely for life insurance. Other types of insurance adopted and used tabulating technology as well but at a somewhat slower rate and somewhat less extensively, because they lacked the immense numbers of policies common in life insurance and had more variation in the policies themselves. Thus I have chosen to limit my user industry analysis to life insurance, and, unless otherwise stated, references to insurance should be taken to mean life insurance.

3 For a standard treatment of Hollerith equipment in computer history, see, for example, Goldstine, Herman H., The Computer from Pascal to von Neumann (Princeton, N.J., 1972).Google Scholar On the other hand, there is no treatment of tabulating equipment in, for example, Williams, Michael R., A History of Computing Technology (Englewood Cliffs, N.J., 1985).Google Scholar

4 Austrian, Herman Hollerith; Beniger, James R., The Control Revolution: Technological and Economic Origins of the Information Society (Cambridge, Mass., 1986)Google Scholar; Norberg, Arthur L., “High Technology Calculation in the Early 20th Century: Punched Card Machinery in Business and Government,” Technology and Culture 31 (1990): 753–79CrossRefGoogle Scholar; Campbell-Kelly, Martin, “Punched-Card Machinery,” in Computing before Computers, ed. Aspray, William (Ames, Iowa, 1990)Google Scholar; and Cortada, Before the Computer.

5 In one of the few studies recognizing the nonmechanical roots of information technology, Campbell-Kelly has studied the emergence of manual methods of large-scale data processing in a British insurance firm during the Victorian era. He notes that “the insurance business is perhaps the purest example of an ‘information-based’ industry—that is, an industry whose sole activity consists of gathering, processing, and distributing informa tion.” Campbell-Kelly, Martin, “Large-Scale Data Processing in the Prudential, 1850—1930,” Accounting, Business and Financial History 2, no. 2 (1992): 118.CrossRefGoogle Scholar

6 In the management of innovation literature, von Hippel, Eric, in The Sources of Innovation (New York, 1988)Google Scholar and other work, has highlighted the role of lead users in innovation. In economic history, Ross Thomson has studied the “learning by selling” process by which users influence ongoing technological innovation (“Learning by Selling and Invention: The Case of the Sewing Machine,” Journal of Economic History 47 [June 1987]: 433–45; The Path to Mechanized Shoe Production in the United States [Chapel Hill, N.C., 1989]); Christine MacLeod has examined the innovation and diffusion practices of machine makers and machine users, finding that users played a more important role in innovation and makers in diffusion (“Strategies for Innovation: The Diffusion of New Technology in Nineteenth Century British Industry,” Economic History Review 45 [May 1992]: 285–307); and Janet T, Knoedler has examined user-initiated joint consumer-producer efforts in the United States to innovate in steel products (“Market Structure, Industrial Research, and Consumers of Innovation: Forging Backward Linkages to Research in the Turn-of-the-Century U.S. Steel Industry,” in this issue of the Business History Review). Of course, studies of the social construction of technology have also shifted focus from the individual inventor to the broader social system influencing innovation, in which users may be considered one of the relevant social groups; see, for example, Bijker, Wiebe E., “The Social Construction of Bakelite: Toward a Theory of Invention,” in The Social Construction of Technological Systems, ed. Bijker, Wiebe E., Hughes, Thomas P., and Pinch, Trevor (Cambridge, Mass., 1987), 159–87.Google Scholar

7 As defined in Thomas P. Hughes, “The Evolution of Large Technological Systems,” in Bijker, , Hughes, , and Pinch, , eds., The Social Construction of Technological Systems, 51–82.Google Scholar

8 Chandler, Alfred D. Jr, Scale and Scope: The Dynamics of Industrial Capitalism (Cambridge, Mass., 1990).Google Scholar This article does not, for example, go into competitive developments in the European market, which generally had little direct effect on the American market; an exception is a brief foray into Great Britain to deal with a significant technical innovation that occurred in conjunction with a British life insurance firm.

9 I do not use co-evolution in the biological sense invoked by Constant, Edward W. (“On the Diversity and Co-evolution of Technological Multiples: Steam Turbines and Pelton Water Wheels,” Social Studies of Science 8 [May 1978]: 183– 210)CrossRefGoogle Scholar; rather, I use it to denote the contemporaneous and interacting developments of a technology and its use and of the vendor and user industries. In particular, many of the developments result from the conscious actions and reactions of managers, inventors, or other individuals or groups involved. Since writing this article, I have discovered a group of researchers in technology and innovation management who have recently introduced what seems to be a similar use of the term. See, for example, Lori Rosenkopf and Michael L. Tushman, “The Co-Evolution of Technology and Organization,” and Joel A. C. Baum and Jitendra V. Singh, “Organization-Environment Coevolution,” both in Evolutionary Dynamics of Organizations, ed. J. A. C. Baum and J. V. Singh (New York, forthcoming).

10 Zelizer, Viviana A. Rotman, Morals and Markets: The Development of Life Insurance in the United States (New York, 1979).Google Scholar The background information on the Ufe insurance industry in this paragraph and the next comes from this source and from Keller, Morton, The Life Insurance Enterprise, 1885–1910: A Study in the Limits of Corporate Power (Cambridge, Mass., 1963).CrossRefGoogle Scholar

11 The largest industrial insurance firms fit both characteristics of lead users defined in Hippel, von, The Sources of Innovation (p. 107)Google Scholar: 1) they face information-handling needs that will be experienced by all life insurance firms (and many other types of firms, as well), but they face them significantly earlier than other firms; and 2) they will benefit significantly by obtaining solutions to those needs. The large number of transactions required by industrial insurance brought on problems of data management much sooner than in ordinary insurance. And, though many of the firms were nonprofit mutuals, as discussed later in this section, they all wanted to continue growing. Reducing the information-handling costs of industrial insurance would both expand the total market for such insurance and give a particular company an opportunity to increase market share.

12 The Armstrong Commission Hearings of 1905, conducted by a joint committee of the New York legislature, were a watershed event in the history of insurance in America. In the wake of the investigation, which exposed widespread abuses in insurance finance, state regulators passed a series of statutes strictly limiting the investment activities of the firms and mandating stricter controls over products and operations. During this period, many firms also mutualized. See Keller, , The Life Insurance Enterprise, 265–92Google Scholar; North, Douglass C., “Life Insurance and Investment Banking at the Time of the Armstrong Investigation of 1905–1906,” Journal of Economic History 14 (Summer 1954): 209–28.CrossRefGoogle Scholar

13 Yates, Control through Communication; and “Evolving Information Use in Firms, 1850–1920: Ideology and Information Techniques and Technologies,” in Information Acumen: The Understanding and Use of Knowledge in Modern Business (forthcoming from Routledge Press). See also Beniger, The Control Revolution.

14 For example, James, Marquis, The Metropolitan Life: A Study in Business Growth (New York, 1947), 168.Google Scholar

15 Stalson, J. Owen, D.C.S., , Marketing Life Insurance: Its History in America (Cambridge, Mass., 1942), 800.Google Scholar

16 Campbell-Kelly (personal communication, January 1993) notes that British insurance firms did not handle all of these types of information. In Britain, a single annual valuation of policies in force was the major actuarial task, for example, with no need to value and compute reserve requirements for each state as in the United States. There was little focus on mortality studies, with most British firms continuing to use nineteenth-century mortality tables well into the twentieth century. Moreover, they did not maintain internal cost accounting systems to monitor costs and to aid in pricing products. Whereas the first of these differences reflects the state-level regulation of U.S. insurance industry, the others reflect the preoccupation with systematization and statistical analysis prevalent in the American business community around the turn of the century; see Yates, , Control through Communication, 1–20.Google Scholar

17 B. F. Dvorak, untitled address to Life Office Management Association (LOMA), Fort Wayne, Ind. (undated, but with materials from c. 1926), in Home Office Study Committee, cabinet 13, Metropolitan Life Insurance Company Archives, New York, N.Y.

18 Austrian, , Herman Hollerith, 238Google Scholar; Norberg, , “High Technology Calculation,” 764.Google Scholar

19 The following description of early Hollerith equipment is based on Campbell-Kelly, “Punched-Card Machinery”; Campbell-Kelly, , ICL: A Business and Technical History (Oxford, England, 1989)Google Scholar; Norberg, “High Technology Calculation”; Bashe, Charles J., et al., IBM's Early Computers (Cambridge, Mass., 1986)Google Scholar; Truesdell, Leon E., The Development of Punch Card Tabulation in the Bureau of the Census: 1890–1940 (Washington, D.C., 1965)Google Scholar; and Austrian, Herman Hollerith.

20 Hughes, Thomas P., Networks of Power: Electrification in Western Society, 1880–1930 (Baltimore, Md., 1983), 14.Google Scholar

21 Campbell-Kelly, “Large-Scale Data Processing in the Prudential.”

22 Prudential's number of policies calculated roughly from data in James, , The Metropolitan Life, 94–95.Google Scholar Prudential's announcement that it would try the equipment is from New York Tribune, 25 April 1890, as cited in Austrian, , Herman Hollerith, 83.Google Scholar

23 Moorhead, E. J., Our Yesterdays: The History of the Actuarial Profession in North America, 1809–1979 (Schaumberg, Ill., 1989), 338Google Scholar; Fackler, David Parks, “Regarding the Mortality Investigation, instituted by the Actuarial Society of America and now in progress,” Journal of the Institute of Actuaries 37 (1903): 1–15.Google Scholar Prudential's archives are closed to the public, so I can only speculate about Gore's motives.

24 Austrian, , Herman Hollerith, 82–83.Google Scholar

25 Campbell-Kelly, , “Punched-Card Machinery,” 133.Google Scholar

26 Hollerith corresponded with Louis F. Butler and H. J. Messenger of Travelers Insurance; Emory McClintock of Mutual Life; John Tatlock of the Actuarial Society of America; John B. Lunger, Rufus Weeks, and A. R. Grow of New York Life; and D. H. Wells of Connecticut Mutual Life in the period from December 1900 through May 1901; container 10, Hollerith Collection, Library of Congress, Washington, D.C.

27 Gore to Hollerith, 23 May 1901, container 10, Hollerith Collection.

28 Austrian, , Herman Hollerith, 177Google Scholar; Campbell-Kelly, ICL, and “Punched-Card Machinery.”

29 John Tatlock, Jr., secretary of the Actuarial Society of America [writing for McClintock], to Hollerith, 20 May 1901, and Gore to Hollerith, 23 May 1902, both in container 10, Hollerith Collection.

30 Campbell-Kelly, , “Punched-Card Machinery,” 140.Google Scholar

31 When the Prudential realized that the Gore sorter could not keep up with subsequent tabulating developments (Earl Chapin May and Will Oursler, The Prudential: A Story of Human Security [Garden City, N.Y., 1950], 308), it contracted with Royden Peirce for an alternative system similar to the one he was developing for Metropolitan Life (discussed in the next section). Although the contract stipulated a 1918 installation, the order was not fulfilled until 1925, and even then incompletely, with the machines requiring the full-time efforts of a Peirce mechanic to keep running; see “A Study of the Peirce Machines,” 5–6, Peirce Machine Matters, cabinet 2, Metropolitan Life Archives. Judging by the Metropolitan's experience with Peirce, catalogued in the same report, the installation was probably not very satisfactory and could not be adapted to changes in the types of policies or data gathered. The Prudential's own accounts of its tabulating experience omit mention of the Peirce machinery (Moorhead, , Our Yesterdays, 338)Google Scholar, reinforcing this supposition.

32 Austrian, , Herman Hollerith, 134.Google Scholar

33 Louis F. Butler to Hollerith, 13 March 1901, container 10, Hollerith Collection.

34 Hollerith to John B. Lunger of New York Life, 28 April 1900, container 10, Hollerith Collection. At this time, card files with tabs, notches, or holes were just coming into use as systems for storage and retrieval of structured data; Yates, , “Information Systems for Handling Manufacturing and Marketing Data in American Firms, 1880–1920,” Business and Economic History, 2d ser., 18 (1989): 207–17.Google Scholar Travelers and other firms were simply using the Hollerith punch to create such a system.

35 Hollerith to H. E. Davidson of the Library Bureau, 8 May 1901, container 10, Hollerith Collection.

36 Hollerith's correspondence with various members of the committee from December 1900 through May 1901 (see note 26) makes clear that lack of adequate attention on Hollerith's part affected the decision. Although the sorter must have been under development for the U.S. Census at this point, he did not mention it in this correspondence. Hollerith insisted on having as many committee members as possible view the new accumulating tabulators he had recently installed in the New York Central Railroad, but he was too busy to travel up to New York to demonstrate it himself. This equipment, it turned out, was less relevant to the needs of the mortality study than that used by the census. Misunderstandings also arose. For example, some committee members questioned whether Hollerith's equipment could count cards with combinations of holes, not single holes, simultaneously. In his correspondence Hollerith stated that it would, but the issue was alluded to later as a decision factor in favor of the Gore equipment (Hollerith to D. H. Wells of Connecticut Mutual Life, 31 Dec. 1900; John Tatlock of the American Soci ety of Actuaries to Hollerith, 21 May 1901— both in container 10, Hollerith Collection). According to Austrian, , Herman Hollerith, 180Google Scholar, the sorters that Hollerith introduced later that year did not sort on combinations, but the tabulators allowed tabulation of combinations. This fact was not, apparently, clear to the committee.

37 Hunter, Arthur, “Method of Making Mortality Investigations by Means of Perforated Cards, Sorting and Tabulating Machines with Special Reference to the Medico-Actuarial Mortality Investigation,” Transactions of the Actuarial Society of America 11 (1909–1910): 252–75.'Google Scholar

38 Hunter, Arthur, “Note on an Approximate Method of Making Mortality Investigations,” Transactions of the Actuarial Society of America 10 (1907–1908): 361Google Scholar; Hunter, “Method of Making Mortality Investigations.” See also New York Life Insurance Co., A Temple of Humanity (New York, 1909).Google Scholar New York Life's rank (by value of insurance in force) is from Stalson, , Marketing Life Insurance, 800.Google Scholar

39 Hunter, , “Method of Making Mortality Investigations,” first quotation, 268–69Google Scholar; second, 265.

40 Ibid., 253. The next passage quoted is from 252–53.

41 For example, Metropolitan Life's Ordinary Insurance Section; Metropolitan Life Insurance Company, The Metropolitan Life Insurance Company: Its History, Its Present Position in the Insurance World. Its Home Office Building and Its Work Carried on Therein (New York, 1914), 70.Google Scholar

42 Kaufman, Henry N., “Some Uses for the Hollerith Machines,” Transactions of the Actuarial Society of America 11 (1909–1910): 276–95Google Scholar; quotation is from 291. Kaufman's company is not identified in this article, but a 9 Nov. 1911 letter from Kaufman to Herman Hollerith on the latter's retirement (container 10, Hollerith Collection) reveals his title and affiliation a year later, when they were probably the same. Phoenix was ranked nineteenth out of 214 firms in 1910 (Stalson, , Marketing Life Insurance, 800Google Scholar, 821).

43 Kaufman, , “Some Uses for the Hollerith Machines,” 278.Google Scholar It is generally assumed that machines could always be rewired, but the detailed description provided in Kaufman's article on accumulator configuration and attachments makes clear that at this point fields defined for accumulating (as opposed to counting) could not be changed by simple rewir ing.

44 C. L. Hayes to Mr. Braitmayer, 25 Nov. 1912; Braitmayer to Hayes, 3 Dec. 1912, container 10, Hollerith Collection.

45 Kaufman, , “Some Uses for the Hollerith Machines,” 279.Google Scholar

46 Metropolitan Life, The Metropolitan Life Insurance Company, 70.Google Scholar

47 Chandler, , Scale and Scope, 8Google Scholar, 35.

48 Austrian, , Herman Hollerith, 238–56.Google Scholar

49 Bashe, , et al., IBM's Early Computers, 9–10Google Scholar; Campbell-Kelly, , ICL, 69.Google Scholar

50 “Report on the Royden System of Perforated Cards,” 9 March 1912, addressed to tlie President's Commission on Economy and Efficiency, signed by M. O. Chance, F. H. Tonsmeire, and E. H. Maling, located in container 3, Hollerith Collection; Prospectus for a new company, the Royden Company, which intended to take over these patents, with 24 Sept. 1912 cover letter from Arthur C. Sherwood to James R. Morse, located in container 10, Hollerith Collection. This attempt probably failed, since I have found no further references to a company of that name, but Peirce Patent Company continued to exist and to manufacture machines.

51 W. E. Freeman, “Automatic Mechanical Punching, Counting, Sorting, Tabulating and Printing Machines Adaptable to Various Lines of Accounting and Statistical Work Essential for Public Service Corporations with Particular Reference to Improvements in the Art of Mechanical Accounting,” paper presented at the annual convention of the National Electric Light Association, San Francisco, Calif, 7–11 June 1915.

52 Papps, Percy C. H., “The Installation of a Perforated Card System with a Description of the Peirce Machines,” Transactions of the Actuarial Society of America 15 (1914): 49–61Google Scholar; for Mutual Benefit's earlier interest in tabulating, see Austrian, , Herman Hollerith, 324.Google Scholar Mutual Benefit ranked eighth in value of insurance in force in 1910 and 1915 (Stalson, , Marketing Life Insurance, 800).Google Scholar

53 The company's public account of when Peirce was retained was contradictory (Dublin, Louis I., A Family of Thirty Million: The Story of the Metropolitan Life Insurance Company [New York, 1943], 253Google Scholar and 397), suggesting as early as 1907 and as late as 1913, but the absence of anyone from the Metropolitan on the 1912 prospectus's list of individuals who had approved the Peirce system suggests that the latter date is more likely.

54 June 1914 letter quoted in Austrian, , Herman Hollerith, 332–33.Google Scholar

55 “Tabulating Machine Co. Operating Revenue and Operating and Selling Expense, 1909 to 1913,” container 10, Hollerith Collection.

56 Papps, “The Installation of a Perforated Card System.”

57 Metropolitan Life, The Metropolitan Life Insurance Company, 71–72.Google Scholar

58 Series of contractual letters between J. Royden Peirce and J. M. Craig, actuary for Metropolitan Life, dated 29 May through 23 June 1913, Peirce Machine Matters, cabinet 2 Metropolitan Life Archives; Commentary by Craig, J. D. on Papps, Percy C. H., “The Installation of a Perforated Card System with a Description of the Peirce Machines,” Transactions of the Actuarial Society of America 15 (1914): 409–13.Google Scholar

59 Undated, unsigned internal report (probably from Mr. Washington) to the third vice-president (Mr. Henry Bruere), in Peirce Machine Matters, cabinet 2, Metropolitan Life Archives. Context and other references make 1916 the likely date. In it, he notes that “Mr. Pierce [sic] is now nearly finished with the work he is doing for the Actuarial Division.” He also notes the final cost of the key machine.

60 Vice-President Hegeman to J. Burn, actuary of Prudential Assurance Company, Ltd., 12 Feb. 1918, Peirce Machine Matters, cabinet 2, Metropolitan Life Archives.

61 “A Study of the Peirce Machines,” 5–6, Peirce Machine Matters, ibid.

2 Copies of the report to the President's Commission on Economy and Efficiency and the prospectus for the Royden Company were in Hollerith's papers and stamped “Personal/H.H.,” indicating that Hollerith was following Peirce's progress. The letter from Gershom Smith to Thomas J. Watson attests to the concern of Hollerith's successors. Both in container 10, Hollerith Collection.

63 Austrian, , Herman Hollerith, 272–74.Google Scholar

64 A certain amount of patent litigation took place between the Hollerith and Powers organizations in the United States and in Europe, but ultimately a cross-licensing arrangement was reached for use of the basic tabulating patents, removing them from the competitive arena. Some later patents that were relevant to the competition will be noted as necessary. See Connolly, James, A History of Computing in Europe (New York, 1967), 13Google Scholar; Campbell-Kelly, , ICL, 35Google Scholar, 64, 88–90.

65 Norberg, , “High Technology Calculation,” 765.Google Scholar

66 For example, see samples in Home Office Study Committee, cabinet 13, Metropolitan Life Archives.

67 “Powers Machinery: Origin and Development,” The Prudential Bulletin, Dec. 1934, 2664–65.

68 [Henry] Bruere, third vice-president, to James D. Craig, actuary, 20 Nov. 1924; Craig to Bruere, 22 Nov. 1924; Bruere to Craig, 25 Nov. 1924—all in Home Office Study Committee, cabinet 13, Metropolitan Life Archives.

69 Campbell-Kelly, , “Large-Scale Data Processing in the Prudential,” 128–30.Google Scholar

70 J. Burns to Mr. Hegeman, vice-president, Metropolitan Life, 7 Nov. 1917, Peirce Machine Matters, cabinet 2, Metropolitan Life Archives.

71 Campbell-Kelly, , ICL, 44–45.Google Scholar

2 Freeman, “Automatic Mechanical Punching, Counting, Sorting, Tabulating and Printing Machines.”

73 Campbell-Kelly, , ICL, 35–36.Google Scholar

4 Bashe, et al., IBM's Early Computers, 7–8.Google Scholar

75 An internal comparison of Powers, Hollerith, and Peirce equipment at the Metropolitan in the mid-1920s revealed a clear preference for the verifier over other methods of verifying accuracy, but other issues, including printing and alphabetical printing capability, were more important. Report to Mr. Dobbins, 2 Feb. 1926, file copy unsigned, in Home Office Study Committee, cabinet 13, Metropolitan Life Archives.

76 Bashe, et al., IBM's Early Computers, 9Google Scholar, dates the introduction to 1921, whereas Campbell-Kelly, , “Punched-Card Machinery,” 141Google Scholar, says it was introduced in 1920. Campbell-Kelly speculates that the long delay between the introduction of the Powers printing tabulator in 1915 and the first IBM printing tabulator, the Type I with automatic control, in 1920 reflected not just the First World War and the time required to develop printing and automatic control devices, but also IBM's intentional holding back from the market to delay making its nonprinting machines obsolete (Campbell-Kelly, , ICL, 63–64).Google Scholar Jordan, George, in “A Survey of Punched Card Development” (MA thesis, Massachusetts Institute of Technology, 1956)Google Scholar, argues that C-T-R continued to produce its standard machinery to fill high war-time demand, then faced a postwar crisis of surplus machinery on its hands. The development of its own printing tabulator responded to this crisis, again expanding demand.

77 Campbell-Kelly, , “Punched-Card Machinery,” 142.Google Scholar

78 Undated document from A. C. Carpenter in material from 1923–26, Home Office Study Committee, cabinet 13, Metropolitan Life Archives. Although Metropolitan Life continued to work with Peirce, as the next section will show, it also rented some equipment from the two commercial vendors.

79 Quoted in 22 Nov. 1924 memo from James D. Craig to Mr. Bruere, Home Office Study Committee, cabinet 13, Metropolitan Life Archives.

80 Dvorak, untitled address to LOMA; an earlier document refers to Dvorak as a Powers representative, but his status at the time of the paper is unclear.

81 A 21 Dec. 1925 report on the Addressograph from Mr. Washington to Mr. Dobbins, fourth vice-president of Metropolitan Life, discusses the use of these systems by other insurance firms (Home Office Study Committee, cabinet 13, Metropolitan Life Archives). He notes that “nearly all of the smaller companies and most of the larger companies use the Addressograph,” and he discusses its use by five specific companies, all of which adopted it between 1913 and 1920. See also, for example, Equitable Life Insurance Society, The Home Office of the Equitable: Description of Departments (New York, 1916), 124–25.Google Scholar For a brief history and description of addressing equipment, see Leffingwell, William Henry, ed., Office Appliance Manual (National Association of Office Appliance Manufacturers, 1926), 409–50.Google Scholar

82 Drawings marked “Confidential” and “Peirce Patents Company,” 20 May 1913, Peirce Machine Matters, cabinet 2, Metropolitan Life Archives.

83 J. Royden Peirce to James M. Craig, actuary, 18 Dec. 1916, Metropolitan Life, Peirce Machine Matters.

84 Ibid.

85 The history of Peirce's other contracts is traced in a 1 March 1926 internal Metropolitan Life report, “A Study of the Peirce Machines,” Peirce Machine Study, cabinet 2, Metropolitan Life Archives. See also Peirce to Metropolitan Life, 21 Jan. and 5 Feb. 1919, with attached notes from Metropolitan executives, Peirce Machine Matters, ibid.

86 Unsigned memorandum addressed to Home Office Study Committee, 7 Nov. 1924, in Home Office Organization Study, cabinet 13, Metropolitan Life Archives.

87 J. D. Craig to Mr. Henry Bruere, third vice-president and chairman of the Home Office Study Committee, 22 Dec. 1924, in Home Office Organization Study, ibid.

88 “A Study of the Peirce Machines,” 1 March 1926, 58–59, Peirce Machine Study, cabinet 2, Metropolitan Life Archives.

89 Campbell-Kelly, , ICL, 43–45.Google Scholar

90 “Powers Machinery: I. Origin and Development” (first of a six-part series), The Prudential Bulletin, Dec. 1934, 2665.

91 Campbell-Kelly, , ICL, 48.Google Scholar

92 Ibid., 45.

93 “Report on Perforated Card Systems of English Companies, January 1924” (dated 13 Dec. 1923 at the end of the report itself; presumably it was written in December, possibly in England, but typed and delivered only on the contingent's return), cabinet 2, Metropolitan Life Archives.

94 Campbell-Kelly, , ICL, 45Google Scholar, 51–52.

95 “Report on Perforated Card Systems of English Companies.” It is unclear whether Powers AMC ever marketed this thirteen-charaeter version.

96 “Report on Perforated Card Systems of English Companies,” 2. Actually, even “full” alphabetic tabulators generally printed only twenty-four letters, using the same character for U and V and the same character for S and 5.

97 Home Office Organization Study, Jan. 1924, cabinet 13, Metropolitan Life Archives. See especially entries for 13 June-12 July, pp. 52–56. This notebook, assembled by the office of Henry Bruere, third vice-president and chairman of the committee, includes logs of meetings and phone calls as well as copies of memos and reports exchanged. The study was initiated to address a space crisis in the home office. Mechanization of clerical processes was seen as one way to save space.

98 From document dated 2 April 1925, Metropolitan Life Archives. The date probably refers to the day on which the sample sheet was created, not the day on which this machine was first successfully run. Standard treatments of the history of tabulating equipment date introduction of alphabetical tabulating by Powers AMC to 1924 (Campbell-Kelly, , “Punched-Card Machinery,” 142Google Scholar; Norberg, , “High Technology Calculation,” 768)Google Scholar, and the log of the Home Office Study Committee places the first successful demonstration of the alphabetical attachment on 19 Nov. 1924; see pp. 66–71.

99 Entries for 16 June-12 July 1924, 55–56, Home Office Organization Study, Jan. 1924, cabinet 13, Metropolitan Life Archives. Later entries in the committee logs indicate that patent issues became more problematic later, as IBM informed Metropolitan Life that the Peirce patents it had purchased in 1922 invalidated Powers AMC's patents (9 July, p. 56; 1–8 Sept., p. 61; 1–31 Dec, p. 72). The meeting between Mr. Bruere and Mr. Pritchard of Powers AMC is described in the entry dated 1–31 Dec, p. 72.

100 Bashe, et al., IBM's Early Computers, 9–10.Google Scholar For the next several years, Peirce continued to work independently with the Metropolitan, overseeing work in the separate machine shop they had set up for him, at the same time that he worked for IBM on developing alphabetical tabulating. Eventually, IBM took over Peirce's Metropolitan contract, as well as Peirce's contracts with the American Prudential and with the Veterans Bureau.

101 Bashe, et al., IBM's Early Computers, 10.Google Scholar

102 Campbell-Kelly, , “Punched-Card Machinery,” 142.Google Scholar

103 Connolly, , A History of Computing in Europe, 25.Google Scholar

104 Campbell-Kelly, , ICL, 82.Google Scholar

105 Log book entry for 1–31 Dec. 1924, p. 72, Home Office Study Committee, Jan. 1924, cabinet 13, Metropolitan Life Archives.

106 Norberg, , “High Technology Calculation,” 771.Google Scholar This momentum was more a business than a technological momentum, in the sense of Hughes, Thomas P. (Networks of Power, 15Google Scholar, 140–74). Both technological approaches to punched-card tabulating, electrical and mechanical, remained available, but IBM dominated the market.

107 Engelbourg, Saul, International Business Machines: A Business History (New York, 1976), 316–17.Google Scholar

108 Campbell-Kelly, , ICL, 65Google Scholar; Jordan, , “A Survey of Punched Card Development,” 31–32Google Scholar, 52; “Parade of the Business Machines,” Business Week, 19 Oct. 1935, 11 (quotation is from this source); “Ways and Means” [an in-house newsletter issued by Metropolitan Life's Standardization Bureau] 4 (August 1948): 1.

109 “Renewal Premium Notices and Receipts and Premium Billing Routines,” pre pared by LOMA Staff Office, Report No. 2 of the Office Machinery and Equipment Committee of the Life Office Management Association, 1 Sept. 1938.

110 For example, LOMA Bulletin 17 (1951)Google Scholar includes three items about conversions of billing operations to punched cards: 15 May, p. 25; 15 July, p. 40; 15 Nov., pp. 63–64.

111 I have not been able to locate figures reflecting what percentage of the tabulating market the life insurance industry accounted for, but it certainly comprised some of the largest single users of the equipment, including Metropolitan Life. As a possible index to its importance, we can look at the first firms to computerize, all of which were major users of tabulating equipment. After government agencies and universities, insurance was the second commercial customer to have a Univac (the first commercially available large digital computer) installed; the life insurance industry accounted for one-quarter of all acquirers of Univaes in 1954, the first year of commercial deliveries, and 1955 (Yates, “From Tabulators to Early Computers in the U.S. Life Insurance Industry: Co-evolution and Continuities,” Sloan School of Management working paper no. 3618–93, Cambridge, Mass., Oct. 1993). Although this figure clearly overstates life insurance's share of the total tabulating market, which included many firms and organizations with much smaller installations that would not have jumped into the computer market right away, it does indicate how significant insurance was among large users of data-processing equipment.

112 Campbell-Kelly, , ICL, 54–55.Google Scholar

113 Ibid., 129–30.

114 MacLeod, , “Strategies for Innovation,” 301.Google Scholar

115 Campbell-Kelly, , ICL, 63–64.Google Scholar