This paper was researched and written while Richard Sutch was Visiting Professor of Economics, California Institute of Technology. Data collected for this paper has been deposited in an archive at the Laboratory for Historical Research at the University of California, Riverside. Communications and requests for data should be addressed to the authors in care of the University California History of Saving Project, Institute of Business and Economic Research, University of California, Berkeley, CA 94720.Google Scholar

Lance Davis, John James, Martha Olney, B. Michael Pritchett, Richard Sylla, and Samuel Williamson offered helpful suggestions. Financial support was provided by the National Science Foundation, the John Simon Guggenheim Foundation, the Division of Humanities and Social Sciences of the California Institute of Technology, and the Laboratory for Historical Research of the University of California, Riverside.Google Scholar

¹ Homans, Sheppard and Phillips, George W., “Tontine Dividend Life Assurance Policies,” Equitable Life Assurance Society of the United States, pamphlet dated December 22, 1868,Google Scholar as quoted in Buley, R. Carlyle, The Equitable Life Assurance Society of the United States: 1859–1964, 2 vols. (New York, 1967), vol. 1, p. 93.Google Scholar Sheppard Homans was the Actuary of the Mutual Life Insurance Company from 1855 to 1871. Despite the fact that he was in Mutual's employ, he consulted for other firms including, in this case, Equitable. George Phillips was Equitable's Actuary.

² Clough, Shepard B., A Century of American Life Insurance: A History of the Mutual Life Insurance Company of New York, 1843–1943 (New York, 1946), Pp. 59–75;Google ScholarBuley, Equitable, vol. 1, pp. 60–63, 92–93, 98–99.Google Scholar

³ Buley, R. Carlyle, The American Life Convention, 1906–1952: A Study in the History of Life Insurance, 2 vols. (New York, 1946), vol. 1, pp. 86–192;Google ScholarNorth, Douglass C., “Capital Accumulation in Life Insurance between the Civil War and the Investigation of 1905,” in Miller, William, ed., Men in Business: Essays on the Historical Role of the Entrepreneur (Cambridge, 1952);Google ScholarPope, Daniel, The Making of Modern Advertizing (New York, 1983), pp. 57–60, 221–23.Google Scholar The total assets held by life insurance companies grew at an annual rate of over 7.2 percent from $203 million in 1868 to $2,706 million in 1905. The 1868 figure comes from Pritchett, B. Michael, Financing Growth: A Financial History of American Life Insurance Through 1900, Huebner, S.S. Foundation Monograph, no. 13 (Homewood, 1985), table 10, p. 22.Google Scholar The 1905 figure is based on the compilation by the Spectator Company, The Insurance Year Book for 1906: Life, Casualty and Miscellaneous, 34th annual vol. (1906), Exhibit III, p. 499. Although the title varies somewhat, these volumes are hereafter cited as Spectator Year Book.Google Scholar

⁴ Hughes defeated William Randolph Hearst for Governor of New York in 1906. He resigned during his second term to accept an appointment to the United States Supreme Court. He resigned from the court in 1916 to run for President against Woodrow Wilson. He served as wen Harding's Secretary of State and was appointed Chief Justice of the Supreme Court by Herbert Hoover.Google Scholar

⁵ New York State Legislature, Testimony, Exhibits, Report and Index of the Joint Committee of the Senate and Assembly of the State of New York to Investigate and Examine into the Business and Affairs of Life Insurance Companies Doing Business in the Slate of New York,7 vols. (Albany, 1906), vol. 7, Pp. 322–24 [hereafter Armstrong Committee].Google Scholar

⁶ See, for examples, Stalson, J. Owen, Marketing Life Insurance: Its History in America (Cambridge, 1942), p. 405;Google Scholar and Krooss, Herman E. and Blyn, Martin R., A History of Financial Intermediaries (New York, 1971), P. 110.Google Scholar

⁷ North, Douglass, “Entrepreneurial Policy and Internal Organization in the Large Life Insurance Companies at the Time of the Armstrong Investigation of Life Insurance,” Explorations in Entrepreneurial History, 5 (03 1953), pp. 139–61.Google Scholar

⁸ The distinguishing feature of a tontine is that only survivors share in the distribution. The name “tontine” recalled the scheme adopted in 1689 by Cardinal Mazarin to raise money for Louis XIV of France. Mazarin's plan had been suggested by Lorenzo Tonti a Neapolitan physician and banker living in Paris. Shares were sold at 300 livres each. The principal earned interest but was never repaid. The survivors inherited the interest of those who died, thus as time went on the whole annuity fund was divided among fewer and fewer survivors, the last of whom drew an annual income of 73,500 livres.Google Scholar

⁹ Buley, Equitable, vol. 1, pp. 94, 126–28; and Buley, American Life Convention, vol. 1, pp. 92–96.Google ScholarHudnut, James M., Semi-Centennial History of the New York Life Insurance Company, 1845–1895 (New York, 1895), pp. 150–51, 155–59.Google Scholar Equitable's share of the industry's sales rose from 7.6 percent to 11.0 percent between 1870/71 and 1872/73, meanwhile, New York Life saw its share of business increase from 4.8 to 5.6 percent; Spectator Year Book, 20 (1892), pp. 274–75, 298–99;Google Scholar and Hoffman, Frederick L., “Fifty Years of American Life Insurance Progress,” Journal of the American Statistical Association, new series, 95 (09 1911), pp. 667–760, table XV, p. 727.CrossRefGoogle Scholar

¹⁰ At least 55 companies out of an initial 128 disappeared between 1872 and 1878; Pritchett, Financing Growth, table 1, p. 6. As far as we have been able to determine, none of the companies that failed had issued tontine policies. By contrast, the Equitable and New York Life passed through the crisis period unscathed;Google ScholarAlexander, William, My Half-Century in Life Insurance (New York, 1935), p. 46;Google Scholar and Hudnut, New York Life, chap. 9.Google Scholar

¹¹ Williamson, Harold and Smalley, Orange, Northwestern Mutual Life: A Century of Trusteeship (Evanston, 1957), pp. 101–2.Google Scholar

¹² Clough, Mutual Life, pp. 141, 144–147; Armstrong Committee, Testimony, vol. 2, p. 1724.Google Scholar

¹³ The New York Insurance Commissioner reported the volume of tontine insurance in force for 26 companies at the end of 1907; New York, Insurance Commission, Forty-Ninth Annual Report of the Superintendent of Insurance of the State of New York, Part II, “Life Insurance” (Albany, 1908).Google Scholar

¹⁴ The data on total terminations were taken from the Spectator Year Book, 36 (1908), Exhibit XXXI, pp. 790–95. The relationship was estimated by linear regression techniques as explained in the source notes to Table 1.Google Scholar

¹⁵ The smaller companies omitted from Table I were heavily involved in selling deferred- dividend policies; Ransom, Roger L. and Sutch, Richard, “Swindle or Security? A Reexamination of Tontine Insurance, 1871–1905,” Working Papers on the History of Saving, No. 6 (Berkeley, November 1986). Table 1 presents data only for non-industrial insurance. Industrial insurance, which is excluded, was at that time about 17 percent of all life insurance;Google ScholarHoffman, “Fifty Years,” tables V, VI, and VII, pp. 717–19. In 1905 industrial insurance was issued by 20 different companies, but three (Metropolitan, Prudential, and John Hancock) accounted for 95 percent of this business. All three issued industrial tontine insurance; Best Reporting Company, Best's Insurance Reports [Life] Upon All Legal Reserve Companies, Assessment Associations and Fraternal Societies Transacting Business in the United States, 1st annual vol. (1906–1907), pp. 252–57, 165–70, 132–33. Of these three, we have been able to ascertain the fraction of the industrial insurance which was issued as tontine policies only for Prudential. For that company, according to the 1908 Report of the New York State Insurance Commission, 63.2 percent of its industrial insurance was on a deferred- dividend basis at the end of 1907, p. 773.Google Scholar

¹⁶ Hoffman, “Fifty Years,” table V, p. 717;Google Scholar and Goldsmith, Raymond, A Study of Saving in the United States, 3 vols. (Princeton, 1956), vol. 3, table W-2, p. 17.Google Scholar

¹⁷ U.S. Bureau of the Census, Historical Statistics of the United Stares, Colonial Times to 1970, 2 vols. (Washington, D.C., 1975), vol. 1, Series A288, p. 41.Google Scholar Our estimate of the number of tontine policies implies that the average size of a tontine policy in 1905 was about $1,000. That may be high since the average value of a policy calculated from data on all insurance was about $600; Hoffman, “Fifty Years,” table V, p. 717.Google Scholar

¹⁸ The Armstrong evidence includes information on the performance of many tontine funds issued by many companies. Examination of this data has convinced us that the relative performance of the Equitable tontine of 1871 is representative.Google Scholar

¹⁹ Elsewhere we present evidence that suggests that the forecasted gains from a tontine policy made by Equitable in 1871 were reasonable projections and not deliberate exaggerations; Ransom and Sutch, “Swindle or Security?”Google Scholar

²⁰ Ransom and Sutch, “Swindle or Security?”Google Scholar

²¹ On the relatively high incidence of retirement during this period see Ransom, Roger L. and Sutch, Richard, “The Labor of Older Americans: Retirement of Men On and Off the Job, 1870–1937,” this Journal, 46 (03 1986), pp. 1–30.Google Scholar

²² The economic theory of saving for old age is known as the “life cycle hypothesis” and is associated with the work of Franco Modigliani; see Abel, Andrew, ed., The Collected Papers of Franco Modigliani, 3 vols. (Cambridge, 1980), vol. 2, “The Life Cycle Hypothesis of Saving”Google Scholar and Modigliani, Franco, “Life Cycle, Individual Thrift, and the Wealth of Nations,” American Economic Review, 76 (06 1986), pp. 297–313.Google Scholar

²³ The efficiency advantages of a straddle may help explain why pure tontine funds—without the life insurance—were not introduced during this period.Google Scholar

²⁴ American insurance companies are subject to state rather than federal regulation. A decision of the U.S. Supreme Court in Paul v. Virginia in 1868 declared that insurance was not an act of interstate commerce; Buley, Equitable, vol. I, p. 92. A “race to the bottom” between states in the matter of insurance regulation left accounting methods and control largely up to the companies themselves;Google ScholarNorth, Douglass, “The Large Life Insurance Companies Before 1906” (Ph.D. diss., University of California, Berkeley, 1952), chap. 8.Google Scholar

²⁵ Buley, Equitable, vol. 1, p. 172; vol. 2, p. 736.Google ScholarCarter, Ellerbe W., “Suits for Accounting on Tontine Life Insurance Policies,” Virginia Law Review, 2 (1914), pp. 18–32.CrossRefGoogle Scholar

²⁶ Lawson's, Thomas articles were later reprinted as Frenzied Finance: The Crime of Amalgamated (Boston, 1905).Google Scholar See Lawson, pp. 414–15; North, “Large Life Insurance Companies,” pp. 35–36;Google ScholarCarosso, Vincent P., Investment Banking in America: A History (Cambridge, 1970), pp. 112–15;Google Scholar and Buley, American Life Convention, vol. 1, 199–201.Google Scholar

²⁷ Armstrong Committee, Report, vol. 7, pp. 322–24. We have rearranged the sentences somewhat and supplied the numbering.Google Scholar