X 879-917. General note.

There are three general sources of primary data about life insurance as a whole: The various State insurance departments through their reports of the life insurance companies operating within their jurisdictions; commercial publishers of life insurance company data; and the trade and other associations of the life insurance companies. Probably the most widely used of the State insurance department reports are those published annually by the New York Insurance Department. For the approximate period 1860-1880 these reports, which give data on the companies domiciled in the State and the companies of other States authorized to transact business in the State, are most frequently made use of to exhibit the progress of life insurance. The data presented in these reports for this period represent a very high percentage of the total life insurance business. Other State reports often consulted by researchers are those of Massachusetts and Connecticut.

Of the commercial publications, the most frequently used to study the progress of life insurance as a whole is the Spectator Insurance Year Book, published annually since 1873 by the Spectator Company, Philadelphia. (For 1873-1937, the publication was known as The Insurance Year Book. There were separate "Life" volumes from 1923-1963 and "Property, casualty" volumes through 1960.)

A number of the trade and other associations in the life insurance business prepare industrywide statistics on different aspects of life insurance. The Institute of Life Insurance, New York, a public relations organization formed by the life insurance companies, compiles a number of such statistics and publishes these, as well as data from other associations and from commercial publishers, annually in the Life Insurance Fact Book. Two major sources of insurance statistics included in the Institute compilations are the American Life Insurance Association and the Life Insurance Agency Management Association.

To obtain a series of figures over a long period it is not necessary to consult each annual edition of the publications mentioned above. Many of the New York Insurance Reports (known also as the Annual Report of the Superintendent of Insurance) contain a chronology which gives some of the salient statistics over a long period (sometimes only for selected years). Prior to 1963, the annual Spectator Insurance Year Book often gave the aggregates for all available companies for the preceding 10 years and, for the early years of this publication, a summary of data was presented for the companies operating in New York State. Thereafter, the monthly Spectator Magazine presents data on life, property, accident, and health companies.

The most recent Life Insurance Fact Book will generally give most of the preceding statistics compiled, as well as historical statistics from other sources, as far back as 1890 for some series.

In addition to the Life Insurance Fact Book, there are two compilations of historical statistics which are often consulted: J. Owens Stalson, Marketing Life Insurance, Its History in America, Harvard University Press, Cambridge, 1942 (the appendixes give data on the number of companies, life insurance sales, life insurance in force, and income as well as many other items from earliest available figures to 1937); and Frederick L. Hoffman, "Fifty Years of American Life Insurance Progress," Quarterly Publications of the American Statistical Association, New Series, No. 95, vol. XII, Boston, 1911 (tables of salient statistics, 1860-1910). The statistics presented in these publications do not always agree with the figures given here because in some cases different sources have been used, and in some cases adjustments and corrections of the source material have been made by the Institute of Life Insurance.

Because it represents the exception rather than the rule, it may be of interest to note two instances in which data on life insurance were collected in the decennial census of the United States. Statistics of the United States in 1860, 1866, pp. 293–294, contains some statistics on the number of life insurance companies, the amount of life insurance, the number of persons insured, and the annual premium income for 1860. Data on the life insurance business are also shown in the Report on Insurance Business in the United States at the Eleventh Census: 1890, Part 2, Life Insurance, 1895. This report contains statistics on life insurance for the decade 1880-1890 for the companies in operation as of December 31, 1889. It does not, however, reflect the business in this decade of companies which ceased to do business before December 31, 1889.

The basic reporting form utilized by all three types of primary sources in preparing their statistics is the annual statement convention blank. This is the prescribed accounting statement which each company must submit to the insurance department of each State in which it is licensed to transact business, setting forth the company's balance sheet, income and disbursement accounts, policy exhibit, and many supporting schedules. The collecting agencies supplement the data from the annual statement form from time to time through mail questionnaires, mostly among the life insurance companies.

An understanding of the historical statistics of life insurance requires some knowledge of the annual statement convention blankthe accounting methods used in preparing the form and changes in the form and methods over the years and some knowledge of the history of life insurance.

Uniformity in the annual statement convention blank required by the States has been achieved through the efforts of the National Association of Insurance Commissioners. This association is a national organization composed of the officials of the various States who have supervision of insurance affairs within their respective States. It was formed in 1871 (under the name of the National Convention of Insurance Commissioners) and adopted its first convention blank in 1874. This organization has also achieved a degree of uniformity in insurance legislation and departmental rulings among the different States.

The convention blank has undergone revisions from time to time. The most recent significant revision in the annual statement convention blank took place in the form used for reporting the operations for 1951. Where these changes have affected the statistics shown, they are discussed below in the text for the specific series. For a complete discussion of the annual statement form now in use, and a comparison with the superseded form, see E. C. Wightman, Life Insurance Statements and Accounts, Life Office Management Association, New York, 1952 and J. C. Noback, Life Insurance Accounting, Irwin, Homewood, Ill., 1969. For a detailed discussion of two of the earlier forms, see Life Insurance Accounts, 1935 and 1941, by Wight

man.

There are many nonstatistical histories of life insurance. A few that may be consulted are: Charles K. Knight, The History of Life Insurance in the United States to 1870, unpublished thesis, University of Pennsylvania, 1920; Marketing Life Insurance, Its History in America (cited above); and The Bible of Life Insurance, George W. Wadsworth, 1932.

The data presented here cover only life insurance as it relates to the insurance companies which are usually referred to as the legal reserve life insurance companies. These are life insurance companies operating under insurance laws specifying the minimum basis for the reserves a company must maintain on its policies. Other types of

life insurance include fraternal life insurance which is provided by societies, lodges, and similar fellowship organizations; life insurance with assessment associations, mutual aid groups, and burial societies; life insurance available through savings banks in three States; and veterans life insurance (consisting of U.S. Government Life Insurance and National Service Life Insurance) issued by the Federal Government to members of the Armed Forces and veterans of World Wars I and II.

Though in very recent years the greatest part of all life insurance in force in the United States has been provided by the legal reserve life insurance companies, veterans insurance at its peaks during or immediately after the World Wars exceeded or nearly equaled the totals achieved by the life insurance companies. Fraternal and assessment life insurance combined for the period 1879-1928 was a significant proportion of the life insurance company total (actually exceeding it for a year or two in the 1890's and never amounting to less than 10 percent of the life insurance company total for the period stated).

For historical statistics of veterans, fraternal, and assessment life insurance, see Stalson, Marketing Life Insurance, cited above, pp. 806-808 and 816-819.

The data for legal reserve life insurance companies which are presented here are subject to three types of limitations: (a) Changes in the annual statement convention blank on which the companies report their operations; (b) incompleteness of the data in terms of the number of companies for which information is available; and (c) lack of uniformity among the companies in the allocation of certain items to the categories of the convention blank, changes in allocation, and changes made by the publishers of life insurance data in their reporting methods.

Changes in the annual statement blank over the years have been discussed previously. With regard to the completeness of the statistics available, it is extremely difficult to obtain data for any given period on the operations of all the life insurance companies operating in the United States. Theoretically, one should be able to compile complete statistics by consulting the insurance reports of each State and the District of Columbia, but in practice this is not feasible. State insurance reports began in the 1850's, but it was not until 1919 that all States (and the District of Columbia) were issuing reports. (A list of the first reports on insurance companies by State departments of insurance is given in Stalson, Marketing Life Insurance, cited above, pp. 775-776.) Therefore, until 1919, there is no way of obtaining data from State reports for companies which operated in only those States for which reports were not available. Subsequent to 1919, the difficulties in compiling complete statistics arise from the lack of uniformity in the various reports with regard to the selection of items to be presented and the basis of reporting, and from the failure of some States to issue reports on a regular annual basis.

The life insurance companies omitted from the sources utilized are very small in size relative to those for which data are available. Therefore, even when a fairly large number of these very small companies are omitted, they account for a very small percentage of the total business. For example, in 1970, according to the Institute of Life Insurance, the 1,390 companies for which life insurance in force data were available accounted for 99.96 percent of the total which would have been obtained from the 1,792 companies in existence at the end of 1970. This percentage is doubtlessly lower for the earlier years and for some of the other categories, but it is highly probable that even the oldest figures presented here represent 90 percent or more of the total for all companies. This is true both with regard to the figures taken from sources, such as the Spectator Insurance Year Book, which collect data from all available companies, and for the figures for about 1860-1880 which are taken from the reports of the New York Insurance Department. (For a discussion of the percentage of total business accounted for by the New York Insurance Department reports, see Hoffman, "Fifty Years of American Life Insurance Progress," cited above, pp. 11-13.)

The third limitation with regard to the data of legal reserve life

insurance companies pertains to the lack of uniformity in allocation of certain items to the categories of the convention blank and changes in allocation. There are many instances where neither the categories of the annual statement convention blank nor the instructions for filing the blank are detailed enough to specify clearly how a certain transaction is to be allocated, so that the treatment becomes a matter of the company's judgment. Thus, for example, of two companies writing monthly debit insurance (a form of life insurance with some of the features of both ordinary and industrial insurance), one may classify it as ordinary and one as industrial. Moreover, a company may decide to change the classification of an item; for the example just cited, a company may transfer at some point its monthly debit business from the industrial to the ordinary classification. Such problems can arise in all the series presented. Even when an accounting procedure tends to become widespread, it is often adopted by different companies at different times.

A further problem arises from the fact that the sources which compile industrywide statistics must often combine the many categories of the annual statement convention blank into broader classifications. From time to time, the manner of combining the categories may be altered or the manner of treating special categories, which are sometimes found in a few companies' convention blank, may be changed.

Related to the problems of changes in the annual statement convention blank and variations in the allocation of items is the problem of changes in method of valuation of policy reserves and assets. The amount of policy reserves reported in a company's convention blank is determined by the types of policies issued, the length of time they have been in force, and the age at issue. The policy reserves are also affected by the mortality table used, the interest assumption, and the reserve basis specified by the various States as the minimum basis for valuation. The assets of a company, and hence its surplus, are also affected by the method of valuation of assets. The problems of changes in valuation of assets and reserves do not appear to be factors of major significance, however, with regard to long-term historical trends of these series.

The general procedure used in preparing these statistics was to examine the various sources and compare the series available as to bases of reporting, completeness of coverage, etc. In those cases where alternative series were available, the selection was determined by completeness of coverage in terms of the number of companies for which data were obtainable, and the basis of reporting most consistent with current practice, on two conditions: (a) That the series be available for a sufficiently long period to preserve the trend, and (b) that component items could be obtained on the same basis or level of coverage as the totals. An illustration might make this clear. For 1879-1887, total assets can be obtained for all the companies operating in New York State. For the same period, totals for a larger group of companies can be obtained from the 1888 Spectator Insurance Year Book. The distribution of assets, by type, however, is available only for the companies operating in New York State. Rather than estimate a distribution for the larger asset totals or report a distribution which would not add to the total shown, the New York State figures were used for the total and for the distribution by type. For the period under discussion, the assets of companies operating in New York State represented from 92 to 97 percent of the assets given by the Spectator Insurance Year Book for all available companies.

In most cases, the various sources were identical with regard to bases of reporting and completeness of coverage. In these cases, the procedure was to compare the various sources presenting the same data for the same period. Thus for the early period, comparisons were made among the individual New York Insurance Reports and the summaries of these reports given in various issues of the Spectator Insurance Year Book, Marketing Life Insurance, and "Fifty Years of American Life Insurance Progress." For later years, comparisons were made among the various issues of the Spectator Insurance Year Book which covered the same period (mainly the 10-year aggregates

as compared with the aggregates given in each Year Book), Marketing Life Insurance, and the Life Insurance Fact Book (which utilizes a great deal of material from the Spectator Insurance Year Book). Where the figures in the various sources were in agreement, the data presented were accepted unless some limitations were uncovered while making the comparisons. Where the sources were not in agreement, the reasons for the differences were investigated and the figures considered to be most accurate and complete were accepted. Some of the figures presented here are original in the sense that they represent adjustments by the Institute of Life Insurance of existing figures for errors in addition, for omissions, or for changes in definition. Wherever possible, published material has been utilized.

X 879. Number of life insurance companies, 1759-1970.

Source: J. Owen Stalson, 1759-1936, Marketing Life Insurance, Its History in America, Harvard University Press, Cambridge, 1942, pp. 748-753. Institute of Life Insurance, 1937-1939, estimates; 1940-1970, Life Insurance Fact Book, 1974, p. 87, and unpublished data.

The figures comprise the total number of companies in operation at the end of the year and domiciled in the United States. This number is larger than the number of companies for which life insurance in force data are available (see general note for series X 879-917). For 1941-1949, figures do not include companies which started and then ceased operations within this period. For data on the number of companies formed, discontinued, and in operation, classified by stock and mutual for 1759-1937, see Stalson, cited above, pp. 748-753.

X 880-889. General note.

For 1854-1894, the series were derived by deducting from the insurance in force figures of U.S. life insurance companies the amount of their Canadian and other foreign business, and adding thereto the U.S. business of Canadian and other foreign companies. Data for 1895-1948 were derived from the totals of individual State estimates given in the "Life Insurance in Force by States" section of each Spectator Insurance Year Book.

For ordinary life insurance, the figures for 1815-1850 are for all available companies; for 1854-1877, the figures are for life insurance companies reporting to the New York Insurance Department. Beginning with 1878, the data are for all available companies. All the data for group, industrial, and credit life insurance are for all available companies.

Life insurance in force is the sum total of the face amounts (plus additions purchased with dividends) of the life insurance outstanding at a given time. The additional amount of life insurance payable under accidental death provisions (providing for payment of an additional death benefit in case of death as a result of accidental means, often called double indemnity) is not included.

Life insurance in force figures have been adjusted to represent insurance in force on the lives of residents of the United States whether issued by U.S. or foreign companies. For statistics of life insurance in force with U.S. life insurance companies, whether the policyholders are residents of the United States or of some other country, and for the number of policies outstanding, for 1900-1970, see Life Insurance Fact Book, 1974, pp. 25, 27, 30, 33, 35. Estimates by States are available from the "Life Insurance in Force by States" section of the annual Spectator Insurance Year Book and the Life Insurance Fact Book. For information on life insurance in force by plan of insurance, 1950, 1954, 1957, 1962, 1966, and 1970, see The Tally of Life Insurance Statistics, January 1959, pp. 1 and 2, March 1968, pp. 1 and 2, and November 1971, pp. 1 and 2.

For an alternative series of life insurance in force in the United States, for selected years, 1815-1937, see Marketing Life Insurance, cited above, pp. 816-817. The alternative series includes fraternal, assessment, and other types of life insurance, and is derived from aggregate figures of U.S., Canadian, and foreign companies, rather than as totals of State figures.

X 880. Number of life insurance policies in force in the United States, 1895-1970.

Source: Institute of Life Insurance, Life Insurance Fact Book, various issues.

Data represent all life insurance in force with U.S. life companies, including both direct business and reinsurance acquired. Data include group certificates and credit life insurance.

X 881. Coverage per family of life insurance in force in the United States, 1930-1970.

Source: See source for series X 880.

Families include the units defined by the Bureau of the Census as families, subfamilies, and unrelated individuals.

X 882. Total life insurance in force in the United States, 1815-1970. Source: 1815-1850, see Stalson, cited above for series X 879, p. 787 (1850 estimate corrected for addition error); 1854-1899, a summation of series X 883 and X 885. 1900-1970, Institute of Life Insurance, Life Insurance Fact Book, 1974, p. 23, and unpublished data.

X 883. Ordinary life insurance in force in the United States, 18151970.

Source: 1815-1850, see Stalson, cited above for series X 879, p. 787. Institute of Life Insurance, 1854-1894, unpublished data; 18951970, Life Insurance Fact Book, 1958, p. 25, and 1974, p. 23.

The 1854-1894 figures were compiled from the following sources, using the method described in the general note for series X 880-889: Ordinary insurance in force of U.S. companies: 1854-1858, Spectator Company, Spectator Insurance Year Book, 1878, p. 71; 1859-1877, Stalson, cited above for series X 879, p. 820; 1878-1894, Spectator Insurance Year Book, various issues (for certain years, adjustments were made). Ordinary business of U.S. companies in Canada: 18691894, Stalson, cited above for series X 879, pp. 833-834 (1873 figure adjusted; 1885-1894, industrial business in Canada of U.S. companies subtracted to get ordinary business in Canada). Ordinary business of U.S. companies in foreign countries other than Canada: 1868–1885, Stalson, cited above for series X 879, p. 824; 1886-1888, Hoffman, "Fifty Years of American Life Insurance Progress," cited above in general note for series X 879-962, p. 86; 1889-1894, Spectator Insurance Year Book, 1899, p. 466. Ordinary business of Canadian companies in the U.S.: 1889-1894, Stalson, cited above for series X 879, p. 839. Ordinary business of other foreign companies in the U.S.: 1854-1870, series for U.S. branches of British companies estimated by the Institute of Life Insurance; 1871-1881, 1885-1886, State of New York Insurance Department, New York Insurance Report, various issues; 1882-1884, data not available, but probably insignificant.

Ordinary life insurance refers to life insurance usually issued in amounts of $1,000 or more, with premiums payable on an annual, semiannual, quarterly, or monthly basis.

X 884. Group life insurance in force in the United States, 1911-1970. Source: Institute of Life Insurance, Life Insurance Fact Book, 1958, p. 27, and 1974, p. 23.

Group life insurance is life insurance issued, usually without medical examination, on a group of persons under a master policy. It is usually issued to an employer for the benefit of employees. The individual members of the group hold certificates as evidence of their insurance.

X 885. Industrial life insurance in force in the United States, 18761970.

Source: 1876-1894, Institute of Life Insurance, unpublished data; 1895-1970, see source for series X 884, 1958, p. 31 and 1974, p. 23.