Two striking indications from the stockholding data should be noted: (1) There has been no significant growth in the stockownership of foundations relative to the total market since 1949; and (2) there has been a small decline in the share of college and university endowments. The total share of all tax-exempt organizations (other than pension funds) was almost unchanged but down slightly.

TABLE 7.-Data on total assets of foundations and higher education endowments [Dollar amounts in billions]

1 This refers only to the endowment in investment assets. Physical plant of colleges and universities also serves as endowments, yielding services rather than cash. If these were included, higher education endowments would exceed those of foundations.

* 1964 Treasury Department Survey of Private Foundations.

Cols. (1) and (2):

SOURCES

1930: "American Foundations and Their Fields," Twentieth Century Fund. The tabulation contained in this report lists foundations with assets of $853,000,000, but 17 of the 122 foundations did not submit asset figures. The report contains the estimate that for all 122 foundations an asset figure of $950,000,000 "is probably not wide of the mark."

1944: "American Foundations for Social Welfare," Harrison and Andrews, Russell Sage Foundation, 1946, p. 58.

1950: "Philanthropic Giving," Andrews, Russell Sage Foundation, 1953, p. 93.

1954: "American Foundations and Their Fields," 7th ed.

1959: "Foundation Directory 1," Russell Sage Foundation, 1960.

1962: "Foundation Directory 2," Russell Sage Foundation, 1964.

Col. (3):
1930-59: Office of Education,

1962: "Giving U.S.A.," 1963 ed., p. 14, American Association of Fund Raising Counsel. Col. (4):

1930-54: "Studies in National Balance Sheet of United States," Goldsmith, vol. II, pp. 124-125. The 1930 figures were interpolated between Goldsmith's estimates for 1929 and 1932 on the basis of aggregate value of corporate shares.

1959-62: "Flow of Funds Accounts," FRB. Total assets were estimated using observed trend of ratio of total to intangible in Goldsmith's data.

TABLE 8.-Estimated holdings of New York Stock Exchange listed stocks by certain exempt institutions

Source: "NYSE Fact Book," 1963 and 1964. Ford figures obtained from Ford Foundation. The 1949 figure was obtained using the book equity of the Ford Motor Co.

The two sets of data in tables 7 and 8 seem to suggest two different conclusions about the relative growth of foundations. The total estimates in table 7 suggest a growth in the relative share continuing through the 1950's. The stockholding data in table 8, however, suggest a cessation in the growth in the relative share of foundations after 1950.

The quality of the data available does not admit of any precise reconciliation of these two sets of statistics. The early survey was admittedly incomplete as to coverage of foundations, and this coverage gradually improved. Also, the later surveys reflected a mixture of market values and ledger values. The stockholding data are based on a limited sample.

A large part of the discrepancy is accounted for by the fact that foundations have a very large portion of their investment in common stock compared to individuals and even compared to higher education endowments. Common stock has advanced far more in price in the last 15 years than other assets. This has been caused by both the growth in dividends and an increase in the price-earnings ratio. The implications of the stockholding data are that stock investments of foundations were not growing faster than the stock investments of other stock investors. All stock investors were gaining compared to people who owned just bonds, bank accounts, and insurance. Since foundations are heavily invested in stocks, this resulted in better than average growth for foundations, compared to total individual wealth.

If foundations were growing faster than other investors due to either an increasing flow of contributions or due to a parsimonious policy of distribution to charity, this should show up in the NYSE data as growth relative to other stock investors. It is significant that there is so little growth of this sort in the NYSE data.

42-663-65- 6

Another evidence of foundation growth is afforded by recording the organization dates of presently large foundations. This serves to identify the 1940's and 1950's as the period of rapid foundation growth, although it is striking that the foundations established since 1950 are relatively small compared to those established before 1950. These data are contained in table 9.

On the basis of the meager evidence available, the following conclusions are suggested about private foundation growth:

(a) There was some growth of foundations relative to the rest of the economy in the 1930's and 1940's. This can be associated with the adoption of increased progressivity in estate and income taxes in the early 1930's plus the charitable contribution deduction under each tax.

(b) Since 1950, the total wealth of foundations has grown faster than the rest of the economy, but in this period the faster growth was probably due to the fact that their principal assets and corporate stocks were increasing in price faster than other assets. In terms of values of shares owned, the proportion owned by foundations appears to have been quite stable.

TABLE 9.-Period of establishment of 5,050 foundations, by decades after 1900: by latest asset classes 1

1 The 5,050 foundations tabulated here are those that had at least $100,000 of assets in 1962 and were thus included in the "Foundation Directory" and which also provided information to the Foundation Library Center as to date of organization.

2 Less than 0.5 percent.

3 Record incomplete; the fragmentary 1960-record (45 foundations) not included in table. Source: "Foundation Directory," ed. 2, p. 13.

4. 1964 survey of foundations

In 1964 the Treasury Department conducted a survey of certain financial aspects of private foundations. The survey involved initially selecting a sample of approximately 1,300 organizations whose Form 990-A was available (principally at the Foundation Library Center office in Washington, D.C.).

Certain parts of the information return, Form 990-A, are required by law to be made available to the public. The Foundation Library Center, a private, nonprofit organization, maintains a file of copies of this public part of the tax return for those exempt organizations which meet their definition of a foundation. The "Foundation Directory,"

In the conduct of this survey assistance was obtained from the Internal Revenue Service and the Foundation Library Center offices in Washington, D.C., and New York City.

edition 2, page 9, published in 1964, explains the definition of a foundation used by the Center as follows:

For purposes of this directory a foundation may be defined as a nongovernmental, nonprofit organization having a principal fund of its own, managed by its own trustees or directors, and established to maintain or aid social, educational, charitable, religious, or other activities serving the common welfare. Both charitable trusts and corporations are included. As previously, the new directory excludes "foundations" which make a general appeal to the public for funds; which act as trade associations for industrial or other special groups; which are restricted by charter solely to aiding one or several named institutions; or which function as endowments set up for special purposes within colleges, churches, or other organizations and are governed by the trustees of the parent institution. Obviously, many "foundations" fall in a gray area, with most of the characteristics of regular foundations but some disqualifications; edition 2 interprets the exclusions more rigidly than did its predecessor.

The "Foundation Directory" published by the Foundation Library Center omits "very small" foundations. The files of the Foundation Library Center do, however, contain copies of the Form 990-A for many of these very small foundations.

Since the word "foundation" is not technically defined for tax purposes, there is no ready way to separate those organizations called foundations from other tax-exempt organizations so far as tax information returns are concerned. As a means of obtaining a body of statistical information, it seemed necessary to utilize the classification which had been established by the Foundation Library Center. Data have been added for certain very large organizations which one might want to define as a foundation where these could be identified.10 No effort was made to expand the center's definition in the other size categories. The Foundation Library Center indicates that their records show that there were approximately 15,000 foundations, according to their definition, in existence around the end of 1962. Of these, an estimated 9,000 were below $100,000 in total assets.

A stratified sampling design was adopted that would produce a sample of about 1,300 foundations. It developed that the 1962 Form 990-A was available in the Foundation Library Center for only about one-half of the total number of foundations. This was principally due to delays involved in obtaining and reproducing the returns. The sampling rates for the foundations below $1 million in size were accordingly doubled, and in the group of foundations with assets size of over $10 million other sources were utilized to obtain the Forms 990-A for the year 1962 in order to carry out the plan to have 100 percent coverage in this area. Information was taken from the Form 990-A, and a supplementary questionnaire was sent to each of the foundations whose return was selected." In the aggregate a response rate of close to 98 percent was realized."

10 Since the particular concern of the present study was private foundations, several community foundations which could be readily identified were omitted from the tabulation.

11 Copies of the Form 990-A (including instructions) and the supplemental questionnaire are attached as exhibits 1 and 2, respectively.

12 When the initial machine tabulation of results was run, the response rate to the questionnaire was about 96 percent. Those organizations from which a questionnaire was not received were tabulated in a special category called unclassified. The results of the initial run were adjusted in the very large category so as to shift several foundations from unclassified to the appropriate donor influence category on the basis of the questionnaire when it was received. Further, for several tabulations of market value asset data, the 2 percent of questionnaires received after the initial tabulations were taken into account. In the remaining cases where negligible effects would be involved, these last 2 percent of questionnaires received were not reallocated from the unclassified category tabulated. The total market value of assets of the unclassified category was calculated, where necessary, by raising the ledger values on stockholdings on the basis of market to ledger ratios for stockholdings on those foundations reporting market values.

The discussion in the following sections is based upon a tabulation of the return forms and questionnaire results. The statistics collected in the sample have been blown up to provide an estimate of the data for all 15,000 foundations. In the tables the small foundations are those whose assets at the beginning of 1962 were under $100,000. The medium foundations had assets of $100,000 to $1 million. The large foundations had assets of $1 million to $10 million. The very large foundations had assets over $10 million.

5. The income of foundations

In 1962 foundations in the aggregate had $1,065 million of total income after investment expenses, but including capital gains. Some material on the aggregate income of foundations is given in table 10.