123. 2

91.4

88.9

80. 2

19 77. 0

72.0

Chicago, Milwaukee, St. Paul and Pacific R. R. Co____

The Illinois Central R. R. Co____

Missouri Pacific R. R. Co....

Chicago & Northwestern Ry. Co.............

13 To allow for unconsolidated subsidiaries, in 1929 the total asset figure given in Moody's was increased by a small percentage. The 1935 figure was increased by the same percentage.

14 Total assets, less reserve for depreciation, of New York, Chicago & St. Louis R. R. Co.; The Wheeling & Lake Erie Ry. Co.; Erie R. R. Co.; Chesapeake & Ohio Ry. Co.; and the Pere Marquette Ry. Co.

15 Total assets, plus 50 percent of total assets of Chicago, Burlington & Quincy R. R. Co., and 50 percent of total assets of Spokane, Portland & Seattle Ry. Co. These two companies are controlled jointly by the Great Northern Ry. Co. and Northern Pacific Ry. Co.

16 Assets of Atlantic Coast Line R. R. Co., less depreciation, plus assets of Louisville & Nashville R. R. Co., less depreciation.

Florida East Coast Ry. Co...
Chicago Union Station Co...

Chicago & Western Indiana R. R. Co..

Chicago & Eastern Illinois Ry. Co.............
Terminal Railroad Association of St. Louis.
Minneapolis & St. Louis R. R. Co............

17 Assets of Reading Co., less depreciation, plus assets of Central R. R. Co. of New Jersey, less investment of Reading Co. in affiliated companies.

18 Total assets of Western Pacific R. R. Corporation, less investment in Western Pacific R. R. Co., plus total assets of Western Pacific R. R. Co., less depreciation. 19 Owned by 15 roads which use the terminals in St. Louis.

1. Methods and Procedures

Definitions

In the basic data for this study, a corporation is simply a corporate unit filing an income-tax return with the Bureau of Internal Revenue. In some years corporations were permitted to submit consolidated income-tax returns, and hence income statements and balance sheets, which included all subsidiaries, 95 percent or more of whose voting stock was held by the parent or other 95 percent-controlled subsidiaries of the same parent. All such subsidiaries whose returns were consolidated with those of their parents are herein called consolidated subsidiaries.

When the term 200 corporations is used in this study what is intended is really 200 corporate units of control, and consolidated subsidiaries are therefore included. In considering units of control, all subsidiaries controlled should be included. The Interstate Commerce Commission considers the ownership of over 50 percent of the voting stock a sufficient condition to call a corporation a subsidiary. The Securities and Exchange Commission is legally permitted to call any company an "actively controlled" subsidiary if there is any evidence that actual control is exercised, whether 50 percent of the voting stock is held by the parent or not. In practice, the Securities and Exchange Commission ordinarily uses the term to apply to companies 50 percent or more of whose voting stock is held by one corporate unit of control. All such companies, however, do not appear upon their records. In this report all companies in which a majority of the voting stock is held by any corporation or combination of corporations in or controlled by the 200 largest are called subsidiaries. If the income tax return of such a company is not consolidated with that of its parent it is called an unconsolidated subsidiary. A corporation, then, is a corporate unit of control and is composed of a parent. corporation, its consolidated subsidiaries, if any, and its unconsolidated subsidiaries which meet the majority control criterion. Actually there may be, and undoubtedly often are, other actively controlled corpora

1 Appendix 11 was prepared by Ezra Glaser and Betti Goldwasser; some preliminary work done by Robert L. Smith. While all of the data in this appendix have been compiled directly from income-tax records except where specifically stated otherwise, the returns to be compiled have been selected and classified on the basis of independent information derived from other sources so that the compiled figures are nowhere available in the published or unpublished records of the Bureau of Internal Revenue.

tions in the unit, but this criterion does not class them as subsidiaries. They are, therefore, excluded from the corporation, as here defined, and so from the totals for the 200 largest.

It should be emphasized that in employing this definition of subsidiary there is no intention to imply that a given proportion of stock ownership carries with it actual control. Majority stock ownership is an indication of ability to control, but is evidence neither of the minimum amount of ownership necessary for control nor yet of complete domination. Defining subsidiary in this way carries no implications of complete ownership. A parent may actually control the policies of a subsidiary even if it owns much less than 50 percent of its assets. Majority stock ownership is assumed to be evidence of ability to control, not of actual control, nor of complete ownership.

From practical considerations of statistical procedure and with a limited amount of time and clerical assistance available, all unconsolidated subsidiaries with total assets under 10 million dollars in 1933 were eliminated from the tabulation of the 200, no matter what percent of the voting stock was owned by the parent. The consolidated subsidiaries were included, however, since their assets could not be isolated from those of the parent corporation in the consolidated returns.

For 1929 a corresponding minimum for total assets was derived, the figure being close to 14 million dollars. All unconsolidated subsidiaries below the minimum were discarded. The details of the method of arriving at the corresponding minimum for 1929 and its interpretation are to be found in section 3.

Derivation of Lists of 200 Largest
Corporations in 1929 and 1933

The lists of the 200 largest corporations were determined independently for 1929 and 1933. They were chosen on the basis of total assets shown by corporate returns filed with the Bureau of Internal Revenue plus total assets of all unconsolidated subsidiaries with assets of more than the minimum described in the previous section. Pure holding companies and financial companies were excluded, unless they were subsidiaries of parents on the list.

While this report uses total assets minus taxable investments as a general measure of size, the lists were

* See section 4 for the significance of this measure of size.

2

compiled on the basis of total assets (including taxable investments). The use of total assets results in the inclusion of some companies which would have been too small had total assets less taxable investments been the criterion of choice. Certain other companies, whose taxable investments formed a smaller proportion of their total assets, were excluded from the 200 because their total assets were not large enough. Therefore, the list compiled is not actually the 200 largest measured by total assets less taxable investments.3

The procedure for deriving the lists was as follows: For 1929, the returns of nonfinancial corporations with assets of 50 million dollars or more were inspected, and the companies arranged in order of size. This list was supplemented by comparisons with lists of large companies from other sources, such as the Interstate Commerce Commission and Moody's Manuals. This cross-comparison acted as a dragnet for corporations which submitted no balance sheets, and consequently no figure for total assets, to the Bureau of Internal Revenue. The 250 largest corporations, measured by total assets, were chosen for further inspection. (Consolidated subsidiaries were included, by necessity, in these corporations.)

The same sources were used to check all nonfinancial corporations with assets over $14,023,000 for which Bureau of Internal Revenue data were available, to discover subsidiaries of the 250 largest corporations with total assets over the minimum. A compilation of total assets for each of the 250 largest corporations and its unconsolidated subsidiaries with assets over $14,023,000 was made. From this, the 200 largest corporations were selected, measuring size by the sum of the assets of each parent and its unconsolidated subsidiaries with assets over the minimum.

Essentially the same procedure was used for 1933, except that the minimum for unconsolidated subsidiaries was set at 10 million dollars (10 million dollars bearing the same relation to total assets in 1933 as $14,023,000 did to total assets in 1929), and the records of the Securities and Exchange Commission were used to supplement the other sources consulted.

4

An important exceptional case should be noted. One large corporation, with assets of over three billion dollars, according to Moody's Manuals, did not submit a balance sheet in 1933. The procedure of searching through Bureau of Internal Revenue materials for corporations with large total assets failed to reveal a balance sheet for this corporation, although the corporation had submitted a return. Since the company did file a

See p. 282 for an appraisal of the error thus introduced.

A further exceptional case, in both 1929 and 1933, is that of a company which existed on paper only, representing a proposed merger through exchange of securities, of two large existing companies and their subsidiaries. Actually, the merger was never completed; for the purposes of this tabulation, however, the two independent companies were considered as active subsidiaries of an inactive parent (the paper company) which had no income statement or balance sheet apart from its subsidiaries. The two companies were operated as a unit.

balance sheet in 1929, its absence from the 1933 list was conspicuous. Estimates of its 1933 balance-sheet items had been made by the Bureau of Internal Revenue by combining the balance-sheet items of its constituent companies, and these figures were used, as omission of the company and its subsidiaries would have caused a serious distortion in the total asset item for the 200 largest corporations-a difference of over 3 percent.

Comparison of the 1929 and 1933 lists of the 200 largest corporate units with each other as well as with the lists of the largest corporate units compiled directly from Moody's Manuals and discussed in Appendix 10 did not disclose any other case in which a very large corporation failed to file a balance sheet in the years mentioned. If a small company, near the lower limits set by the total assets criterion, failed to submit a balance sheet but was on a list compiled from other sources, it made little difference if it was omitted and another small company substituted. The discrepancy which may have been introduced in this way is undoubtedly smaller than the error which would have been involved in an estimate of the missing balance-sheet items for companies which did not submit them.

The lists therefore include:

(a) The 200 largest nonfinancial parent corporations in each year, chosen on the basis of total assets as shown by the consolidated returns, plus total assets of all unconsolidated subsidiaries with total assets over the minimum;

(b) Subsidiaries of the 200 largest parent corporations whose returns are consolidated with those of their parents. (Parent companies were allowed to submit consolidated returns for all subsidiaries over which they had 95 percent or more control, measured by ownership of voting stock, but this was not compulsory. As a result, some subsidiaries subject to 95-100 percent control are probably not included. Subsidiaries thus consolidated may be of any size, and the majority have smaller total assets than the minimum):

(c) Subsidiaries, showing assets greater than the minimum, which did not submit consolidated returns. Control of these by the parents ranged from 50 to 100 percent.

The procedure employed probably understates the 200 largest corporations, as no unconsolidated subsidiaries were included in the 200 unless balance sheets were submitted for them. That several such omissions were made is obvious from the most cursory search of Moody's Manuals. Other omissions may have been made because the information on stock control in the sources consulted did not reveal all the corporations which should have been classified as subsidiaries by the present definition. Where the subsidiary status of a company was doubtful, the error is deliberately on the side of conservatism. While the influence of missing companies is impossible to estimate accurately, the errors of omission were estimated to be very small.

From the discussion above it is evident that there may be one or two more or less than 200 corporations on the list. The former is likely if a corporation has been counted as a subsidiary which should have been considered independent. The latter is likely if a corporation has been counted as independent when it should have been considered a subsidiary of one of the corporations on the list. The error thus introduced is of slight importance. The ten smallest companies on the list contributed only 1.0 percent to total assets of the 200 in 1929, and 0.9 percent in 1933.5 Omission or addition of one or two corporations at the very bottom of the list would therefore subtract or add a very small percentage of the total assets, and this percentage would be well within the range of error of the study. Similarly, if unconsolidated subsidiaries have been omitted because of missing balance sheets or insufficient information in the sources consulted, the choice of the 200 companies may have been affected. Again, the companies affected would probably be near the borderline, and the error would be insignificant. A rough guess puts the error due to these sources at something less than 2 percent, probably in the direction of understatement, as classification of the companies as subsidiaries was made as conservative as possible.

Classification by Industrial Group

The breakdown of the list of corporations into industrial groups can be compared with Bureau of Internal Revenue classification, as follows:

All nonfinancial corporations include corporations classified in Statistics of Income under the heading "Aggregate", less those classified under "Finance".

Manufacturing is classified by the same definition as "Total manufacturing" in Statistics of Income.

Transportation and other public utilities are classified by the same title in Statistics of Income. This classification has been broken down for the present tabulation according to the Bureau of Internal Revenue's subclassifications, "Transportation and related activities" and "Other public utilities", as shown in table 14 of the 1933 Statistics of Income.

"Other nonfinancial corporations" include corporations classified in Statistics of Income under the headings. "Agriculture and related industries," "Mining and quarrying", "Construction", "Trade", "Service", and "Nature of business not given."

The correspondence with the Bureau of Internal Revenue industrial classification was maintained for the calculation of concentration ratios.

Consolidated subsidiaries are, of course, subject to the industrial classification of their parents. This is true both for the tabulations of the 200 corporations and for Statistics of Income. Unconsolidated subsid

5 Four other items were investigated and showed substantially the same percent contribution for the 10 smallest corporations.

• The tables in Statistics of Income which were used to derive the concentration ratios did not have this breakdown. The necessary totals were taken from office worksheets of the Bureau of Internal Revenue.

iaries were tabulated for the 200 according to the industry of their parents. This is not true of the Bureau of Internal Revenue practice; for tax purposes these subsidiaries are independent and hence are classified independently. This difference in classification distorts the concentration ratios of the industrial groups by an unknown amount. No evidence of change in bias could be found over the period 1929-33.

The number of parent corporations in the 200 included in each industrial group in each year, and the corresponding information for unconsolidated subsidiaries, are shown in table I together with the percentage distribution of the corporations by industrial group.

TABLE I.-Number and percentage distribution of returns of parents and unconsolidated subsidiaries tabulated for the 200 largest nonfinancial corporations, by industrial groups, 1929 and 1933 1

The items tabulated have been defined, where necessary, in footnotes to the tables. They are comparable to items in Statistics of Income. Further description follows:

(a) Inventories.-Although Statistics of Income shows a total for inventories for all corporations submitting balance sheets, the total does not represent comparable contributions from the various industrial groups, because of the different usage of the word "inventory" by them. Except for one classification, the term primarily signifies inventory to be sold. For transportation and other public utilities, the term is used primarily to represent inventory for use (actually working capital in the form of fuel supplies, reserves of equipment, etc.). For transportation and other public utilities it represents an asset item which is not comparable in economic significance to the same asset item for other industrial classifications.

(b) Taxable investments.-This item is comparable to the item "Investments other than tax-exempt" in Statistics of Income. Stocks of corporations were in

cluded in this classification because the securities themselves were not tax-exempt. Dividends paid on these stocks to individuals were taxable. However, the tax laws did not require dividends to be included in the taxable income of a corporation in the period studied. Hence, "Taxable investments" includes all investments except the obligations of political units, although actually corporations did not pay income tax on dividends received from corporate stocks.

(c) Miscellaneous assets.-Miscellaneous assets is, for the 200 and for Statistics of Income, a remainder, derived by subtracting the specific asset items from total assets. When the balance-sheet statements are given in detail by a company, miscellaneous assets represents primarily good will and patent rights, and certain liquid assets not elsewhere listed. However, for those corporations which submit sketchy balance sheets, miscellaneous assets may contain in whole or in part asset items which should be reported elsewhere in the balance sheet. The figures tabulated are comparable to Statistics of Income for this item; the sole difficulty lies in interpreting the totals for the item.

(d) Total assets less taxable investments.-As has been mentioned, this item is considered more indicative of the assets controlled by a corporation than total assets. Total assets is subject to a heavy inflation because of security holdings of related corporations, but the worst part of this duplication of assets is removed by subtracting taxable investments. Further comment will be found in section 4.

(e) Receipts. The item tabulated as receipts represents the combination of gross sales (where inventories are an income-determining factor) with receipts from other operations. This seems desirable because of the arbitrary division of these two items in Statistics of Income. According to Bureau of Internal Revenue practice, income of transportation and other public utilities, and of finance companies, is never classified as gross sales, no matter what the source, even when it results from sales of inventory. The difficulty of comparing receipts for the various industrial groups when the rigid Bureau of Internal Revenue practice is followed makes it advisable to sum the two items.

(f) Income tax.-It should be remembered that excess-profits taxes were not paid before 1933, and are therefore not included with income tax before that year. For 1933, the Federal income tax and the excess-profits tax were summed, as both are taxes on income. tax thus paid is paid on an income larger than that revealed in the item tabulated as statutory net income less statutory net deficit in Statistics of Income." The

The

7 The accompanying tables include compiled net profit or loss, which differs from statutory net income in that it includes interest on tax-exempt investments and dividends received. Statutory net income less statutory net deficit may easily be computed from the accompanying tables by subtracting tax-exempt interest and dividends received from compiled net profit or loss.

result of subtracting the deficit is to conceal the amount of income actually taxed.

(g) Interest received from tax-exempt investments.-The relation of this item to the actual holdings of tax-exempt investments as shown by balance sheets is not clear. The balance-sheet item represents the estimated values of holdings as of December 31. The interest recorded is an income item covering the whole year, and does not, of course, necessarily correspond to security holdings of a given date. It is, however, an absolute and measurable quantity, unlike the value of holdings of tax-exempt investments. The totals for the latter depend on the basis used for valuation.

Two other items included in the accompanying tables were derived from items recorded on tax returns and in Statistics of Income. A measure of the income from operations was derived by combining compiled net profit and interest paid, and subtracting from the result income tax, interest received on taxable investments, interest received on tax-exempt investments, and cash dividends received. This amounts to taxable (or statutory) net income, which is not tabulated, plus interest paid, less income tax, less interest received on taxable investments.

A measure of corporate savings was also derived. This equaled compiled net profit, less income and excessprofits taxes paid, less cash dividends paid. This item is frequently negative, showing that dividends were paid out of reserves, not out of current income.

Tabulation of the Bureau of Internal Revenue Data

The income-statement and balance-sheet items were tabulated for 1929 and 1933 for the 200 largest nonfinancial corporations by industrial group, and for their unconsolidated subsidiaries, arranged by the industrial group of the parent. Totals were computed for each item for each industrial group, and these were summed to give the total for the 200.

It should be emphasized that the records from which the tabulations were made were compiled from data available in the Statistical Section of the Income Tax Unit of the Bureau of Internal Revenue, which were compiled from unaudited returns. Tables in Statistics of Income are compiled from the same records, so there is no lack of comparability from this source. However, should the returns of large corporations be more likely to be subjected to change after auditing than the returns of smaller corporations, it becomes apparent that the totals from audited returns would show a much greater percentage change from the present totals for the 200 than for all nonfinancial corporations. No quantitative estimate of the shift can be made. In this case, it is probable that the concentration ratios for certain income items would be raised.