Flagrant Example of Padded Income Account. In comparatively rare occasions, managements resort to padding their income account by including items in earnings that have no real existence. Perhaps the most flagrant instance of this kind that has come to our knowledge occurred in the 1929—1930 reports of Park and Tilford, Inc., an enterprise with shares listed on the New York Stock Exchange. For these years the company reported net income as follows:
1929—$1,001,130 = $4.72 per share.
1930— 124,563 = 0.57 per share.
An examination of the balance sheets discloses that during these two years the item of Good-will and Trade-marks was written up successively from $1,000,000 to $1,600,000 and then to $2,000,000, and these increases deducted from the expenses for the period. The extraordinary character of the bookkeeping employed will be apparent from a study of the condensed balance sheets as of three dates, shown on page 436.
These figures show a reduction of $1,600,000 in net current assets in 15 months, or $1,000,000 more than the cash dividends paid. This shrinkage was concealed by a $1,000,000 write-up of Good-will and Trade-marks. No statements relating to these amazing entries was vouchsafed to the stockholders in the annual reports or to the New York Stock Exchange in subsequent listing applications. In answer to an individual inquiry, however, the company stated that these additions to Good-will and Trade-marks represented expenditures for advertising and other sales efforts to develop the business of Tintex Company, Inc. a subsidiary.
| Balance Sheet | Sept. 30, 1929 | Dec. 31, 1929 | Dec. 31, 1930 |
|---|---|---|---|
| Assets | |||
| Fixed assets | $1,250,000 | $1,250,000 | $1,250,000 |
| Deferred charges | 132,000 | 163,000 | 32,000 |
| Good-will and Trade-marks | 1,000,000 | 1,600,000 | 2,000,000 |
| Net Current Assets | 4,797,000 | 4,080,000 | 3,154,000 |
| Liabilities: | |||
| Bonds and mortgages | 2,195,000 | 2,195,000 | 2,095,000 |
| Capital and Surplus | 4,984,000 | 4,898,000 | 4,341,000 |
| Total of Assets and Liabilities | $7,179,000 | $7,093,000 | $6,426,000 |
| Adjusted earnings | First 9 months, 1929 | Last 3 months, 1929 | Year, 1929 | Year, 1930 |
|---|---|---|---|---|
| Earnings for stock as reported | $929,000 | $ 72,000 | $1,001,000 | $125,000 |
| Cash dividends paid | 463,000 | 158,000 | 621,000 | 453,000 |
| Charges against surplus | 229,000 | |||
| Added to capital and surplus | 466,000 | decrease 86,000 | 380,000 | decrease 557,000 |
| Earnings for stock as corrected (excluding increase in intangibles and deducting charges to surplus) | 929,000 | 528,000(d) | 401,000 | 504,000(d) |
The charging of current advertising expense to the good-will account is inadmissible under all canons of sound accounting. To do so without any disclosure to the stockholders is still more discreditable. It is difficult to believe, moreover, that the sum of $600,000 could have been expended for this purpose by Park and Tilford in the three months between September 30 and December 31, 1929. The entry appears therefore to have included a recrediting to current income of expenditures made in a previous period, and to that extent the results for the fourth quarter of 1929 may have been flagrantly distorted. Needless to say, no accountant’s certificate accompanied the annual statements of this enterprise.
Balance-sheet and Income-tax Checks upon the Published Earnings Statements. The Park and Tilford case illustrates the necessity of relating an analysis of income accounts to an examination of the appurtenant balance sheets. This is a point that cannot be stressed too strongly, in view of Wall Street’s naïve acceptance of reported income and reported earnings per share. Our example suggests also a further check upon the reliability of the published earnings statements, viz., by the amount of the federal income tax accrued. The taxable profit can be calculated fairly readily from the income-tax accrual, and this profit compared in turn with the earnings reported to stockholders. The two figures should not necessarily be the same, since the intricacies of the tax laws may give rise to a number of divergences. We do not suggest that any effort be made to reconcile the amounts absolutely but only that very wide differences be noted and made the subject of further inquiry.
The Park and Tilford figures analyzed from this viewpoint supply the suggestive results as shown in the table on page 438.
The close correspondence of the tax accrual with the reported income during the earlier period makes the later discrepancy appear the more striking. These figures eloquently cast suspicion upon the truthfulness of the reports made to the stockholders during 1927—1929, at which time considerable manipulation was apparently going on in the shares.
This and other examples discussed herein point strongly to the need for independent audits of corporate statements by certified public accountants. It may be suggested also that annual reports should include a detailed reconcilement of the net earnings reported to the shareholders with the net income upon which the federal tax is paid. In our opinion a good deal of the information relative to minor matters that appears in registration statements and prospectuses might be dispensed with to general advantage; but if, in lieu thereof, the S.E.C. were to require such a reconcilement, the cause of security analysis would be greatly advanced.
| Period | Federal Income Tax Accrued | Rate of tax, per cent | Net income before federal tax as indicated by the tax accrued | Net income before federal tax as reported to the stockholders |
|---|---|---|---|---|
| 5 mo. to Dec. 1925 | $36,881 | 13 | $283,000 | $297,000 |
| 1926 | 66,624 | 13½ | 493,000 | 533,000 |
| 1927 | 51,319 | 13½ | 380,000 | 792,000 |
| 1928 | 79,852 | 12 | 665,000 | 1,315,000 |
| 1929 | 81,623 | 11 | 744,000 | 1,076,000 |
Another Extraordinary Case of Manipulated Accounting. An accounting vagary fully as extraordinary as that of Park and Tilford, though exercising a smaller influence on the reported earnings, was indulged in by United Cigar Stores Company of America, from 1924—1927. The “theory” behind the entries was explained by the company for the first item in May 1927 in a listing application that contained the following paragraphs:
The Company owns several hundred long-term leaseholds on business buildings in the principal cities of the United States, which up until May, 1924, were not set up on the books. Accordingly, at that time they were appraised by the Company and Messrs. F. W. Lafrentz and Company, certified public accountants of New York City, in excess of $20,000,000.
The Board of Directors have, since that time, authorized every three months the setting up among the assets of the Company a portion of this valuation and the capitalization thereof, in the form of dividends, payable in Common Stock at par on the Common Stock on the quarterly basis of 1¼% on the Common Stock issued and outstanding.
The entire capital surplus created in this manner has been absorbed by the issuance of Common Stock at par for an equal amount and accordingly is not a part of the existing surplus of the Company. No cash dividends have been declared out of such capital surplus so created.
The present estimated value of such leaseholds, using the smae basis of appraisal as in 1924, is more than twice the present value shown on the books of the Company.
The effect of the inclusion of “Appreciation of Leaseholds” in earnings is shown herewith:
| Year | Net earnings as reported | Earned per share of common ($25-par basis) | Market range ($25-par basis) | Amount of “Leasehold Appreciation” included in earnings | Earned per share of common excluding lease appreciation |
|---|---|---|---|---|---|
| 1924 | $6,697,000 | $4.69 | $64-43 | $1,248,000 | $3.77 |
| 1925 | 8,813,000 | 5.95 | 116-60 | 1,295,000 | 5.05 |
| 1926 | 9,855,000 | 5.02 | 110-83 | 2,302,000 | 3.81 |
| 1927 | 9,952,000 | 4.63 | 100-81 | 2,437,000 | 3.43 |
In passing judgement on the inclusion of leasehold appreciation in the current earnings of United Cigar Stores, a number of considerations might well be borne in mind.
-
Leaseholds are essentially as much a liability as they are an asset. They are an obligation to pay rent for premises occupied. Ironically enough, these very leaseholds of United Cigar Stores eventually plunged it into bankruptcy.
-
Assuming leaseholds may acquire a capital value to the occupant, such value is highly intangible, and it is contrary to accounting principles to mark up above actual cost the value of such intangibles in a balance sheet.
-
If the value of any capital asset is to be marked up, such enhancement must be credited to Capital Surplus. By no stretch of the imagination can it be considered as income.
-
The $20,000,000 appreciation of the United Cigar Stores leases took place prior to May 1924, but it was treated as income in subsequent years. There was thus no connection between the $2,437,000 appreciation included in the profits of 1927 and the operations or developments of that year.
-
If the leaseholds had really increased in value, the effect should be visible in larger earnings realized from these favorable locations. Any other recognition given this enhancement would mean counting the same value twice. In fact, however, allowing for extensions of the business financed by additional capitalization, the per-share earnings of United Cigar Stores showed no advancing trend.
-
Whatever value is given to leaseholds must be amortized over the life of the lease. If the United Cigar Stores investors were paying a high price for the shares because of earnings produced by these valuable leases, then they should deduct from earnings an allowance to write off this capital value, by the time it disappears through the expiration of the leases. The United Cigar Stores Company continued to amortize its leaseholds on the basis of original cost, which apparently was practically nothing.
The surprising truth of the matter, therefore, is that the effect of the appreciation of leasehold values—if it had occurred—should have been to reduce the subsequent operating profits by an increased amortization charge.
- The padding of the United Cigar Stores income for 1924—1927 was made the more reprehensible by the failure to reveal the facts clearly in the annual reports to shareholders. Disclosure of the essential facts to the New York Stock Exchange was made nearly three years after the practice was initiated. It may have been compelled by legal considerations growing out of the sale to the public at that time of a new issue of preferred stock, underwritten by large financial institutions. The following year the policy of including leasehold appreciation in earnings was discontinued.
These accounting maneuvers of United Cigar Stores may be fairly described, therefore, as the unexplained inclusion in current earnings of an imaginary appreciation of an intangible asset—the asset being in reality a liability, the enhancement being related to a previous period and the proper effect of the appreciation, if it had occurred, being to reduce the subsequent realized earnings by virtue of higher amortization charges.
The federal-income-tax check, described in the Park and Tilford example, will also give interesting results if applied to United Cigar Stores as shown in the table below.
Moral Drawn from Foregoing Examples. A moral of considerable practical utility may be drawn from the United Cigar Stores example. When an enterprise pursues questionable accounting policies, all its securities must be shunned by the investor, no matter how safe or attractive some of them may appear. This is well illustrated by United Cigar Stores Preferred, which made an exceedingly impressive statistical showing for many successive years but later narrowly escaped complete extinction. Investors confronted with the strange bookkeeping detailed above might have reasoned that the issue was still perfectly sound, because, when the overstatement of earnings was corrected, the margin of safety remained more than ample. Such reasoning is fallacious. You cannot make a quantitative deduction to allow for an unscrupulous management; the only way to deal with such situations is to avoid them.
| Year | Federal Tax Reserve | A. Income before tax as indicated by tax reserve | B. Income before tax as reported to stockholders | C. Income before tax as reported to stockholders, less leasehold appreciation |
|---|---|---|---|---|
| 1924 | $700,000 | $5,600,000 | $7,397,000 | $6,149,000 |
| 1925 | 825,000 | 6,346,000 | 9,638,000 | 8,343,000 |
| 1926 | 900,000 | 6,667,000 | 10,755,000 | 8,453,000 |
| 1927 | 900,000 | 6,667,000 | 10,852,000 | 8,415,000 |
| 1928 | 700,000 | 5,833,000 | 9,053,000 | 9,053,000 |
| 1929 | 13,000 | 118,000 | 3,132,000 | 3,132,000 |
| 1930 | none | none | 1,552,000 | 1,552,000 |
Fictitious Value Placed on Stock Dividends Received. From 1922 on most of the United Cigar Stores common shares where held by Tobacco Products Corporation, an enterprise controlled by the same interests. This was an important company, the market value of its shares averaging more than $100,000,000 in 1926 and 1927. The accounting practice of Tobacco Products introduced still another way of padding the income account, viz., by placing a fictitious valuation upon stock dividends received.
For the year 1926 the company’s earnings statement read as follows:
| Net income | $10,790,000 |
| Income tax | 400,000 |
| Class A dividend | 3,136,000 |
| Balance for common stock | 7,254,000 |
| Earned per share | 11 |
| Market range for common | 117-95 |
Detailed information regarding the company’s affairs during that period has never been published (the New York Stock Exchange having been unaccountably willing to list new shares on submissions of an extremely sketchy exhibit). Sufficient information is available, however, to indicate that the net income was made up substantially as follows:
| Rental received from lease of assets to American Tobacco Co. | $2,500,000 |
| Cash dividends on United Cigar Stores common (80% of total paid) | 2,950,000 |
| Stock dividends on United Cigar Stores commont (par value $1,840,000), less expenses | 5,340,000 |
| $10,790,000 |
It is to be noted that Tobacco Products must have valued the stock dividends received from United Cigar Stores at about three times their face value, i.e., at three times the value at which United Cigar charged them against surplus. Presumably the basis of this valuation by Tobacco Products was the market price of United Cigar Stores shares, which price was easily manipulated due to the small amount of stock not owned by Tobacco Products.
When a holding company takes into its income account stock dividends received at a higher value than that assigned them by the subsidiary that pays them, we have a particularly dangerous form of pyramiding of earnings. The New York Stock Exchange, beginning in 1929, has made stringent regulations forbidding this practice. (The point was discussed in Chap. 30, which is on accompanying CD.) In the case of Tobacco Products the device was issued in the first instance to represent a fictitious element of earnings, i.e., the appreciation of leasehold values. By unscrupulous exploitation of the holding-company mechanism these imaginary profits were effectively multiplied by three.
On a consolidated earnings basis, the report of Tobacco Products for 1926 would read as follows:
| American Tobacco Co. lease income, less income tax, etc. | $2,100,000 |
| 80% of earnings on United Cigar Stores common | 6,828,000* |
| $8,928,000 | |
| Class A dividend | 3,136,000 |
| Balance for Common | $5,792,000 |
| Earned per share | $7.27 |
- Excluding leasehold appreciation.
The reported earnings for Tobacco Products common given as $11 per share are seen to have been overstated by about 50%.
It may be stated as a Wall-Street maxim that where manipulation of account is found, stock juggling will be found also in some form or other. Familiarity with the methods of questionable finance should assist the analyst and perhaps even the public, in detecting such practices when they are perpetrated.
This title introduces our second general type of adjustment of reported earnings. When an enterprise controls one or more important subsidiaries, a consolidated income account is necessary to supply a true picture of the year’s operations. Figures showing the parent company’s results only are incomplete and may be quite misleading. As previously remarked, they may either understate the earnings by not showing all the current profits made by the subsidiaries, or they may overstate the earnings by failure to deduct subsidiaries’ losses or by including dividends from subsidiaries in excess of their actual income for the year.
Former and Current Practices. In earlier years disclosure of subsidiaries’ results was a matter of arbitrary election by management, and in many cases important data of this kind were kept secret. For some time prior to 1933 the New York Stock Exchange had insisted in connection with new listings that the results of subsidiaries be presented either in a consolidated statement or separately. But since passage of the 1934 act, all registered companies are required to supply this information in their annual reports to the Commission, and therefore practically all follow the same procedure in their statements to stockholders.
Degree of Consolidation. Even in so-called “consolidated statements” the degree of consolidation varies considerably. Woolworth consolidates its domestic and Canadian subsidiaries but not its foreign affiliates. American Tobacco consolidates only its wholly owned domestic subsidiaries. Most utilities now issue consolidated reports including all companies controlled by them (by ownership of a majority of the voting stock) and deduct the portion of the earnings applicable to others under the heading of “minority interest.” In the railroad field results are rarely consolidated unless the subsidiary is both 100% owned and also operated as an integral part of the system. Hence, Atlantic Coast Line does not reflect its share of the results after dividends of Louisville and Nashville, which is 51% owned but separately operated. The same is true with respect to the 53% voting control of Wheeling and Lake Erie held by the Nickel Plate (New York, Chicago, and St. Louis Railroad Company).
Allowance for Nonconsolidated Profits and Losses. It is now frequent procedure for industrial companies to indicate either in the income account or in a footnote thereto their equity in the profits or losses of nonconsolidated subsidiaries after allowance for dividends.
Examples: The 1938 report of American Tobacco Company showed by way of footnote that dividends received from nonconsolidated subsidiaries exceeded their earnings by $427,000. Hercules Powder reported a similar figure of $257,714 for that year, in footnote form, whereas prior to 1937 it had included its share of the undistributed earnings of such affiliates under the heading “Other Income.” Railroad companies handle this matter differently. The Atchison, for example, now supplies full balance sheet and income account data of affiliates in an Appendix to its own report, which continues to reflect only the dividends received from these companies.
The analyst should adjust the reported earnings for the results of non-consolidated affiliates, if this has not already been done in the income account and if the amounts involved are significant. The criterion here is not the technical question of control but the importance of the holdings.
Examples: On the one hand it is not customary, nor does it seem worth while, to make such calculations with respect to the holdings of Union Pacific of Illinois Central and other railroads. These holdings, although substantial, do not bulk large enough to affect the Union Pacific common stock materially. On the other hand, the adjustment is clearly indicated in the case of the ownership of Chicago, Burlington, and Quincy stock by Northern Pacific and Great Northern, each holding less than a controlling interest (48.6%).
| Year | Du Pont earnings per share | Adjustments to reflect Du Pont’s interest in operating results of General Motors | Earnings per share of Du Pont as adjusted |
|---|---|---|---|
| 1929 | $6.99 | $+$2.07 | $9.06 |
| 1930 | 4.52 | +0.04 | 4.56 |
| 1931 | 4.30 | -0.51 | 3.79 |
| 1932 | 1.81 | -1.35 | 0.46 |
| 1933 | 2.93 | +0.43 | 3.36 |
| 1934 | 3.63 | +0.44 | 4.07 |
| 1935 | 5.02 | +1.30 | 6.32 |
| 1936 | 7.53 | +0.77 | 8.30 |
| 1937 | 7.25 | +0.57 | 7.82 |
| 1938 | 3.74 | +0.61 | 4.35 |
Similarly, the interest of Du Pont in General Motors, representing about 23% of the total issue, is undoubtedly significant enough in its effect on the owning company to warrant adjustment of its earnings to reflect the results of General Motors. This is actually done by Du Pont each year in the form of an adjustment of surplus to reflect the previous year’s change in the book value of its General Motors holdings. The analyst would prefer, however, to make the adjustment concurrently and to include it in the calculated earnings of Du Pont. The effect of such adjustment on the earnings of Du Pont for 1929—1938 is shown in the table on p. 445.
The report of General Motors Corporation for 1931 is worthy of appreciative attention because it includes a supplementary calculation of the kind suggested in this and the previous chapter i.e., exclusive of special and nonrecurring profits or losses and inclusive of General Motors’ interest in the results of nonconsolidated subsidiaries. The report contains the following statement of per-share earnings for 1931 and 1930:
Earnings per share, including the equity in undivided profits or losses of nonconsolidated subsidiaries
| Year | Including nonrecurrent items | Excluding nonrecurrent items |
|---|---|---|
| 1931 | $2.01 | $2.43 |
| 1930 | 3.25 | 3.04 |
Suggested Procedure for Statistical Agencies. Although this procedure may seem to complicate a report, it is in fact a salutary antidote against the oversimplification of common-stock analysis which resulted from exclusive preoccupation with the single figure of per-share earnings. The statistical manuals and agencies have naturally come to feature the per-share earnings in their analyses of corporations. They might, however, perform a more useful service if they omitted a calculation of the per-share earnings in all cases where the company’s reports appear to contain irregularities or complications in any of the following directions and where a satisfactory correction is not practicable.
-
By reason of nonrecurrent items included in income or because of charges to surplus that might properly belong in the income account.
-
Because current results of subsidiaries are not accurately reflected in the parent company’s statements.
-
Because the depreciation and other amortization charges are irregularly computed.
Special Dividends Paid by Subsidiaries. When earnings of non-consolidated subsidiaries are allowed to accumulate in their surplus account, they may be used later to bolster up the results of a poor year by means of a large special dividend paid over to the parent company.
Examples: Such dividends, amounting to $11,000,000, were taken by the Erie Railroad Company in 1922 from the Pennsylvania Coal Company and Hillside Coal and Iron Company. The Northern Pacific Railway Company similarly eked out its depleted earnings in 1930 and 1931 by means of large sums taken as special dividends from the Chicago, Burlington, and Quincy Railrod Company, the Northern Express Company, and the Northwestern Improvement Company, the last being a real-estate, coal and iron-ore subsidiary. The 1931 earnings of the New York, Chicago, and St. Louis Railroad Company included a back dividend of some $1,600,000 on its holdings of Wheeling and Lake Erie Railway Company Prior Preferred Stock, only a part of which was earned in that year by the Wheeling road.
This device of concealing a subsidiary’s profits in good years and drawing upon them in bad ones may seem quite praise-worthy as a method of stabilizing the reported earning power. But such benevolent deceptions are frowned upon by enlightened opinion as illustrated by the more recent regulations of the New York Stock Exchange which insist upon full disclosure of subsidiaries’ earnings. It is the duty of management to disclose the truth and the whole truth about the results of each period; it is the function of the stockholders to deduce the “normal earning power” of their company by averaging out the earnings of prosperity and depression. Manipulation of the reported earning by the management even for the desirable purpose of maintaining them on an even keel is objectionable none the less because it may too readily lead to manipulation for more sinister reasons.
Distorted Earnings through Parent-subsidiary Relationships. Examples are available of the use of the parent-subsidiary relationship to produce astonishing distortions in the reported income. We shall give two illustrations taken from the railroad field. These instances are the more impressive because the stringent account regulations of the Interstate Commerce Commission might be expected to prevent any misrepresentation of earnings.
Examples: In 1925 Western Pacific Railroad Corporation paid dividends of $7,56 upon its preferred stock and $5 upon its common stock. Its income account showed earnings slightly exceeding the dividends paid. These earnings consisted almost entirely of dividends aggregating $4,450,000 received from its operating subsidiary, the Western Pacific Railroad Company. The year’s earnings of the railroad, itself, however, were only $2,450,000. Furthermore its accumulated surplus was insufficient to permit the larger dividend that the parent company desired to report as its income for the year. To achieve this end, the parent company went to the extraordinary lengths of donating the sum of $1,500,000 to the operating company, and it immediately took the same money back as a dividend from its subsidiary. The donation it charged against its surplus; the receipt of the same money as dividends it reported as earnings. In this devious fashion it was able to report $5 “earned” upon its common stock, when in fact the applicable earnings were only aobut $2 per share.
In support of our previous statement that bad accounting practices are contagious, we may point out that the Western Pacific example of 1925 was followed by the New York, Chicago, and St. Louis Railroad Company (“Nickel Plate”) in 1930 and 1931. The details are briefly as follows:
In 1929 Nickel Plate sold, through a subsidiary, its holdings of Pere Marquette stock to Chesapeake and Ohio, which was under the same control. A profit of $10,665,000 was realized on this sale, which gain was properly credited to surplus. In 1930 Nickel Plate needed to increase its income; whereupon it took the $10,665,000 profit out of its surplus, returned it to the subsidiary’s treasury and then took $3,000,000 thereof in the form of a “dividend” from this subsidiary, which it included in its 1930 income. A similar dividend of $2,100,000 was included in the income account for 1931.
These extraordinary devices may have been resorted to for what was considered the necessary purpose of establishing a net income large enough to keep the company’s bonds legal for trust-fund investments. The result, however, was the same as that from all other misleading accounting practices, viz., to lead the public astray and to give those “on the inside” an unfair advantage.
Broader Significance of Subsidiaries’ Losses. We have suggested in this chapter that security analysis must make full allowance for the results of subsidiaries, whether they be profits or losses. But the question may well be raised: Is the loss of a subsidiary necessarily a direct offset against the parent company’s earnings? Why should a company be worth less because it owns something–in this case, an unprofitable interest? Could it not at any time put an end to the loss by selling, liquidating or even abandoning the subsidiary? Hence, if good management is assumed, must we not also assume that the subsidiary losses are at most temporary and therefore to be regarded as nonrecurring items rather than as deductions from normal earnings?
This point is similar to that discussed in the previous chapter relative to idle-plant expense and similar also to the matter of unprofitable divisions of a business, to be touched upon later. There is no one, simple answer to the questions that we have raised. Actually, if the subsidiary could be wound up without an adverse effect upon the rest of the business, it would be logical to view such losses as temporary–since good sense would dictate that in a short time the subsidiary must either become profitable or be disposed of. But if there are important business relations between the parent company and the subsidiary, e.g., if the latter affords an outlet for goods or supplies cheap materials or absorbs an important share of the overhead, then the termination of its losses is not so simple a matter. It may turn out, upon further analysis, that all or a good part of the subsidiary’s loss is a necessary factor in the parent company’s profit. It is not an easy task to determine just what business relationships are involved in each instance. Like so many other elements in analysis, this point usually requires an investigation going well beyond the reported figures. The following examples will illustrate the type of situation and analysis with which we have been dealing.
Example A: Purity Bakeries Corporation. This large maker of bread and cake operates through a number of subsidiaries, of which one of the largest is Cushman’s Sons, Inc., of New York. Cushman’s has outstanding $7 and $8 cumulative preferred stock, not guaranteed by Purity. The annual reports of Purity are on a consolidated basis and show earnings after deduction of full dividends on those Cushman’s preferred shares not owned by Purity, whether earned or paid. The separate reports of Cushman’s reveal that between 1934 and 1937 its operations resulted in a considerable loss to Purity, on its accounting basis, viz.:
| Year | Purity net income as reported | Loss of Cushman’s after full preferred dividends | Purity earnings excluding Cushman’s operations |
|---|---|---|---|
| 1937 | $463 | $426 | $889 |
| 1936 | 690 | 620 | 1,310 |
| 1935 | 225 (d.) | 930 | 678 |
| 1934 | 209 | 173 | 382 |
| Average 4 years | 278 | 537 | 815 |
| Per share of Purity | 0.36 | 0.71 | 1.06 |
The earnings are thus seen to be three times as large excluding Cushman’s as they were including Cushman’s. Could the analyst have reasoned that the former provides the truer measure of Purity’s earning power, since the company can be expected either again to earn money from that subsidiary (as it had earned it in the past up to 1934) or to drop it? The question of inter-corporate relationships would have to be considered. A note in the 1937 report of Cushman’s indicated that Purity was making a fairly large service charge in connection with its subsidiaries’ operations, which suggests that Cushman’s might be of some extra value in absorbing overhead. This matter would call for a careful inquiry.
But the report for the next year, 1938, showed, first, that Cushman’s had earned the preferred dividend deduction, and secondly, that two unprofitable retail plants (in Philadelphia and Chicago) had been closed. Subject to further investigation, therefore, the analyst might well infer that the subsidiary’s losses were nonpermanent in nature and that the reported results for 1934—1937 are to be viewed with this point in mind.
Example B: Lehigh Coal and Navigation Company. This enterprise has derived its income from various sources, chief of which has been the lease of its railroad property to the Central Railroad of New Jersey for an annual rental of $2,268,000. Its next largest holding consists of anthracite coal mines, which since 1930 have been operated at a loss. In 1937 this loss was equivalent to about 90 cents per share of Lehigh stock. As a result the company reported a consolidated net loss of $307,000 for the year, as contrasted with a profit on a parent-company basis only of $1,125,000, or 64 cents per share.
But in this case the analyst could not safely make the assumption that the Lehigh stock was not worth less by reason of its ownership of the mining properties than it would be worth without them. Operation of the mines supplied an important tonnage to the railroad division. If the mines were shut down, the ability of the Jersey Central to pay the annual rental might have been critically impaired, especially since the lessee road had been doing poorly for some years past. (In fact the claim was later made by the Jersey Central that the Lehigh Coal and Navigation was obligated in connection with the lease to supply a certain tonnage from its coal properties). Hence, in this rather complicated set-up the investor could not safely go behind the consolidated results, including the losses of the anthracite subsidiary.
Example C: Barnsdall Oil Company. We have here a situation opposite from the other two. Barnsdall Oil owned both refining and producing properties, the latter profitable, the former unprofitable. In 1935 it segregated the refineries (and marketing units) in a separate company, of which it distributed the common stock to its own stockholders, retaining, however, the preferred stock and substantial claims against the new company. In 1936–1938 the refineries and stations continued to lose; Barnsdall Oil advanced considerable sums to cover these losses and wrote them off by charges first against capital surplus and then against earned surplus. On the other hand, its income account, freed from the burden of these refining losses, showed profits from producing operations at a steady rate from June 1, 1933, to the end of 1938.
In 1939, however, the New York Stock Exchange called upon the company to correct its statements to stockholders by advising them of the effect upon the reported profits of charging there-against the write-downs of the investment in the refining company. These losses would have reduced the indicated profits by more than one-third.
It is clear, from the standpoint of proper accounting, that as long as a company continues to control an unprofitable division, its losses must be shown as deductions from its other earnings. The analyst must decide what the chances are of terminating the losses in the future, and view the current price of the stock accordingly. The method followed by the Barnsdall Oil Company appears therefore clearly open to criticism, since it served merely to terminate the reporting of its refining losses without really terminating the losses themselves. (At the end of 1939 the company set steps into motion for an apparent complete divorcement and sale of the refining and marketing divisions.)
Summary. To avoid leaving this point in confusion, we shall summarize our treatment by suggesting:
-
In the first instance, subsidiary losses are to be deducted in every analysis.
-
If the amount involved is significant, the analyst should investigate whether or not the losses may be subject to early termination.
-
If the result of this examination is favorable, the analyst may consider all or part of the subsidiary’s loss as the equivalent of a nonrecurring item.