Jenks holds that the competition that arises in this way is more far-reaching than under a competitive régime, since the exact cost and efficiency at the separate plants are known, and the pressure to make a good showing is great.1 The argument has weight, yet it is a question whether a salaried manager, even with a stake in the profits, would strive as earnestly to make a good showing as an independent business man for whom inefficiency means not only the failure to secure large profits, but the actual sustaining of losses. Be that as it may, the gain that accrues from competition among plants is open to the combination as well as the trust, though doubtless to a lesser extent. However, no gain on this score is possible, even to the trust, when the practice has been followed of specializing each plant on the production of a given shape or size. Obviously if each plant is devoted to the manufacture of a single machine or size to secure to the fullest extent the economies of large-scale production, no basis exists for a close comparative study of the separate plants. This point illustrates what many writers have apparently not seen, namely, that to enumerate a list of trust economies and to illustrate each by some one trust leaves the reader with the mistaken impression that each economy may be realized by all trusts. The fact is, as will be pointed out in discussing other economies, that the resort to one saving often precludes the employment of another; and that a number of economies may be availed of only by a comparatively few trusts.

(6) Utilization of by-products. Trusts can effect as complete a utilization of by-products as is feasible. Yet this saving results from production on a large scale rather than from combination. That this is the case is indicated in the fact that many writers fail to mention the utilization of by-products in enumerating the advantages of trusts. Those writers who do mention it commonly cite the meat-packing plants, and refer to the conspicuous absence of waste in these establishments. The fact is, however, that such combination as exists in this business is of the nature of a pool rather than of a trust. The leading companies have 1 Jenks, Trusts and Industrial Combinations, Bulletin of the Department of Labor, vol. V, no. 29, p. 675.

1

undoubtedly largely eliminated competition, yet they have done so through agreement rather than through combination; and it is by virtue of the large scale of their individual operations rather than by virtue of their agreement that they realize these economies. Moreover, in some industries there are no by-products, and therefore there is no need of a trust to insure their effective utilization. Thus, in the manufacture of many steel and tobacco products there are no by-products; and in the refining of cane sugar there are only two of importance (molasses and sirup).1

(7) Insurance. Trusts can save in insurance, particularly fire insurance, because of the wide distribution of their plants. This possibility exists in some instances, yet not in all. Thus, the cash register trust with only one factory has no such distribution; and other concerns, the sugar trust, for example, have only a few plants, and therefore find it necessary to continue their insurance. Some combinations, such as the American Agricultural Chemical Company, have a wider distribution of plants, and thus of risks, than many trusts. It is quite obvious that the saving in insurance, when it exists, has not been of sufficient consequence to exert any considerable influence on the trust movement.

(8) Smaller fixed charges per unit of product. The trust with its large output has an advantage because its fixed charges are distributed over more units, and thus the expense per unit is less. This, however, is really an argument for large-scale production, and not for the trust. In every industry that has a national market, plants of a certain size secure full economic efficiency; and any expansion of the plant beyond this point, unless it takes the form of a complete duplication of the plant, is more likely to increase operating costs than to reduce them.2

1 Report of the United States Tariff Commission on Costs of Production in the Sugar Industry, p. 23.

This opinion is generally held by leading economists. See Taussig, Principles of Economics, I, p. 59; Ely, Monopolies and Trusts, p. 165; Bullock, Quarterly Journal of Economics, 15, p. 198; and Durand, The Trust Problem, p. 69.

This follows from the fact that a plant of a given size permits machinery to be employed in the most effective manner, and permits the most advantageous subdivision of labor; represents, in a word, the most effective combination of the productive factors, any variation from which increases rather than decreases costs. Failure to attain this effective combination of the productive factors means high costs and perhaps bankruptcy. Naturally, the size of the most economic plant unit varies greatly in different industries. Thus, the president of the large steel firm of Jones and Laughlin stated in 1901 that the minimum cost in manufacturing steel was not secured until an output of 2,000 to 2,500 tons per day was reached, but that a mill with this large an output could manufacture as cheaply as one that made 5,000 tons per day. An independent refiner of cane sugar after considerable experimentation concluded that an output of 7,500 barrels a day was the most economic unit in this industry. If the trust chooses the plant unit of maximum efficiency, it secures the economies of large-scale production; if it fails to do so, it does not secure these economies. Inefficiency in this regard, be it noted, may more than counterbalance all the savings that the trust can effect in other ways, namely, those economies that it realizes through the fact of being a trust. In other words, the reduction in fixed charges per unit of product above referred to is an economy of large-scale production and not of combination. If there be any industry in which the most efficient plant unit is one producing the total output, there will be in that industry a strong, probably irresistible, tendency toward monopoly; 3 yet the gains of monopoly will arise, not from combination, as with the trust, but from large-scale production. In this industry as in all others there should be no interference with the economic tendency toward large-scale production, yet public policy would doubtless demand that this industry, like all other natural

1 Industrial Commission, XIII, p. 505. On the tin plate industry, see Dunbar, The Tin-Plate Industry, p. 102.

2 Hearings on the American Sugar Refining Company, 1911-1912, pp. 1151-1152.

3 The cash register industry may be one of this type. See p. 529.

monopolies, be brought either under the control or the ownership of the state.

III. Economies in Selling

The economies in selling may be discussed under the following headings: (1) advertising; (2) traveling salesmen; (3) export trade; (4) cross freights; (5) bad debts; (6) smaller stock of goods. It is in selling costs that the trust apparently finds the best opportunity for effecting economies that represent social gain.

L

(1) Advertising. The expenditure for advertising under conditions of competition is enormous. If this expenditure had the effect of stimulating the demand for commodities and thus of permitting large-scale production, it is possible that the expense of advertising might be compensated for by reduced production costs; and the cost and price of the article need not be increased on account of advertising expense. While this is sometimes the case, more commonly, no doubt, the effect of advertising is to increase the sales of one article at the expense of another. When this is the result, that is, when advertising does not increase purchases but merely turns them into different channels, the outlays for advertising must be regarded as economic waste, unless indeed the advertised article happens to be better than the one that would have been purchased had it not been for the advertising. The social costs of competition on the selling side are undoubtedly great; and their reduction is ardently to be desired.

Herein lies one of the principal advantages of the trust. The trust with the greater part of the market in its control need not expend such large sums to induce people to buy its product, since in large measure they must buy from it as the principal source of supply. Thus, the advertising expenditures of the tobacco trust declined as its monopoly control increased; and the advertising expenditures of the companies that succeeded to the trust upon its dissolution increased as compared with those of the trust. Illustrating the first point, the tobacco trust prior to 1900 controlled approximately 55 per cent of the output of

little cigars, and spent on advertising about 10 per cent of its net receipts (from little cigars) less tax; and between 1905 and 1908 it controlled approximately 85 per cent of the output, and spent only about 1 per cent of the net receipts less tax.1 In part the large apparent saving resulted from the fact that the expenditures during the period when the trust was building up its monopoly control were abnormally large,-larger than they would be under normal competitive conditions. Thus, between 1898 and 1903 the trust was striving to obtain a monopoly of the manufacture of cigars, and in the latter year it actually expended on advertising 33.4 per cent of its net receipts less tax.2 But having come to the conclusion that economic conditions did not favor monopoly in the manufacture of cigars, the trust gave up its competitive campaign, and by 1907 was spending on advertising only 4.7 per cent of its net receipts less tax.3 Illustrating the second point, the advertising expenses of the tobacco trust in 1910 in all branches except cigars were $10,895,132, while those of the successor companies (the companies that succeeded to the business of the trust upon its dissolution) ✓ amounted in 1913 to $23,623,564, or more than double. Based upon rates per thousand or per pound the advertising expenditures of the successor companies exceeded those of the trust by percentages varying from 14 per cent (Turkish cigarettes) to 155 per cent (plug cut smoking).5

These increases in cost seem to be properly laid at the door of dissolution; they represent some of the inevitable wastes of competition. The competitive system, of course, is by no means perfect; if it were none would seriously think of restraining its influence. And chief among its imperfections is the waste involved in inducing customers to buy the product of one concern rather than of another, even though the product of the latter be as good or actually better than that of the former.

Conceding that trusts may effect savings in advertising, it

1 Report of the Commissioner of Corporations on the Tobacco Industry, part III, p. 5.

2 Ibid. 3 Ibid.

4 Ibid., p. 18.

5 Ibid.