the public at the time these investments were made. It is probably correct to say that no more definite understanding could have been implied than that the service would be supplied at the cost of production. Cost of production here means the actual cost of producing the service, including interest and normal profit, but excluding monopoly gains. Whether interest and normal profit were to be based on actual cost or on cost of reproduction was probably seldom considered, and there has certainly been no authoritative statement that could justify a conclusion that either the one or the other method would prevail.

If it were generally true that public utility properties could now be reproduced at less than actual cost, the argument for the acceptance of actual cost as a normal standard for fair value would appeal very strongly to the public utility interests. As, however, prices of land, labor, and material have in general advanced enormously since 1896, most utility enterprises can only be reproduced to-day at a cost considerably in excess of the actual necessary cost. It is natural, therefore, that public utility interests should incline strongly toward the reproduction method.

It may be argued that it has, at least for a considerable number of years, been recognized by the highest courts that the company is entitled to earn a fair return on the fair value of its property. The courts have used the terms "value," "present value," "fair value," "reasonable value," and "just value" in relation to the amount upon which the fair rate of return should be based. That they have not usually had current exchange value in mind is clearly apparent from the fact that the chief weight in determining value has been given to cost, either actual cost or reproduction cost. They have usually used "value" in the sense of "normal value," as that term is understood by economists, i. e., a value corresponding to the cost of production. It seems at first thought that reproduction cost corresponds more nearly to any proper use of the term "value" than does actual cost. It does have a closer relation to exchange value, but not necessarily a closer relation to normal value or cost of production, which is the only sense in which the term "value" can be properly used in this connection.

It may be argued, nevertheless, that constant use of the term "value" in this connection has, whether rightly or not, produced the impression in the minds of investors and others that utilities

would be allowed to earn upon a value represented by their reproduction cost. This argument seems particularly forceful in regard to land. It is argued that as the company owns land and could, if it so desired, dispose of it and pocket the proceeds, it amounts to confiscation not to permit it to earn on the present market value of the land. This statement needs qualification. In the first place, most land that is used for public utility purposes is improved to such an extent that if sold for other uses the loss due to the scrapping of the improvements would offset the increase of selling price over cost price. If a railroad right of way were sold for farm purposes, the loss due to the scrapping of the roadbed would more than offset any increment in the selling price of the land. In case the company desired to go out of business it could, of course, sell its entire property to any purchaser, but such purchaser would necessarily base his purchase price on probable earnings under the existing scheme of regulation. If, however, the company decided to abandon its franchise, scrap its plant, and sell its land for what it would bring for other uses, it could of course realize on the increment in the value of the land, but such increment would be far from sufficient to offset the loss in the value of the plant and equipment as based on actual cost.

But the determination of fair value is only one step in the process of determining a reasonable rate of charge. We have already found that a reasonable rate of charge for an appropriate and normally successful utility enterprise is the equivalent of the normal cost of production. The normal cost of production is the amount which in the long run it is necessary to pay to secure the utilities demanded by the public. It is the amount that will secure an equilibrium between demand and supply. Increments and profits of every kind enjoyed by the company must necessarily be considered a part of the total compensation that the company receives from the public. In so far as there are increments and profits arising from increase in land values, it is clear that such increments and profits should in a rate proceeding be considered either as income or as an offset in fixing the rate of return. Any apparent advantage secured by the company under the reproduction method is in a measure counterbalanced. The reproduction method as thus applied is therefore a roundabout method of accomplishing what can be more simply and effectively accomplished by basing the rate of

return on the actual capital cost. This does away with the necessity of accounting for appreciation or reducing the nominal rate of return.

Most of the objections to the adoption of any standard of value arise from a consideration of the numerous difficulties and complexities created by the case of poorly located and unsuccessful enterprises. It is apparent that in such cases rates must be fixed without much regard to cost of production. Rates in such cases must be largely based on a fair judgment of the value of the service to the consumer. In such cases the existing property has little or nothing to do with the estimate. The entire problem is approached from the standpoint of what is fair to the consumer. It is recognized that a rate which would fairly compensate the company from the standpoint of its outlay is improper, as such rate would be higher than the consumer could justly be called upon to pay. Such cases are not, properly speaking, cases for valuation at all. They are cases in which the rates must be determined largely without regard to what is normally meant by value for rate purposes. It is true that no standard of rate-making can be adopted for such cases. The usual procedure is to determine actual cost, reproduction cost, accrued depreciation, capitalization, and perhaps other factors, and then to fix an amount as fair value which at the rate of return determined upon will permit of the rate which seems to correspond to the fair value of the service. The reasoning here is somewhat circular. The fair rate of charge must be first determined and then a fair value and fair rate of return that will seem to justify the rate of charge already determined upon. In considering a normal standard of fair value for rate purposes the abnormal case of the poorly located, unsuccessful, or partially obsolete enterprise should be excluded. Such abnormal enterprises must necessarily be put in a class by themselves in considering rate and valuation problems. If the discussion is limited to the normally located and successful enterprise it is clear that a controlling standard for the determination of fair value can and should be developed. It is essential that the relations between the company and the public be put on a more permanent and dependable basis. In justice both to the company and to the public the determination of this important matter should not be left wholly to the judgment or predilection of the court in each particular case.

Most judges and individuals as well, whether consciously or not, use some standard as actually controlling. They may test their judgment of reproduction cost by actual cost, or they may test their judgment as to normal actual cost by reproduction cost, but in either event it is one or the other of these standards that is actually controlling.

The determination of a standard of value applicable to existing investments will be worked out if at all by the slow and piecemeal process of court decision in numerous cases. The final answer can only be given by the Supreme Court of the United States. It would seem, however, that as to the future, legislative bodies and commissions might at once adopt a standard. This standard would apply to future investments and to future fluctuations in existing investments. If normal actual capital cost were adopted as the rule for the future, accounting methods and rate regulation would be much simplified and the relations between the utilities and the public placed on a much more equitable and dependable basis. The adoption of this as the normal standard for appropriately located and successful enterprises would not mean that exceptional efficiency shown in the construction or operation of an enterprise could not be properly rewarded, or, on the other hand, that exceptional inefficiency could not be penalized. Such reward or penalty, however, is more properly reflected in the rate of return allowed.

NEW YORK CITY.

Robert H. Whitten.

THE CLASSIFICATION OF TRUSTS AS EXPRESS, RESULTING, AND CONSTRUCTIVE.

IN

N view of the provisions of the Statute of Frauds, the common classification of trusts according to their mode of creation "into express trusts, implied trusts, resulting trusts, and constructive trusts," 1 and the proper definition of the terms used in that classification, are of interest and of value. An analytical examination of this particular classification of trusts leads to somewhat surprising conclusions and has its practical bearing on the applied law of trusts.

A conventional statement of the matter and a common arrangement of the terms were given by Maitland as follows:

"Trusts are created (1) by the act of a party, (2) by the operation of law. I do not think that these terms are unexceptionable, still they are well known and useful. A further classification has been made:

Before scrutinizing closely Maitland's analysis, however, we must get some definitions of the terms, or at least must discuss details about the terms.

A. THE NATURE OF EXPRESS TRUSTS, IMPLIED TRUSTS, RESULTING TRUSTS, AND CONSTRUCTIVE TRUSTS.

I.

Express and Implied Trusts.

Express trusts are, of course, trusts stated fully in language, when the language used is properly construed. There are (a) oral