3. ALLOCATION OF CONSTRUCTION COSTS TO VARIOUS FEATURES

Since Federal multiple-purpose projects have so many different features, it is considered necessary by various Federal agencies to allocate a project's construction costs to the various features in order to determine the value relationships as between features and the estimated costs of each feature. This allocation of construction costs among various features is assumed to be important; for, by doing so, it is thought possible to determine the costs for each portion of the project and to isolate those costs which must be met from various sources of funds.

After allocating construction costs to the various features, the specific and joint costs allocable to each of such features are then broken down into two major groupings, reimbursable costs and nonreimbursable costs. The reimbursable costs of construction are primarily those costs allocable to power and a portion of the costs allocable to irrigation. Costs allocable to the other features are considered to be nonreimbursable.

The methods used by various construction agencies in computing the allocations of construction costs vary considerably. Among the theories which have been advanced are the following:1

a. Allocation in proportion to benefits.

b. Alternative expenditure method.
c. Allocations based upon vendibility.

d. Allocations based upon assessability.

e. Proportionate use of capacity.
f. Separate projects method.
g. Residual value method.
h. Equal distribution method.

The sharp divergencies inherent in different methods of allocating construction costs can be illustrated by the following extracts from the 1939 Hearings on the Tennessee Valley Authority (Report of the Joint Committee Investigating the Tennessee Valley Authority, S. Doc. 56, pt. 1, April 1939, p. 154):

The testimony given before the committee revealed a wide divergence of opinion on allocation theories, which may be summarized below:

The nonallocation theory was advanced by Charles W. Kellogg, president of the Edison Electric Institute. Mr. Kellogg stated that in the past the Federal Government has forced power companies constructing dams in navigable streams to bear the total cost. Regardless of the facts as to the Authority's program or any future multiple-purpose program, he claimed that the precedent of the past should be controlling and that the total investment should be charged to power. The alternative justifiable expenditure theory formed the basis for the allocation reports of the Authority. The difference resulting from a single construction NOTE.-Numbered references 1, 2, and 3 appear in the General Notes, page 54.

for all purposes, compared with a series of separate constructions, each for one purpose, is apportioned to each function.

The incremental power investment theory was advanced by Dean Moreland in his testimony as a witness for the Commonwealth & Southern Corp. Under this theory, power is charged with the total investment balance after deducting the estimated cost of the most economical alternative plan for accomplishing the other purposes.

In a system devoted to several different purposes, such as the Authority, the cost assigned to each purpose must necessarily be to a large degree a matter of opinion. The problem is similar to that which would confront an architect if he were asked to allocate the cost of a house among its various uses. The best allocation is that which appears to be the most reasonable in terms of any proposed alternative. The difficulties of allocation, which cannot be based on exact scientific considerations, have created confusion and lack of uniformity among engineering and accounting experts, in both public and private fields. This problem is particularly aggravated in the Authority's operations by the magnitude of the common costs.

Proponents of divergent theories, similar to those shown above in the case of TVA, have argued vehemently over these theories for years; and the controversy remains unsolved. The reason for their vehemence can be explained by reverting to the discussion of reimbursable and nonreimbursable costs.

Reimbursable construction costs are those on which the Government anticipates repayment of the capital it provided (frequently with interest) from the revenues of the project. Nonreimbursable construction costs, on the other hand, are costs attributable to those features which produce no revenue and, hence, on which no repayment of the Government's investment is even contemplated. This means that, if the allocations of construction costs to nonreimbursable features like flood control are heavy, the reimbursable features like power and irrigation need not produce such large revenues in order to warrant construction. A corollary of this is the fact that it is sometimes possible to sell electrical energy or water at lower prices, if the allocations to nonreimbursable costs are heavy.

Given the large number of alternative methods of allocating construction costs, confusion and inconsistencies have been inevitable. Among the criticisms which can be made of the present situation are: a. Allocations comparisons between various projects are difficult to make.

b. Material presented to Congress regarding allocations is frequently either so inadequate or so technical that judgments are seriously hampered.

c. The extent to which a project should be required to carry its weight financially is influenced greatly by an arbitrary decision of the construction agency on cost allocations.

To put it more simply: Allocations ultimately determine the amount which will be repaid on the Government's investment. Unless they are reached by a clear, fair and standardized method, Congress and the public will not have adequate information to permit intelligent evaluation of a proposed project. Unless cost allocations are made properly, it is virtually impossible to determine the true economic justification for a project.

B. The Present Concept of Economic Justification and the Cost-Benefit Relationship

1. GENERAL

Given the diverse objectives of multiple-purpose projects, it is admittedly difficult to determine whether a specific project is feasible or justifiable from the standpoint of the national economy. Therefore, it has proven necessary for the construction agencies to devise a method whereby the feasibility of a project can be measured. In order to do so, it has become customary, first, to estimate the average annual benefits to which the project would give rise. Against these benefits are matched the average annual costs of the project; and, if such benefits exceed such costs, a determination is made to the effect that the project is economically feasible or justifiable.

This procedure is susceptible of widespread misinterpretation, because the average annual benefits do not represent cash income and are not available to meet the average annual costs.

Only power and irrigation (and including the sale of municipal water) are productive of substantial cash income to meet costs and are considered to be reimbursable, and even the costs to be met from these two sources depend on the arbitrary and extremely variable allocation of construction costs described in the previous section of this chapter. If less than a fair proportion of the construction costs is allocated to power and irrigation, then interest and amortization of the power and irrigation investment is unfairly low. In addition, the revenues anticipated to be received from the sale of water for irrigation have been frequently overestimated, so that much of the irrigation revenues may fail to be developed into cash income. The most important problems in attempts to determine the economic justification of a project are: The inexactness of the present method of comparing theoretical benefits, rather than revenues, with estimated costs, and the wide variations inherent in the allocation of construction costs, both throw serious doubt upon the validity of the economic justification of many multiple-purpose projects.

2. CURRENT PROCEDURES FOLLOWED BY CONSTRUCTION AGENCIES

Under present procedures, the department or agency charged with originating multiple-purpose projects prepares the original detailed economic justification for each such project. In some cases, the initial motivation for the project may have arisen within the executive department or agency which requests Congress for study money or for authorization of the project; and in other cases, Congress itself may provide the initial motivation by voting study money and referring the project to an executive department or agency. Later, a statement regarding the economic justification of the project, frequently accompanied by a formal certification of authorization under the Reclamation Act or other legislation, is prepared by the originating department or agency; but the final judgment as to its economic feasibiltiy is that of Congress.2 Historically there has been no independent engineering or economic group in the Executive Branch with final authority to analyze and evalute the material so submitted by the originating executive agency, although sporadic attempts have been made to create such authority.

3. AVERAGE ANNUAL BENEFITS

The first step in the determination of the economic justification of a multiple-purpose project is the computation of the average annual benefits. In making this computation, there appears to be some lack of uniformity among Federal agencies, but, in general, the average annual benefits are hypothetical advantages expected to be derived either locally or by the citizens of the Nation as a whole from the project. They are expressed in dollars and are considered to be representative of the annual increment to the Nation's wealth arising by reason of the project.

Speaking broadly, the average annual benefits may be computed as follows:

a. Power benefit is computed by various methods. Sometimes it represents the anticipated revenue from the sale of the power at the "bus bar" at competitive rates. In the case of the Missouri Basin it is computed by estimating the energy to be generated which, stated in kilowatt hours, is then multiplied by 41⁄2 mills.

b. Irrigation benefit represents the increase in the estimated value of the production of the land to be irrigated.

c. Flood control benefit represents the savings in property values attributable to the cessation or diminution of floods.

d. Navigation benefit is the estimated savings to the national economy from water transportation when compared with the cost of transportation of equivalent quantities at higher rail or road rates.

e. Recreation benefit is computed by estimating such items as additional license fees, the value of new vacation sites, etc.

f. Fish and wildlife benefit is the increase in the number and value of fish and wildlife in the land and water affected by the project.

g. Malaria control, silt control, and salinity control benefits are computed by estimating the increase in national wealth attributable thereto.

Under this procedure, a statement of the average annual benefits is prepared, generally in detail, as a basic consideration in the computation of the economic justification of each project.

There is a lack of uniformity in the principles followed in computing benefits. An example of this lack of uniformity can be found in the case of irrigation benefits. In some cases, the benefits arising from irrigation are based upon the estimated increase in the gross crop income which it is anticipated will be obtained by the furnishing of the new irrigation supply to the area serviced (e. g., Colorado Basin development); and, on the other hand, the benefits from irrigation may be estimated on the basis of the net crop income (e. g., Gila Federal Reclamation project).

4. AVERAGE ANNUAL COSTS

Having computed the average annual benefits, the department or agency then estimates the average annual costs of the project after full development is reached. This is the second of the two fundamental steps involved in determining the economic justification of the project, and it is done in order to determine whether this estimated annual cost is less than the average annual benefits expected to result from the operation of the project.

In computing the estimated annual costs, there again appears to be a lack of uniformity in the practices of the executive departments and agencies, and even among the field offices of some of them. This lack of uniformity has given rise to serious controversy in some areas, including, for example, the two following matters:

a. In the computation of operating costs, resplacement accounting is used in most cases by the Bureau of Reclamation and by the Corps of Engineers as well as by the Federal Power Commission, but estimates of replacement provisions differ among the above. The method used by the Bureau of Reclamation is not described in the published reports studied. The Corps of Engineers uses a sinking-fund method based upon a 3 percent interest rate, while the Federal power Commission uses a sinking-fund method with a 21⁄2 percent interest rate. It is interesting to note that the Bonneville Power Administration uses the replacement-accounting method in its repayment studies but uses depreciation accounting in its annual reports of revenues and

expenses.

b. In the computation of other charges, there is wide diversity in the treatment of interest on the Government's investment, repayment of that investment, and payments in lieu of taxes.