New Guidelines for Water

In June 1969 a special task force of the Water Resources Council released its Procedures for Evaluation of Water and Related Land Resource Projects. This document, which may cast a long shadow, can be appraised only against an extensive background. For more than three decades federal proposals in this area have been subject to explicit economic appraisal of one kind or another.

The first such provision, the Flood Control Act of 1936, specified that in federal flood control projects benefits should "exceed the costs, to whomsoever they accrue." In 1950 the need for more inclusive guidelines to cover all federal water programs led the Federal Inter-Agency River Basin Committee to issue Proposed Practices for Economic Analysis of River Basin Projects, or, as it came to be known, the "Green Book." Though offered for the guidance of the participating agencies (Agriculture, Army, Commerce, Interior, Labor, and the Federal Power Commission), the report was not mandatory. The agencies were not bound by the proposed practices. This document, which was revised in 1958, contained detailed definitions, methodologies, and examples of application to program and project formulation, all reflecting the objective of maximizing the national income. It was an outstanding achievement for its time.

A second draft set of criteria came into existence in December 1952, when the director of the Bureau of the Budget, in Circular A-47, set forth the standards that would be used by the Executive Office of the President in reviewing agency reports and budget estimates. This circular covered some of the same technical benefit-cost issues contained in the "Green Book," but it went into additional matters of policy. Its revision, issued in 1955, enunciated the Eisenhower "partnership policy" relating especially to power development. For seven years A-47 remained the only binding set of guidelines for economic assessment of federal water projects.

President Kennedy in October 1961 requested a review of existing principles, standards, and procedures for economic evaluation, along with recommendations on appropriate federal policy. The resultant Policies, Standards, and Procedures in the Formulation, Evaluation, and Review of Plans for Use and Development of Water and Related Land Resources was approved by the President in May 1962.

Known as Senate Document 97, it was much less detailed than the earlier documents in both technical and policy content. Neither it nor any of the earlier guidelines set forth any uniform policy on cost-sharing or reimbursement by beneficiaries of government projects. On this vital and controversial matter the various agencies had all evolved their own disparate rules. However, the President's letter approving the new guidelines took note of the need for "up-to-date policies, standards, and procedures relating to cost allocation, reimbursement, and sharing. . . ."

The Water Resources Planning Act of 1965 established the Water Resources Council as a permanent coordinating and overall guiding agency for federal water policy. It was empowered, subject to presidential approval, to set the standards and procedures to be followed by federal agencies in planning and evaluating water developments. The council's first major policy recommendation came in response to the rise in interest rates which began in late 1965. From that time it became increasingly clear that the discount rate of three and one-fourth percent then being used was inappropriate. The council's recommendation that the discount rate be related to current government bond yields was approved in December 1968, and the applicable discount rate was raised to four and three-eighths percent. It was further raised to four and five-eighths percent in fiscal year 1970. This increase naturally raised estimated project costs and thus had the effect of "cutting off" many proposals which had formerly exhibited a benefit-to-cost ratio greater than 1.0.

Meantime public anxiety had been increasing concerning those effects of federal investment programs that do not lend themselves to measurement in dollars. The growth of programs whose justification was primarily "noneconomic" (health, education, poverty, foreign aid) served to emphasize that all government programs have some impact on the environment, on the relative growth of regions, on the distribution of wealth among persons, or on other aspects of human welfare that do not fit into the established framework of quantitative economic evaluation. Dissatisfaction with the slighting of these values led to a growing number of legal actions and citizen protests to enjoin public agency projects.

Raising the discount rate, with the consequent dropping of many prospective federal water projects, added to the pressures for a more comprehensive view of water development benefits. In response the council appointed a special task force to develop guidelines for benefit and cost measurement and interpretation which would "insure that all tangible benefits and costs are estimated and their full value included . . . and that both intangible benefits and costs are similarly included."

The covering letter issued with the June 1969 report of the task force recommended that national and regional hearings be held on the document (these were completed in September), that the proposed procedures be tested on representative projects by various government and university teams of investigators (this work is now in progress), and that an enlarged task force be called to resolve remaining problems before acceptance of the document as the new set of evaluation standards and procedures.

The intent of the new evaluation procedures clearly is to broaden the criteria by which federal water-related projects are designed and selected for funding. The report seeks to introduce "multiple objective planning" by adding to the established national economic efficiency criterion three new objectives: (1) regional income, (2) environmental quality, and (3) human well-being. Four "accounts" are set up to correspond with the four objectives. On the debit and credit sides of each account are entered the benefits and costs relating to that objective, stemming directly or indirectly from the project. The benefits and costs may be measurable in dollars or they may be purely descriptive (e.g., a life saved, a more beautiful river, etc.). Benefit-cost ratios can be computed for the national and presumably the regional income accounts, but no simple summary measure can be applied to the other accounts. No attempt is made to add or otherwise aggregate the balances of the four accounts. Thus Congress would be presented with reports on four major dimensions of each project and would have to judge the appropriate trade-offs among objectives in making its decision.

The basic idea of extending the evaluation process to new dimensions of a project's impact is a long-needed innovation. Modern America clearly values objectives other than economic efficiency, and the more light that can be shed on nonmarket impacts, the more faithfully will the decision-making process reflect the broader social point of view. Properly developed, the new approach can lead toward that goal.

Implementation, however, will require further development of the proposed guidelines, especially through basic research and experimentation; expansion of criteria to include regional income, environmental quality, and other aspects of human well-being calls for great extensions of economic and theory and measurement. Naive attempts to apply this much more complex approach could result in misapprehensions that could negate the objectives of the guidelines.

In its present form the task for report appears to have four major shortcomings. The first and most serious is the proposal to remove from the national income account all economic costs that are judged to be attributable to the attainment of the regional, environmental, and human well-being objectives. It would be of interest to know which of the separable costs are attributable to these objectives. But this could be done without removing them from the national income account so as not to render that account completely meaningless while producing only a bogus benefit-cost ratio.

Second, throughout the report, benefits are emphasized and costs underplayed. This bias is found in the discussions of secondary benefits and all types of nonmarket pacts. When conditions are conducive to secondary or nonmarket benefits they are also likely to bring about secondary or nonmarket costs. An example might be the secondary benefits forthcoming from the employment of otherwise unemployed labor in industries indirectly relate to a water project. If this occurs, the financing of the project or the sale of its output may well displace employment in other regions.

Third, the document recommends that in the attempt to measure the impact of projects upon regional income, the analysis be confined to the project region itself. But that area would probably experience net benefits no matter how bad the project might be from a national viewpoint. The analysis would then omit any investigation of offsetting dis-benefits to other regions. An important historical example has been the displacement of southern agriculture by the expansion of western irrigated acreage.

Finally, as in the past, there is a total absence of recommendations on reimbursement policy, i.e. what charges should be made against beneficiaries for project outputs. The regional benefits mentioned in the paragraph above cannot be ascertained unless one knows how much of the cost is borne by beneficiaries in the region and how much by taxpayers throughout the nation. Of much wider importance is the fact that reimbursement policies of the different federal water agencies are not consistent. This leads potential beneficiaries to "shop around" among agencies to get a particular type of project at least local cost, e.g., for flood control. Present policies often lead to pressures for undertaking projects which do not represent the least-cost method of achieving the goal sought. For example, low-flow augmentation through a Corps of Engineers project a is nonreimbursable, while sewage treatment plants must be partially financed by the locality. If a water quality problem can be solved by either method, local interests are likely to press for reservoir construction even if it is the more costly alternative from the national viewpoint.

Since the new guidelines presented by the task force have not been finalized, some of the weaknesses noted above may be eliminated or mitigated in later revisions. The findings of the federal agency and university teams now in the field will be useful in uncovering other points that need research.