Federal Natural Resources Development: Basic Issues in Benefit and Cost Measurement JACK L. KNETSCH The George Washington University ROBERT H. HAVEMAN Grinnell College (On leave with the Joint Economic CHARLES W. HOWE Resources for the Future, Inc. JOHN V. KRUTILLA Resources for the Future, Inc. MICHAEL F. BREWER Resources for the Future, Inc. MAY 1969 Natural Resources Policy Center For well over one hundred and fifty years now, the federal government has undertaken expenditures to develop the nation's natural resources. Federal spending in these areas currently totals over $3.5 billion per year and accounts for some of the nation's largest physical structures. Water resource installations have productive potentials similar to those of industrial investments generally. These projects absorb inputs and produce outputs. Among the primary "products" which these installations produce are irrigation water, the reduction of flood hazards, the provision of transportation services, hydroelectric energy, and the provision of municipal water supply. The inputs which they use are similar to those used in common industrial enterprises: skilled and unskilled labor, steel, cement, bulldozers, and complex electrical generating equipment. The fact that these projects absorb valuable inputs and produce valuable outputs provides the opportunity of measuring the benefits from such government undertakings as well as the costs which they entail. The comparison of the resulting benefits and costs is necessary if prudent public expenditure policy is to prevail. Interest in procedures to determine the contributions of natural resource projects to the nation's economic welfare is not a new one. Indeed, explicit efforts to account for the expected benefits and costs of projects have been made for over 30 years. The basic criterion for determining the worth of proposed projects was formally outlined in the Flood Control Act of 1936, which stated that the federal government was prepared to undertake such investments "if the benefits to whomsoever they accrue exceed the costs." This criterion still guides evaluation efforts in the natural resources develop ment area. Both the Senate Committee on Interior and Insular Affairs and the Senate Public Works Committee have stated that: "The economic analyses of projects should reflect the broadest scope of potential benefits and costs," and that project evaluations "should accurately reflect all primary direct and indirect benefits as well as secondary benefits." We share this concern. It is our special concern, however, that current deliberations to broaden the concept of project benefits not take place in isolation from some basic principles pertaining to appropriate economic concepts of social benefits and social costs. In both the private sector and the public sector, decision-makers who strive to develop good policy for their organizations evaluate uses of funds to insure that the expected returns exceed the costs. While the basic notion of benefit and cost evaluation is similar for both the private and public sectors, there is one basic difference. When decision-makers in the private sector, say in a private business, evaluate the benefits of investing in a new production facility and compare those benefits with associated costs, they are concerned only with the gains and losses which accrue to their firm. Because any other gains and losses which may accrue to outside parties do not show up in the revenues and costs of their firm, they are typically ignored in private investment evaluation. On the other hand, a responsible decision-maker in the public sector cannot adopt so restricted a view. He must conceive of his investment project in a more comprehensive way so that all of the costs and gains associated with the undertaking are accounted for in the investment decision whether or not all appear as receipts of, or disbursements by, his particular agency. Indeed, it is just because market-governed private organizations cannot charge for third-party benefits or be held liable for third-party costs that the public sector must undertake so many resource development activities. But even when some agency of the public sector undertakes a resource development or management project, it is not clear that all impacts on third parties will be accounted for. First it is necessary that the "project" be defined so that all physical impacts are included in the scope of the "project." For example, a hydro power project must be defined to include all downstream effects of the dam's storage and release cycle, as well as effects on the quality of water in the reservoir and downstream from the project. An irrigation project must be defined to include the downstream effects of quantity and quality diminution as well as all drainage facilities which will be needed to keep the project in operation-even if they will not be needed for some period following initial project construction. The second difficulty which can stand in the way of a complete evaluation of a project's impacts even when undertaken by a public body is a jurisdictional one, namely the "accounting stance" or geographical scope of concern assumed by the decision-making body. The "accounting stance" may or may not incorporate all of the effects of the project (properly defined). As an example, downstream impacts of waste treatment may lie entirely outside the jurisdiction of the city responsible for the treatment plant. If a project induces significant changes in market prices, regions quite remote from and possibly even non-contiguous with the project site may incur significant benefits or costs, especially if there are problems of chronic unemployment or immobility of human and capital resources. If, for example, a proposed Bureau of Reclamation irrigation project located in Nevada is evaluated from the state's point of view, it is likely to have an enormous benefit-cost ratio. Clearly, most of the benefits from the project in the form of increased farm incomes will accrue to residents of Nevada. Because of existing repayment (pricing) policies for federal irrigation projects, much of the costs are federal government costs and only a small proportion of them will fall on people living in Nevada. If the increased agricultural production in Nevada lowers farm product prices and displaces agricultural output in other regions with subsequent temporary or long term income losses (a real national cost), the vast majority of the losers are likely to be in other parts of the country. Thus, if evaluation were made from a broader perspective, the benefit-cost ratio obviously would be reduced. Evaluated from this broader view, the project might or might not be in the national interest. When the project is evaluated from the national point of view, gains to all the beneficiaries and losses sustained by all the cost-bearers will be included in the calculation. It is, therefore, not necessarily true that what is good for Region X is good for the United States. One elemental principle for benefit-cost measurement of federal natural resource development expenditures derives immediately from this discussion of accounting stance. By definition, federal natural resources development is sponsored and financed by the national government, which represents all of the people of the United States. For this reason, we conclude that: I. Federally-financed investments require a national While citing this as a basic principle, we are not arguing that analyses of the benefits and costs accruing to more localized regions should not be estimated. We are arguing, however, that federal natural resource agencies have a primary obligation to weigh the costs and benefits of projects from a national point of view and a significantly smaller responsibility to weigh the impacts of projects on regional economies where such local impacts are in fact offset elsewhere in the economy. While this definition of an appropriate federal accounting stance is a primary principle in establishing a correct benefit-cost criterion, it does not provide many clues to the accurate measurement of these national benefits and costs. To address this measurement problem, we require some notion of the basic underlying characteristics of the economy. In large measure, the U.S. economy relies on the freely-arrivedat decisions of consumers and producers in getting private goods and services produced. Under these market arrangements, the economy has repeatedly shown a fluid response to changed conditions and has demonstrated an enormous potential for growth. Changing demands elicit a changed pattern of supply with little time lag, without the need for some central authority first to recognize the existence of the new demands, and then convey them to the produc |