assimilation. Yet there is some evidence that a net accumulation of CO2 is occurring in the atmosphere. If this accumulation were to continue, some experts have suggested that it may give rise to significant and probably adverse weather changes by the end of this, or early in the next, century. Similarly, although the ocean has long been considered a "bottomless pit," we are not yet sure what the long-run effects of dumping presumably leak-proof containers of radioactive wastes will be; or if the basic food chain in the ocean will be significantly altered in the long-run by the discharge of waste materials such as baled solid wastes. Our knowledge of the long-run potentially irreversible impacts on ecosystems is limited. In fact, if sufficient "virgin" ecosystems are not maintained, we will not even have the base lines from which to measure changes in environmental quality. In some areas this has already become a problem, in that background levels of radioactivity, seismicity, and so on were not measured prior to human activities which affected these environmental parameters.

Impacts of environmental quality on receptors

Even if we could specify for each type of residual the time pattern of environmental quality resulting from a specified time pattern of its discharge —which we cannot-the next problem is that of delineating the effect of the resulting time pattern of quality on humans, animals, vegetation, and so on. The unknowns in this area are much larger than the knowns, particularly in the longrun. To cite one example, the evidence of the impact on human health of exposure to sulfur dioxide is inconclusive. Experiments in England which have continued for over ten years have shown no measurable impacts on health. Yet there is still a "residual" belief that such exposure may be harmful. It may be that the impacts are a function not of exposure to SO, itself, but rather to a combination of SO2, particulates, and atmospheric moisture.

We can define, reasonably well, the impacts of exposure of some plants to different levels and durations of concentration of sulfur dioxide. Thus, short-run damage functions (or loss functions), for leafy plants such as spinach, lettuce, and tobacco have been developed. The long-run damages are essentially unknown. Yet it is the long-run impact which may be most critical or some residuals, such as pesticides, where the cumulative effect throughout the food chain over time is the central problem.

Environmental quality management

The integration of the four components discussed above should be accomplished by the application of the various available techniques and methods for environmental quality management. Developing an environmental quality management program for a specific region involves three basic questions.

1. What level(s) of environmental quality are desired what portion (s) of the time? That is, should the dissolved oxygen level in the river be 3 parts per million (3 ppm), 4 ppm, or 5 ppm-and must it be at the given level 100% of the time, or only 98% or 95% of the time?

2. What is the least cost means of achieving any desired level of quality with a specified degree of certainty?

3. What kinds of management institutions can be devised which will enable achieving the desired quality levels most efficiently?

The first question cannot be answered in isolation from the second question, i.e., how much does it cost. This is because there are other demands, particularly in an urban society, on the scarce resources available-demands for schools, transportation, social welfare. Nor can it be answered without considering the third question as well, because the nature of the management institutions and the instruments used by them affect the costs.*

Because all of the benefits of improved environmental quality cannot be measured in dollar terms, we must resort, at a minimum, to defining how much it will cost to achieve different levels of quality." Then the decision must be made regarding how much increased quality is worth-in relation to the benefits foregone by devoting that amount of resources to improving environmental quality rather than to other needs. In a recent public opinion survey, about three-quarters of the respondents felt that more should be done to reduce

The residual discharge itself may be a stochastic variable, i.e., one whose characteristics can be specified only in terms of a probability distribution.

An empirical example relating to these questions is given in the Appendix.

The problem is not only the magnitude of the benefits but the distribution of the benefits among segments of the population. The distribution of costs of environmental quality management is also very relevant.

water and air pollution in the United States, but only a few percent were willing to pay $15 per capita per year toward that end.

Two aspects of the costs of environmental quality merit mention. One, the incremental costs of achieving higher levels of quality with increasing certainty rise rapidly. That is, the unit cost of removing a pound of organic waste from a liquid effluent at the 95% removal level is much higher than the unit cost at the 50% removal level.

Two, a multiplicity of ways exists for improving environmental quality, ranging from activities undertaken in the individual production units-process change, materials recovery, by-product production, treatment-to activities under collectively-reservoirs to provide water in low flow periods, collective waste treatment facilities, imposition of effluent charges or standards. Research and practice have shown sizeable economies of scale in residuals-environmental quality management, and significant efficiencies to be derived from looking at the problem in a systems context.

The importance of policy decisions with respect to the techniques used for environmental quality management merits emphasis. This can be illustrated in the following manner. A region which depends heavily on electric space heating and electrically-powered mass transit, utilizes efficient wet scrubbing of stack gases from industries and power plants, and disposes of its garbage by grinding, transporting in sewers, and discharging untreated to watercourses, would have a high degree of air quality. But there would be a heavy residuals load imposed on the watercourses of the region, with likely severely adverse consequences on water quality. Alternatively, suppose the region treats its municipal and industrial liquid waste streams to a high degree, and relies almost exclusively on incineration of sludges and solid wastes to handle the residuals from these treatments. Protection of the water and land environments would result, but at the expense of a heavy residuals load discharged to the air. If the region were to practice high-level recovery of waste materials, recycling, and by-product production, combined with the stimulation of production processes which resulted in the generation of small quantities of waste per unit, very few residuals might well be discharged into any of the environments.

Finally, knowing what the technological possibilities and economic consequences are, management institutions are needed which can utilize the range of possibilities available. The management agency needs to have authority to undertake the entire range of activities comprising environmental quality management-planning, research, data collection, construction and operation of facilities, regulation of emissions, levying withdrawal and effluent charges. The designing of governmental institutions which would perform the task of environmental quality management effectively, efficiently, and equitably, and which would be politically viable, is the toughest problem facing us. While there are gaps in our technological knowledge requiring research, pilot projects, and demonstration operations-for example, methods of analyzing management systems, technology for reducing wastes generation per unit of product, and the short-run and longrun impacts of residuals discharge-the institutional problem is the most difficult and most important."

APPENDIX

The following data relating to the questions posed on page 10 are taken from a study of the Potomac estuary by Davis.*

Table 1 shows the costs of the least cost system-some combination of reservoirs for low flow augmentation, in-stream aeration, effluent redistribution, advanced waste treatment-to achieve different levels of water quality as measured by dissolved oxygen (D.O.) concentration. The incremental costs of going to the next higher level of quality are also shown.

Institutional experimentation would appear to be the order of the day. The performance of the following agencies having some degree of responsibility for environmental quality management should be observed and evaluated-the interstate-federal Delaware River Basin Commission, Ohio Development Authority, New York State Pure Water Authority, New York City Environmental Protection Service, Miami Conservancy District (whose responsibility has recently been expanded to include water quality management), Ontario Water Resources Commission, Los Angeles County Sanitation District, Metropolitan Sanitary District of Greater Chicago. Likewise, consideration should be given to agencies which have been proposed but are not yet in operation-such as Maryland's Waste Acceptance Service and of agencies not yet proposed but hopefully are being conceived in some fertile imaginations.

Robert K. Davis, 1968. The range of choice in water management: a study of dissolved oxygen in the Potomac estuary, Johns Hopkins Press, Baltimore.

98-999-68-10

1 Costs are present worth of 50-year time stream of capital and operation, maintenance, and normal replacement (OMR) costs discounted at 4 percent.

2 Very large. * Large.

Table 2 compares the cost of the least cost system to achieve a specified level of water quality (D.O.) with the cost of a reservoir only system to accomplish the same end by increasing low flows for the dilution of wastes. These data illustrate one effect of policy decisions-in the planning process as well as in the decision process-concerning the types of quality management techniques which are included for consideration.

TABLE 2.-COMPARATIVE SYSTEM COSTS AND LEVELS OF WATER QUALITY

1 Costs are present worth of 50-year time stream of capital and OMR costs discounted at 4 percent.

Table 3 shows the impact of another type of policy decision on the cost of environmental quality management-the cost-sharing policy of the federal government. The area directly benefited by the environmental quality management system pays more or less of the direct costs, depending on the particular facilities included in the system. All systems provide the same level of water quality, i.e., 4 ppm D.O.

COST SHARING IN ALTERNATIVE WATER QUALITY MANAGEMENT SYSTEMS

1 Costs are present worth of 50-year time stream of capital and OMR costs discounted at 4 percent.

POLLUTION AND A BETTER ENVIRONMENT

(By Allen V. Kneese, Resources for the Future, Inc., Washington, D.C.)

In many ways, the performance of our economy has been impressive in this decade. The sixties have seen the United States emerge from a period of relatively slow economic growth and soul-searching about its inability to keep up with the growth rates of the Western European countries. The economy's annual growth

in GNP of about 4% per year has become one of the highest in the developed world, and man-hour productivity growth of about 6% per year is well above the post-war trend. Productivity gain has been particularly impressive in the manufacturing and agricultural sectors, and prices of goods have been falling relative to services, thus making food, energy, and material goods more and more copiously available to the population. These developments would seem to be good reasons for euphoria and in many ways they certainly are.

There is a dark lining on this silver cloud, however. The vast increase in manufacturing activity and output and the rapidly rising conversion of fossil fuels to energy has imposed an ever-increasing burden of waste residuals on the environment. In fact, the total weight of residuals discharged to the environment tends to rise pari passu with the increase in manufacturing activity and energy conversion unless there are gains in the technical efficiency of converting inputs to useful outputs or the recycle of used goods is increased.

Throughout most of our history, the discharge of residuals to air, water, and the land was of concern only in particular and unusual instances, if at all. Granted some of these instances were spectacular, such as the smoke in Pittsburgh early in this century. But overall, we were endowed with immense space and vast flows of water which could dilute and assimilate residuals with little damage to the natural environment. This was fortunate because the air and water are "common property" resources with respect to which the private market, on which we have relied so heavily to allocate resources to their most valuable uses, cannot function. In recent years, the naturally available assimilative capacity of the natural environment has been rapidly used up and it is becoming more difficult to protect one environment medium, such as water, without damaging another, such as air. In the past, when pollution control efforts were undertaken, it was often assumed that if a liquid or gaseous waste stream was treated, or solid wastes were burned or hauled off the premises, the pollution problem was solved. In recent years we have gradually come to appreciate that air, water, and solid waste problems are closely interdependent and their analysis and control is best viewed as a systems problem relating to the whole process of residuals generation and control.

The overall residuals problem

To clarify and illustrate these points, it is useful to view the residuals problem initially as a materials balance problem for the entire economy. A highly simplified schematic of how the goods and residuals production process works is shown in Chart I (p. 144). The inputs to the system are fuels, foods, and raw materials which are partly converted into final goods and partly become waste residuals. Except for increases in inventory, final goods also ultimately enter the waste stream. Thus, goods which are "consumed" really only render certain services. Their material substance remains in existence and must either be reused or discharged to the ambient environment.

In an economy which is closed (no imports or exports) and where there is no net accumulation of stocks (plant, equipment, inventories, consumer durables, or residential buildings), the amount of residuals which is inserted into the natural environment must be approximately equal to the weight of basic fuels, food, and raw materials entering the processing and production system, plus oxygen taken from the atmosphere. This result, while obvious upon reflection, leads to the, at first rather surprising, corollary that residuals disposal-in terms of sheer tonnage is an even larger operation than basic materials production. In an open (regional or national) economy it would be necessary to add flows representing imports and exports. Similarly, in an economy undergoing stock or capital accumulation the production of residuals in any given year would be less by that amount than the basic inputs. In the entire United States, economy accumulation accounts for about 10-15 per cent of basic annual inputs,' and there is some net importation of raw and partially processed materials amounting to 4 or 5 per cent of domestic production. Table 1 shows estimates of the weight of raw material produced in the United States in several recent years, plus net imports of raw and partially processed materials.

1 Mostly in the form of construction materials.