Cover Image

PAPERBACK
$125.25



View/Hide Left Panel
Click for next page ( 292


The National Academies | 500 Fifth St. N.W. | Washington, D.C. 20001
Copyright © National Academy of Sciences. All rights reserved.
Terms of Use and Privacy Statement



Below are the first 10 and last 10 pages of uncorrected machine-read text (when available) of this chapter, followed by the top 30 algorithmically extracted key phrases from the chapter as a whole.
Intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text on the opening pages of each chapter. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.

Do not use for reproduction, copying, pasting, or reading; exclusively for search engines.

OCR for page 291
ECONOMIC CONSIDERATIONS _ MANAGING CONTAMINATED MARINE SEDIMENTS Thomas A. Grigalunas and James J. Opaluch University of Rhode Island ABSTRACT Contaminated marine sediments pose highly uncertain but potentially serious threats to public health and the environ- ment. However, cleanup of these sites is very expensive and costs increase rapidly with level of effort. Thus, impor- tant tradeoffs are faced in the social decision concerning appropriate cleanup level. This paper discusses the applica- tion of economic analysis as input to the social decision process for managing contaminated marine sediments. Two general approaches are outlined: cost-benefit and cost- effectiveness analysis. Both approaches can provide valu- able input into remedial action decisions at a site and in allocating efforts among multiple sites (fund balancing). However, significant diffi culties and uncertainties charac- terize all approaches for managing contaminated marine sedi- ments and economics is no exception. The difficulties and potential for application of the two economic approaches are discussed, along with the potential role of strict liability for damages in providing incentives for source control to avoid creation of new sites. INTRODUCTION Contaminated sediments occur in marine coastal areas throughout the United States and in the Great Lakes (A. D. Little, 1987; U.S. Environ- mental Protection Agency [EPA], 1985; Office of Technology Assessment [OTA], 1987; U.S. Dept. of Commerce [DOC], 1988~. This contamination stems from a variety of point and nonpoint sources including day-to-day releases from industry, sewerage treatment plants, urban runoff, riv- ers, federal facilities, shoreline erosion, atmospheric sources, and periodic spills from vessels, pipelines, and shoreside facilities. The substances contained in sediments include heavy metals, synthetic or- ganic compounds, petroleum hydrocarbons, and other materials. Concern with contaminated sediments stems from the threat they pose to public health and the environment. At sufficient concentrations, toxic substances give rise to health threats to individuals exposed either directly through contact with contaminated materials or much more likely, indirectly via the food web. Risks to health from con- sumption of contaminated shellfish or finfish have caused public offi- cials to close fishing grounds or restrict the catch of certain 291

OCR for page 291
292 species, thereby imposing economic losses on commercial and recrea- tional users of the affected species (e.g., Freeman, 1987; OTA, 1987 A.D. Little, 1987~. Indeed, even the perception of a possible but uncertain threat to health from consuming fish exposed to toxic sub- stances can impose losses (Schwartz and Strand, 19811. Other losses also can result from contaminated marine sediments. Direct lethal effects on adult fish and shellfish, juveniles, eggs and larvae can cause short- and long-term commercial and recreational fish- ery losses (Grigalunas et al., 1987, 1988~. Indirect or ecological effects can lead to losses of commercial and recreational fisheries, waterfowl, and other marine resources through loss of habitat (Kahn and Kemp, 1985) and via the food web (Grigalunas et al., 1987, 1988~. Con- cern about exposure to toxic materials also may impose losses on recrea- tional beach users, and a reduction in property values can occur as a consequence of a loss in amenity services at or near a contaminated site (e.g., Freeman, 1987~. Although the presence of contaminated sediments can impose a var- iety of losses, remedial actions typically are very costly and in many cases the cost increases rapidly as additional levels of remediation or treatment are sought. For example, remedial actions described or pro- posed for New Bedford Harbor (EBASCO Services, Inc., 1987~; the Hudson River (Mark Brown, New York State Department of Environmental Conserva- tion, personal communication); and Commencement Bay (Lukjanowicz et al., 1988) could cost millions of dollars. An analysis of alternative levels of cleanup of Hudson River polychlorinated biphenyls (PCBs) show rapidly increasing unit costs with increasing levels of cleanup (Na- tional Research Council [NRC], 1979~. Hence, additional degrees of public health protection and environmental benefits can be achieved-- but typically only at far greater cost. Given limited funds to use in cleanup of the rapidly increasing number of contaminated sites through- out the country, higher expenditures on cleanup at one site implies smaller remaining budget for cleanup at other sites. Thus, important tradeoffs are faced in determining the level of cleanup that should be carried out at any particular site. Given the potential significance of public health effects and envi- ronmental costs at contaminated marine sites, on the one hand, and the potential high costs of taking corrective action at all sites, on the other hand, issues relating to the management of contaminated marine sediments are of major national importance. This importance is ref- lected in the passage of the Comprehensive Environmental Response, Com- pensation and Liability Act of 1980, CERCLA (PL 96-510) and is given additional emphasis by the enactment of the Superfund Amendments and Reauthorization Act of 1986, SARA (PL 99-499~. SARA adds a new crite- rion to the hazardous ranking system, which is used to assess sites to determine whether they should be included on the National Priorities List (NPL). Under the act, consideration must be given to "the damage to natural resources which may affect the human food chain" (Sec. 105 (a)~2~. As a consequence, SARA increases the likelihood that marine contaminated sediment sites will be included on the NPL and thereby be eligible for use of fund-financed remedial actions. Deciding how, to what extent, and whether to remediate at a site on ;

OCR for page 291
293 the NPL unavoidably confronts decision makers with very difficult deci- sions and tradeoffs. SARA mandates that the remedies selected must 1. protect human health, 2. be cost-effective, 3. meet federal and state standards, and 4. utilize permanent remedies and alternative technologies ''to the maximum extent practical" (Sec. 121 Obey. However, the level of risk to health will depend on the amount commit- ted to an action and the technology used, and remedial actions will dif- fer in their cost and degree of permanence. SARA expresses a clear preference for permanent remedies, but permanent actions are not re- quired and have been avoided because of their high cost, at least in the short run. Hence, implicit--if not explicit--tradeoffs among goals cannot be avoided in making a decision at any site. The difficulties involved are underscored by the evolving state of the art for remedial action technologies; the formidable problems inherent in quantifying risks to human health and the environment (Lave, 1987~; the need to make remedial action decisions among multiple sites in the context of fund balancing; and the requirement under SARA that the public and potentially responsible parties be actively involved in the decision process. Clearly, site-specific factors are critical considerations in reme- diation decisions at a given location. The case studies presented at the workshop illustrate the particular concerns that drive the desire for remedial actions, the strategies considered, and the institutional factors that influenced the decisions made or the actions considered in specific cases. Nonetheless, there are general principles which tran- scend particular applications. This paper examines some of the economic principles and issues that arise in deciding whether, how, and to what extent to remediate at a site. These approaches could be used to complement current approaches to evaluating sediment management alternatives (EPA, 1985b, 1985c, 1986~. A generally applicable, economics-based methodology using con- cepts from benefit-cost analysis and cost-effectiveness analysis is developed, and suggestions are made concerning how this methodology could be applied so as to capture the special characteristics of parti- cular sites. The potential usefulness and limitations of the economics methodology for assisting in decisions concerning the management of con- taminated marine sediments is described. To make economic considerations more applicable to contaminated marine sediment issues, it is particularly important that the various components be quantifiable. Hence, particular emphasis will be placed on measurement, and the potential for quantifying these arguments for particular applications will be discussed. It is recognized that quan- tification of benefits from remedial actions is exceedingly difficult, and may not be possible in all cases. Hence, the use of economic analy- sis is limited in some cases. However, it also is recognized that considerable uncertainty surrounds the use of any approach--whether based on concepts from the natural or the social sciences--used to make

OCR for page 291
294 decisions at and among sites, and thus substantial use must be made of imprecise information and informed judgment. In keeping with the scope of the workshop, particular attention is given to the role of cost-effectiveness analysis in making management decisions at particular sites. However, remediation decisions at indi- vidual sites are made within a broader framework. This broader frame- work involves, for example, fund balancing among sites and implicit if not explicit judgments concerning relative public health and environ- mental benefits and costs. In another vein, the liability provisions established under CERCLA and the Clean Water Act, as amended, have potentially important implications for encouraging source control to help avoid the creation of new contaminated sediment sites (Grigalunas and Opaluch, 1988~. Recognizing the importance of this broader frame- work, this paper goes beyond consideration of cost-effectiveness and outlines 1. the potential contribution and problems which arise from the use of benefit-cost analysis to help guide remediation decisions, and . 2. the use and limitations of liability as an approach for source control. GENERAL ECONOMIC CONSIDERATIONS As noted, contaminated marine sediments are widespread and can impose a number of public health and environmental costs, and in par- ticular cases, these costs can be substantial. For example, New Bed- ford Harbor, an area heavily contaminated with PCBs, is the marine Superfund site that has been most carefully studied by economists. Research funded by the National Oceanographic and Atmospheric Adminis- tration (NOAA) estimated the present value of damages to marine resources (using a 3 percent real rate of discount) to range from a total of $39.6 million to $52.4 million in 1985 dollars (Freeman, 1987~. These damages resulted from injury to 1. the lobster fishery, 2. public beaches and recreational fishing, and 3. reduced amenity services experienced by people living near the harbor alleged to arise from high concentrations of PCBs. Direct losses to striped bass recreational fishermen alleged to have resulted from Hudson River PCB contamination have been estimated by New York State to be more than $4 million annually (OTA, 1987, p. 43~. New Bedford Harbor and the Hudson River are among the most dramatic examples of contaminated marine sediment sites. However, numerous other cases exist of restrictions imposed on harvesting marine re- sources from areas contaminated with metals and organic chemicals, particularly for shellfish and bottom species near urban and industrial centers (see, e.g., OTA, 1987; A. D. Little, 1987; Haberman et al., 1983~.

OCR for page 291
295 On the other hand, remedial action alternatives can require a major commitment of resources, and the costs are very sensitive to the remov- al and disposal option selected. To illustrate, in the case of Ever- ett, Washington on Puget Sound, it was found that the construction cost of $55 million for upland disposal was almost four times larger than the $14.5 million construction cost of the selected alternative, con- tained aquatic disposal in the deep waters of Puget Sound (Lukjanowicz et al., 1988, p. 389. Further, available estimates suggest rapidly increasing costs with additional degrees of remediation. For example, an NRC report of PCB contamination (NRC, 1979) concludes that cost per pound of PCBs removed from the Hudson River vary from $65 per pound PCBs removed for initial cleanup of hot spots to $3,153 per pound for complete removal of low-concentration river sediments, as shown in Table 1. Ideally, one would like to measure the economic damages at a site prior to remediation, and the reduction in damages (the resultant bene- fits) expected at different levels of remediation. With this informa- tion, it would be possible to compare the increments in benefits from greater levels of remediation with the associated increments in cost. It then would be possible to assess whether remedial action at a site was worthwhile on economic grounds and to use economic principles to help guide the extent of remediation--a textbook solution to the "how clean is clean" dilemma. Difficulties inherent in measuring the full spectrum of damages make such an ideal approach beyond the reach of the state of the art in many cases. Nonetheless, in a number of instances some of the potential benefits from remediation can be quantified, and this information can be used as part of the decision process concerning proposed remedial actions. As part of any analysis, attention must be given to several impor- tant factors. Particularly important is the potential for uncertain future costs. For example, landfilling or disposal of contaminants in marine waters may cause adverse environmental impacts at some future, uncertain date if the substances become re-released. SARA recognizes TABLE 1 Incremental Costs of PCB Control in Hudson River (in 1978 Dollars) Incremental Incremental Incremental quantity control cost controlled costs per kg Policy (kg) ($ million) (dollars) A. Maintenance dredging 23,100 $ 2.5 $ 108 B. Removal of remnants 7,700 $ 0.5 $ 65 C. Removal of stabilized remnant deposits 15,100 $ 3.3 $ 219 D. Hot spot dredging 77,000 $ 22.4 $ 291 E. Removal of all river sediments 55,600 $17S.3 $3,153

OCR for page 291
296 this possibility and requires that in addition to short- term costs, long-term costs must be considered when selecting a remedial action. Such long-term costs include potential adverse health effects, long- term maintenance costs, and potential future remedial action costs should the selected alternative fail (Sec. 121 Ably. A rigorous examination of long-term costs thus would include an analysis encom- passing all feasible options and an assessment of the probability of their failure at points in time. Further, the analysis must include the probability that damages would result, given failure of each alternative; the chance that further remedial action would be taken; and the cost of such action and maintenance costs at each point in time. Clearly, the data requirements for such an analysis impose a truly major research burden on scientists and others charged with assessing remedial action alternatives. Other factors also must be considered. Potential beneficial effects in addition to health and environmental benefits could result in particular cases. For example, it may be possible to use removed materials to construct islands or provide other natural resource en- hancement (e.g., Landin, 1988) or to recover materials for reuse, as is planned for uranium contained in sediments at Port Hope, Canada (Or- chard, 1988~. Another factor to be considered is the availability and capacity of upland disposal sites. Given the limited availability of landfills and the difficulty in siting new facilities, the opportunity costs of use of the disposal site must be considered in evaluating the social impacts of alternative remediation strategies when upland disposal is being considered. Finally, it is important to recognize that in some cases, sediment contaminants may degrade/dilute over time or become covered with clean material as a result of natural deposition. In these cases, the bene- fits to be achieved through remedial action can be negligible--or, in fact, severe health or environmental costs could result should resus- pension occur. The James River kepone case is an important example. Natural sedimentation and dilution have reached the point that commer- cial fishing in the James River will be allowed for the first time in more than a decade (Huggett, 1988~. Thus, the simplified contaminated sediment management problem can be depicted in two stages; 1. public health and environmental effects--sediments are perceived to cause public health and other social losses. 2. removal and remediation--these actions result in health and envi- ronmental improvement but are costly, in terms of monetary costs of the action and possibly environmental costs associated with ecosystem disruption from physical removal and disposal in case of nonpermanent actions. Landfilling, nearshore, and offshore disposal may imply further uncertain costs or possible benefits. Although it is easy to enumerate possible costs and benefits that might arise in particular cases, it is difficult to provide quantita- tive- economic information to assess net social impact. In this regard the potential use of uncertain economic information is on the same

OCR for page 291
297 footing as the use of uncertain information from the natural sciences for making management decisions at sites. Hence, the choice is not use of one approach that provides precise information versus another that yields inexact information. Rather, all available approaches necessar- ily involve important elements of imprecision, subjectivity, and judg- ment. Important social decisions must necessarily be made within this uncertain environment since no amount of research can completely re- solve the uncertainty faced by society. Clearly, a great deal of scientific and technical information is needed to describe the problem and the tradeoffs that result from the management alternatives for addressing the problem. However, the final choice among the alternatives is necessarily a social decision based on a weighing of these tradeoffs. The next section describes two general frameworks that have been used by economists to address such issues. ALTERNATIVE ECONOMIC FRAMEWORKS FOR ADDRESSING THE CONTAMINATED MARINE SEDIMENTS PROBLEM Benef it - Cos t Analys is Benefit-cost (B-C) analysis attempts to quantify all important bene- ficial and detrimental impacts of a proposed action in dollar terms. This approach potentially can be very valuable because it is very flex- ible, and, moreover, it is the only approach that can indicate whether or not remediation is a good investment of society's resources. Hence, to the extent B-C analysis can be used as part of remediation deci- sions, it puts investments in this area on the same economic footing as public investments for environmental improvement in other areas and for public projects in general. However, B-C analysis is limited in its potential applicability due to the difficulty of providing a monetary measure of damages when eval- uating commodities that are not sold in established markets, as is typ- ically true when evaluating many environmental damages. Despite the extreme difficulty in quantifying environmental damages in dollar terms, a great deal of progress has been made in measuring these non- market effects. Economic Methods for Evaluating Nonmarket Environmental Goods Many environmental damages involve nonmarket goods and services-- that is, goods and services that are not traded in the marketplace, such as sports fishing or public beach use. Values of nonmarket goods are sometimes viewed as "subjective," and it is often claimed that these values cannot be measured in monetary terms, as is the case for market goods. However, values for market goods in many respects are no less subjective than those for nonmarket goods. For example, consumer preferences for taste, texture, color or other attributes of salmon are no less subjective than preferences for viewing wildlife; consumer will- ingness to trade off price differences for salmon attributes could

OCR for page 291
298 reveal the value of these attributes (Anderson, 1988~. The difficulty in measuring values for nonmarket goods does not derive from the fact that they are more subjective, but rather from the fact that these preferences are not directly revealed in market transactions. Two general approaches have been developed for evaluating nonmarket commodities: the revealed preference approach and contingent valuation approach. A brief description of each approach follows. Revealed Preference Approaches The basis of the revealed preference concept is that through an individual's actions, his or her preferences are revealed. Use of revealed preference is most straightforward for economic valuation of goods and services sold in the marketplace since it is relatively easy to infer preferences from these market decisions. However, the concept of revealed preference also can be used for nonmarket goods by using related-market techniques. This approach is based on the concept that the value a good or service which is not traded on a market can be inferred from closely-related goods which are sold on the market (Freeman, 19799. Two related-market approaches are outlined and illustrated in the following paragraphs. Travel cost approach. Suppose the problem is to evaluate a non- market recreational experience, such as a fishing trip to a particular site. If a fishing experience could be bought in the marketplace, such as through an entrance fee, then the observations on the decisions individuals made, given this market price of participation, could be used to value the fishing experience. Although there may be no en- trance fee for most fishing sites, the cost of participating can be measured, since in order to fish at a particular site one must travel to the site and incur certain costs in the process. Thus, the travel cost can be viewed as the price of participating, and usual market valuation approaches can be applied by examining participants revealed behavior. A recreational fisherman's value for changes in catch rates, for example, can be measured in terms of willingness to travel longer distances to more remote sites, which have higher catch rates (Brown and Mendelsohn, 1984~. An example of the application of this related-market approach to the problem of contaminated marine sediments is provided by the econ- omic damage assessment study of New Bedford Harbor PCBs. It was hypo- thesized that public awareness of the PCB contamination in the harbor affected recreational beach users and recreational fishermen. Damages would be reflected in a decrease in the demand for these recreational activities relative to the no pollution situation. For recreational beach use, the study focused on three beaches adjacent to waters or sed- iments with significant PCB concentrations. Telephone interviews were conducted with a random sample of residents of nearby communities. The results revealed that "among those aware of the PCB pollution, up to twice as many households would have visited the beach in 1986 if the PCBs had been cleaned up' (Freeman, 1987, ply. The estimated total

OCR for page 291
299 damages to beach users alleged to have resulted from the PCB pollution of the sediments was $8.3 million (Freeman, 1987, p.lO). For recreational fishing, it was estimated that the PCB pollution resulted in the diversion of 41,935 trips to other sites with an aver- age cost of diversion per trip of $1.60. Using this approach, annual damages alleged to result from the PCB pollution were estimated to be $67,100, and the present value of these alleged damages was $3.1 mil- lion in 1985 dollars. Hedonic-price approach. This related-markets approach is based on the concept that a particular market good is composed of various nonmar- ket characteristics, and that given market prices for similar goods with differing levels of the characteristics, one could calculate the implicit '"price" people are willing to pay for the characteristics. For example, the characteristics of a house can be described in terms of square footage, style, age, number of bathrooms, yard size, and neighborhood characteristics. Given data on a large number of house sales, statistical techniques could be used to relate price differen- tials to each of the characteristics in order to identify the implicit price of these characteristics. If one of the characteristics is, for example, sediment quality of the adjacent marine waters, then the value of sediment quality to an individual can be estimated in terms of the additional amount the individual would be willing to pay for a house in an area of high sediment quality, as compared to a house that has iden- tical characteristics, or after correcting for other differences in characteristics, except for the one "neighborhood" characteristic, sedi- ment quality. Note that willingness to pay in this context is based on actual or revealed behavior as reflected in the sales price of homes. This approach is particularly useful for valuing the effects of con- taminated marine sediments and, in fact, was applied in the New Bedford Harbor PCB damage assessment study. Any reduction in the amenity ser- vices of the harbor should be reflected in relative decreases in the prices of nearby residential properties. It was hypothesized that the effect of the pollution on housing prices would be stronger for those houses near the more contaminated harbor waters. The study area was divided into three zones of diminishing levels of pollution, and data were assembled on residential sales prices for single family dwellings located within two miles of the New Bedford harbor shoreline. The results indicated that the estimated total damages (reduction of pro- perty values) alleged to result from PCB contamination of marine sed- iments was between $26.2 million and $39.0 million in 1985 dollars (Freeman, 1987, p.17~. Contingent valuation approach. This survey-based approach asks individuals how they would behave under some set of given circumstances in an attempt to elicit the individual's preferences. A contingent valuation study of a fishing experience, for example, may ask whether an individual would participate in fishing at a particular site if it costs $X to do so; or the individual could be asked whether they would pay $X to experience (or avoid experiencing) a specified increase (decrease) in the catch rate. The responses to the questionnaire are

OCR for page 291
300 then used as though they were actual market behavior to assess the value of the resource issue in question. Note that this approach, in contrast to the approaches outlined in preceding paragraphs, is not based on revealed behavior. A detailed assessment of the state of the art in contingent valuation, including the importance of sources of potential bias and "reference operating conditions" for controlling bias, is contained in Cummings et al. (1985~. Valuing Public Health Risks Public health effects are a primary concern in assessing remedial actions, and SARA establishes a number of important health-related authorities to provide a better understanding of the health effects from exposure to toxic substances (Sec. 110~. From an economic view- point particularly difficult issues arise when contaminated sediments result in a threat to human health. Revealed preference is probably not useful for valuing an individual's life, since this would be tan- tamount to determining what one is willing to pay to continue living, or willing to accept to die, neither of which are reasonable concepts. However, most environmental impacts increase the risk of death for some population, rather than directly incurring death of an particular individual. Revealed preference may be useful for evaluating health risks since individuals make decisions that change risks every day, for example, through the decision to wear seatbelts, drive a car, smoke cig- arettes, etc. It is possible to elicit values related to increments in risk by observing how individuals behave when making decisions that determine the level of risk. One particularly useful related-markets approach to valuing mortal- ity risks is to examine behavior in choosing risky occupations (Fisher et al., 1988~. For example, bridge painters are often paid differing amounts depending on whether they paint the more risky, top parts of the bridge, or the relatively safer, lower parts. Thus, the individual trades off wages for risk of death, and behavior revealed in the labor market can provide information on the individual's preferences for safety, or lack of risk. The key concepts used in economic analysis on mortality risk are the value of a statistical life and cost per life saved, which will be discussed below. Valuing a statistical life is an example of a cost- benefit approach that can be illustrated as follows. Suppose that contaminated sediments lead to human ingestion of a toxic substance through the food chain and that risk analyses show that this level of concentration implies the probability of death through cancer is increased by .001 percent for each of one million people. The number of statistical lives lost by this pollutant is .00001 X 1,000,000 ~ 10 If, on average, individuals are willing to give up $100 in wages in order to reduce the risk of death on the job by .001 percent, then the value of a statistical life is $10 million, as revealed by actions of

OCR for page 291
301 the individuals. This implies that a remedial action that would remove this pollutant from sediments, hence removing the health risk, would result in $100 million in benefits This concept can be applied to the problem of contaminated sediments, given a risk analysis of the increased mortality which result from various exposure pathways. It should be noted that the approach discussed in this section assesses statistical lives saved, or mortality risks, not reductions in sublethal effects, or morbidity. The concepts to be used to evaluate the benefits from reduced morbidity--reduced medical costs, smaller amount of time out of work due to illness, and so forth--are relatively straightforward. Note, however, these approaches do not value the ill- ness per se, such as the discomfort or suffering of the individual, but only the associated monetary costs, and thus would strictly understate the value of reduction in morbidity rates. Additionally, it can be very difficult to establish the incremental improvements in health from total or partial remediation of contaminants at a given site due to the inherent difficulties in isolating the effect of exposure to one or more contaminants from all other effects that influence health. An alternative approach that has been employed to value mortality is the so-called human capital approach. This approach uses the earn- ing potential of the individual over his or her future life as the value of human life. The human capital approach has been widely used in the courts in cases of wrongful or accidental death. While this approach measures potential future earnings, it does not place a value on the loss of life, per se, nor does it measure the individual's will- ingness to accept risk of death. In addition, the approach has an in- nate bias against those who have little or no direct wage income, such as a housewife whose services are not valued through the market. Thus, the human capital approach would be expected to significantly under- state the value of life. Indeed in practice, the value of a statisti- cal life determined from revealed preference studies tends to be signi- ficantly higher than the value determined from the human capital approach. For example, the EPA uses figures of $400,000 to $7 million as a range of reasonable values for a statistical life based on re- vealed preference . A present value of this s ize implies a perpetual income of $32,000 to $560,000 at an 8 percent discount rate . Cost-Effectiveness Analysis General Considerations Generally, health and environmental benefits of remediation at a site are not explicitly measured in dollar terms, but implicitly may be judged to be worth the costs of remedial measures. In these cases, cost-effectiveness (C-E) techniques can be used to guide the selection of remediation alternatives by helping to assess the relative cost and the effectiveness of the alternative removal and disposal strategies. SARA requires that C-E is to be considered in the evaluation of reme- dial actions, and that long-term as well as short-term costs be taken into account.

OCR for page 291
302 C-E analysis provides a systematic way for determining (1) the least-cost approach(es) for achieving a given objective, or (2) the maximum level of the objective which can be achieved for a given cost. Correctly applied, C-E can be a powerful tool because it potentially allows decision makers to screen remedial action alternatives on the basis of a common measure. All else being equal, if remedial action alternative A is less expensive than B. then A clearly is preferred; or, if A and B are equally expensive but A results in a greater public health and environmental improvement than B. then A would be selected. However all else generally is not equal, comparisons rarely are so straightforward, and as a result C-E analysis is subject to several potentially important shortcomings. Potential Problems with Cost-Effectiveness Analysis C-E analysis will lead to misleading results if (1) important costs are ignored or (2) costs for alternative actions at a site or between sites are not estimated using a consistent approach. As noted, SARA expresses a clear preference for permanent actions, involving thermal, biological or chemical treatment to reduce the volume, toxicity or mobility of the substancefs). Permanent remedies typically cost more initially than actions that do not involve treatment (e.g., capping). For example, a recent report found that for the ten cases studied (none marine), the average cost of the five cases involving permanent reme- dies through treatment of the removed materials was $16 million as compared to $7.5 million for nontreatment (impermanent) remedies (OTA, 1988~. Assessing whether permanent remedies are more cost-effective when long-term as well as short-term costs are considered is extremely difficult, as noted above. Nonetheless, long-term costs must be con- sidered if C-E analysis is to be a meaningful guide for remedial action policy. Clearly, for C-E analysis results to lead to appropriate decisions, it also is vital that costs be assessed consistently. A recent report by OTA found that in some cases (none marine) different approaches were used by contractors to estimate costs. To the extent the use of incon- sistent approaches at marine sites causes large differences in apparent costs, the usefulness of cost-effectiveness analysis can be severely compromised. Application of Cost-Effectiveness to Public Health Risks The C-E approach does not place a value on health or other benefits and hence cannot be used to provide an economic argument in support of, or against, remediation at a site. However, C-E analysis can play an important role in helping to guide the extent of remediation at a site and among sites (fund balancing). To illustrate this, the concept of cost-per-life-saved is used. Rather than placing a value on a statis- tical life, the cost-per-life-saved approach ranks options according to the number of statistical lives saved per dollar spent. Similarly, the

OCR for page 291
303 approach could be applied for nonlethal affects, for example by evalu- ating cost per reduced cancer case. To illustrate, suppose there is a fund of $50 million, and six non- mutually exclusive cleanup alternatives are being compared, as shown in Table 2. The alternatives may represent, for example, cleanup of var- ious subareas within some particular contaminated site, such as the four subareas of New Bedford harbor discussed by Ikalainen and Allen (1988~. Additionally, some alternatives may represent subareas from differing contaminated sites, such as four subareas in New Bedford Harbor and two subareas in the Hudson River. For the cases depicted in Table 2, assuming no impacts other than reduced health risks, alternatives one, two, and three would be chosen, resulting in 14 statistical lives saved. Given these alternatives, this is the greatest reduction in risk that would result from this fixed expenditure of the $50 million fund. This approach implies a social willingness-to-pay per statistical life saved between $6.7 mil- lion, the highest cost per life saved for the alternatives chosen, and $8 million, the lowest cost per life saved for the alternatives not chosen. Given the same alternatives, the cost-benefit criterion out- lined in the preceding section would justify alternatives one, two, three, and four, using the $10 million hypothetical value per statis- tical life derived above. This would result in 15 statistical lives saved and would result in expenditures of $58 million, which would require $8 million in addition to the $50 million contained in the fund. SOURCE CONTROL: THE ROLE OF LIABILITY UNDER CERCLA AND THE CWA Under CERCLA and the Clean Water Act, as amended, polluters are liable not only for cleanup and reasonable assessment costs, but also for " damages for inj ury to, destruction of, or loss of natural resources" (Sec.107.(a)~4~(C) (hereafter, injury to natural resources) resulting from a spill . Briefly, CERCI-A provides for two types of TABLE 2 Depiction of Cost-Effectiveness Strategy Using the Cost-per- Life-Saved Approach Alternative One Two Three Four Five Six Cost ($ Million) $10 $20 $20 $8 $12 $15 Number of statis- tical lives saved 6 5 3 1 1 1 Cost per statis- tical life saved $1.7 $4 $6.7 $8 $12 S15

OCR for page 291
304 damage assessment regulations. The type A regulations provide a simpli- fied approach, involving minimal field observation to be used for minor incidents of short duration, while the type B regulations describe meth- ods for site-specific natural resource damage assessments with poten- tially extensive field observations, to be used for major incidents. Since contaminated sediment problems generally arise as a result of chronic releases over an extended period, the type B approach almost always will be the appropriate approach for measuring damages. For example, the economic damage assessment study of New Bedford harbor PCBs was a type B study, the first carried out under CERCLA (e.g., Freeman, 1987~. The two-tiered damage assessment approach mandated by Congress recognizes that undertaking a damage assessment can be very expensive. For example, the economic study of the damages alleged to result from the presence of PCBs in New Bedford Harbor cost $0.5 million (Meade, NOAA, personal communication). For some cases, these assessment costs can exceed the value of the damages that can be ascertained. For example, 5,600 barrel ARGO Anchorage oil spill cost $250,000 for damage assessment, while the resultant damage estimate was $33,000. Clearly, it only makes sense to spend the large amounts of money necessary to carry out field significant investigation when assessing damages for very large incidents, such as the New Bedford Harbor case. The intent of CERCLA is to compensate governments for damages to publicly controlled natural resources in their role as trustees of these resources. Thus, the primary goal of the act is to encourage fairness by compelling the responsible party to pay compensation for the damages resulting from their actions; the amount recovered is to be "available for use to restore, rehabilitate, or acquire the equivalent of such natural resources by the appropriate agencies...'' (Sec.107(f)~. However, as an unintended side-effect, the liability provisions of CERCLA create a legal framework for what is akin to a "tax" on pollu- tion incidents covered under the act. As such, the damage assessment regulations introduce what could be an important new approach for using economic incentives to avoid pollution for a wide range of incidents (Opaluch and Grigalunas, 1984; Grigalunas and Opaluch, 1988~. For exam- ple, in discussing the liability provision, the 1982 version of the Clean Water Act requires that "the Administrator shall . . . conduct a study and report to Congress on methods, mechanisms, and procedures to create incentives to achieve a higher standard of care in all aspects of the management and movement of hazardous substances." The potential importance of incentives embodied in the liability provisions in the CERCLA regulations for source control is made clear by examining its applicability and unique characteristics. The regu- lations apply to virtually all publicly controlled natural resources, and encompass a wide span of pollution discharges. Also, CERCLA holds polluters strictly liable for their actions, so that following an inci- dent, there is no need to establish negligence in a prolonged and cost- ly court trial. Moreover, CERCLA establishes joint and several liabil- ity. Thus, any one polluter can be held liable for all cleanup or remediation costs, even if they contribute only a small share of the total amount released into the environment. Note, however, that these

OCR for page 291
305 incentives are applicable only to accidental spills, and not, for exam- ple, to routine discharges permitted under the National Pollution Dis- charge Elimination System (NPDES) of the Clean Water Act. Another unique and very important characteristic of the CERCLA natural resource damage assessment regulations merits emphasis. The regulations provide an advantage for trustees in that they carry the force of rebuttable presumption (Sec.lll (h)~2~. That is, if the pro cess set out in the regulations is correctly applied by the authorized official following a spill, the resulting measure of ~ sumed to be correct, unless the potentially liable party can show other- wise by a preponderance of the evidence. In most cases it will be very difficult and costly to prove that the results of a damage assessment carried out under the act are incorrect, especially for the type A approach, by virtue of the fact that it is intended to be simplified and based on minimal field observation. Hence, the rebuttable presump- tion provision of the CERCLA can have important implications for the effectiveness of the damage assessment regulations, in general, and especially for the type A approach. It should also be noted that under CERCLA, liability extends beyond custody of the material to include materials spilled by those under con- tract, directly or indirectly, with the firm. Thus, liability under CERCLA provides incentives not only for careful handling of materials, but also for careful choice of parties with whom to contract for waste removal and final disposition. Hence, CERCLA recognizes the importance of choice of contracted parties in order eliminate the obvious finan- cial incentive to hire inexpensive "fly-by-night" contractors who prac- t~ce midnight dumping of hazardous wastes. There is some evidence that liability provisions have been effec- tive in providing incentives for damage reduction. For example, in a study of industries which produce hazardous wastes, Killory (1987) con- cludes "[source reduction] has become an increasingly attractive envi- ronmental policy for the organic chemical industry because of the high costs of waste disposal and, more importantly, the greater liability that producers now incur for generated wastes." Further, the only empir- ical study of economic behavior under liability finds some evidence that is consistent with provision of incentives (Opaluch and Grigalunas, 1984~. Thus, the liability provisions of CERCLA may provide incentives both for source control and for careful handling and disposal of hazar- dous materials. This is an extension of the so-called "polluter pays principle" which holds the polluter financially responsible for costs associated with the harmful effects of pollution emissions. Note, however, that despite the success of incentive-based approaches in Europe, U.S. environmental legislation does not maintain this same incentive system for other sources of environmental pollutants, such as pollution emission under NPDES permits. For many years economists have argued for environmental policy based at least in part on financial incentives. One alternative would be a mixed system of direct regula- tion and financial responsibility that would require the firm to attain some stated treatment percentage, but would also require the firm to pay a fee for the remaining pollutants emitted or would pay a subsidy damages is pre-

OCR for page 291
306 if pollution were reduced below that required by the regulation (Baumol and Gates ? 1975; Roberts and Spence, 1976), perhaps accounting for potential locational differences in impacts (Tietenburg, 1978~. Although the act is widely applicable, the type A approach cannot be used in a number of important cases; and CERCLA itself is of limited applicability in some cases. For example, the type A approach cannot be used to assess damages from chronic releases, although the type B approach can be employed. CERCLA does not apply to releases under NPDES permits, although it does apply when the permitted release is exceeded, nor can the act be used to assess damages from releases of fertilizer or pesticides resulting from normal use. Where CERCLA does not apply, other laws and approaches will have to be used to control releases of contaminants into marine waters. However, these acts do not generally provide incentives, such as those implied by the liabil- ity provisions of CERCLA, the Clean Water Act, or the Outer Continental Shelf Lands Act. The CERCLA natural resource damage assessment regulations are rela- tively new and, in many respects, novel. How effective they will prove to be depends importantly upon on several factors, including how active states are in implementing this approach. To be effective, trustees must be appointed, staff must become familiar with the regulations, and efforts to apply the regulations to releases under the act must be pur- sued. If the act is not implemented, than it will be of little use. It is not clear that trustees have fully explored the potential useful- ness of this approach. Further, liability can only be applied in cases where one or more responsible parties can be identified. For many cases of illegal dumping this may not be possible. A final note is in order. A unique part of CERCLA is the require- ment that the damage assessment regulations be reviewed every two years and updated, as appropriate. Hence, there is the important opportunity to suggest new techniques or data to be included in updated natural re- source damage assessment regulations. Also, the biennial review man- dated for CERCLA may provide an opportunity to explore the feasibility of developing a simplified approach which could be applied to those con- taminated marine sediment cases which may not warrant the high cost of a type B study but which cannot be encompassed within the present type A framework. SUMMARY AND CONCLUDING COMMENTS Contaminated marine sediments are of concern because they can impose a variety of adverse public health effects and environmental losses. At the same time, remediation can be very costly. Hence, whether, how, and the extent to which sites should be remediated are important national issues. However, quantification of public health and environmental effects unavoidably involves considerable scientifi uncertainty and social tradeoffs. In light of the many uncertainties involved and the lack of clear criteria, these decisions concerning remediation at a site necessarily are based on imprecise information and important elements of uncertainty and subjectivity. Hence, these C-

OCR for page 291
307 issues are principally social decisions, and not merely scientific, technological or economic issues. Two economic frameworks are available for contributing to social management decisions at contaminated marine sediment sites. The two economic frameworks presented--benefit-cost analysis and cost- effectiveness analysis--can contribute to the decision process by making explicit the costs and benefits (in the case of cost-benefit analysis) of the alternatives. Despite the many difficulties inherent in quantifying some of the important factors, these approaches can be used to complement the use of scientific information in making deci- sions at or among contaminated sediment sites. These economic frameworks provide a means of organizing the infor- mation in ways that can be helpful to decision makers. The techniques are particularly useful for identifying alternatives that achieve goals at excessive costs. For example, the cost-per-life-saved approach would identify policy options that save few lives at relatively high costs, in favor of alternatives that result in a greater reduction in risk per unit expenditure. To the extent that the benefits from alter- native cleanup policies can be quantified, benefit-cost analysis can provide a perspective on relative benefits and costs to help determine whether goals appear to be reasonable. This can be particularly useful when "conservative" high or low estimates can be consistently utilized and imply an unambiguous solution. For example, if conservative, low estimates of benefits of a remedial action exceed costs, then certainly that level of remediation would be warranted. On the other hand, if costs greatly exceed benefits, even when overstated benefit estimates are used, then it is likely that somewhat less ambitious levels of action may be warranted, with reallocation of funds for expenditure at some alternative site. For example, complete remediation of all tainted sediments from a particular site may be prohibitively expensive and the costs of doing so would be beyond any reasonable level of bene- fits which may result. Finally, if the results of a cost-benefit anal- ysis are not conclusive, then the project can neither be justified nor rej ected on a C-B basis, and other considerations would dominate the .ecls Ion . It must be recognized, of course, that quantifying benefits and costs can be exceedingly difficult. This is particularly true when examining long-term costs associated with nonpermanent solutions. In evaluating costs associated with impermanent solutions, such as in situ capping, long-term costs associated with failure must be considered to make comparisons with costs associated with permanent solutions, such as incineration. EPA calls for a screening of the alternative actions to eliminate those that cost more but do not provide a "commensurate'' public health or environmental benefit. However, determining whether these benefits are "commensurate" places a significant burden on scien- tists and economists. To do so the probability of failure for imperm- anent solutions must be determined, in addition to the consequences of failure considering the potential for, and costs of, any associated remedial action. However, it is possible in many cases to measure the benefits from improvement, as was illustrated by the results presented for the New

OCR for page 291
308 Bedford Harbor damage assessment study. Moreover, it is important also to recognize that noneconomics-based approaches must also consider the same uncertainties and tradeoffs, but will do so in a way that leaves the tradeoffs implicit. In contrast, an economics-based approach speci- fies these tradeoffs explicitly, so that when choosing an action, the decision maker can see the alternatives being given up--the opportunity cost. C-E analysis begs the question of whether remediation ought to take place at a site. However, given that a decision to remediate has been made, C-E analysis can be an important part of remediation decisions, and under SARA C-E is given a central role in designing remedial ac- tions. This section reviewed the C-E approach and illustrated its app- lication to public health effects. Properly applied, C-E analysis can provide a powerful tool for choosing among alternative approaches for remediation at a given site and for allocating resources among sites. However, there is the danger that the concept of cost effectiveness may be confused with the least costly remedial action. Various actions can only be compared on a cost-effective basis if the benefits are equal, but the costs differ; if the costs are equal and the benefits differ; or if the least costly alternative also results in the highest level of benefits. Cost-effectiveness analysis cannot be used to compare two alternatives where one is more costly but leads to greater environmen- tal benefits. Additionally, the role of li ability as an approach for encouraging source control was examined. CERCLA holds polluters strictly liable for their actions. The prospect of paying potentially very consider- able sums for damages, assessment, and remediation actions creates a powerful incentive to reduce the amount and the toxicity of materials potentially spilled, as well as to handle more carefully and dispose of the materials that remain. Another unique, and very important charac- teristic of the act is that the natural resource damage assessment regu- lations carry the force of rebuttable presumption. Thus, the damage assessment process is greatly facilitated by shifting the burden of proof. Given the characteristics of the act and its broad potential appli- cability, the damage assessment regulations established by CERCLA clearly are a major development in environmental policy. CERCLA's dam- age assessment regulations also may represent a major, and perhaps un- precedented, expansion of the use of economic incentives to control pollution. However, to be effective trustees must be appointed and the act must be enforced. It is not clear that states have fully exploited the potential of the act for assessing damages and remediating contam- inated marine sediment sites. REFERENCES Anderson, J. L. 1988. Analysis of the U.S. Market for Fresh and Frozen Salmon. Staff Paper. Kingston, RI: Department of Resource Econo- mics, University of Rhode Island. 65 pp. (plus app.~. A. D. Little, Inc. 1987. An Overview of Sediment Quality in the United

OCR for page 291
309 States. Final report to Monitoring and Data Support Division, Of- fice of Water Regulations and Standards, U.S. Environmental Protec- tion Agency, Washington, D.C. (June). Barnett, H. C. 1985. The allocation of Superfund, 1981-1983. Land Econ- omics 61~3~:255-262. Baumol, W. J. and W. Gates. 1975. The Theory of Environmental Policy: Externalities, Public Goods and the Quality of Life. Englewood Cliffs, N.J.: Prentice Hall. Brown, G. and R. Mendelson. 1984. The Hedonic travel cost method. Rev. Econ. and Stat. 66~3~:427-433. Cummings, R. G., D. S. Brookshire, and W. D. Schulze, eds. 1986. Valu- ing Environmental Goods. Totowa, N.J.: Rowman and Allanheld. 270 p. EBASCO Services. 1987. Detailed Analysis of Remedial Technologies for the New Bedford Harbor Feasibility Study. EPA Contr. 68-01-7250. Washington, D.C.: U.S. EPA. Fisher, A., L. G. Chestnut, and D. M. Violette. 1988. The value of reducing risks of death: A note on new evidence. J. Policy Analysis and Management 8~1~. Freeman, A. M. 1979. The Benefits of Environmental Improvement: Theory and Practice Resources for the Future. Baltimore: Johns Hopkins Press. Freeman, A. M. 1987. Assessing damage to marine resources: PCBs in New Bedford Harbor. Paper presented at the meetings of the Assoc. of Environmental and Resource Economists, Chicago, Dec. 27. 25 p. Grigalunas, T. A., J. J. Opaluch, D. French, and M. Reed. 1987. Meas- uring Damages to Coastal and Marine Natural Resources: Concepts and Data Relevant for CERCLA Type A Damage Assessments. Springfield, Va.: National Technical Information Service (2 vols.~. Grigalunas, T. A., J. J. Opaluch, D. French, and M. Reed. 1988. Meas- uring damages to marine natural resources from pollution incidents under CERCLA: Application of an integrated ocean systems/economic model. Mar. Res. Econ. 5~1~. Grigalunas, T. A. and J. J. Opaluch. 1988. Assessing liability for damages under CERCLA: A new approach for providing incentives for pollution avoidance? Nat. Res. J. 28~3~. Haberman, D., G. B. Mackiernan, and J. Macknis. 1983. Toxic compounds. Chapter 4 In Chesapeake Bay: A Framework for Action. Philadelphia: U.S. Environmental Protection Agency. Huggett, R. J. 1988. Kepone and the James River. Paper presented at the National Research Council Workshop on Contaminated Marine Sediments, Tampa, Fla. Ikalainen, A. J. and D. E. Allen. 1988. New Bedford Harbor Superfund Project. Paper presented at National Research Council Workshop on Contaminated Marine Sediments, Tampa, Fla. Kahn, J. R. and W. M. Kemp. 1985. Economic losses associated with the degradation of an ecosystem: The case of submerged aquatic vegetation in Chesapeake Bay. J. Environ. Econ. and Manage. 12~3~. Killory, H. C. 1987. Getting to the source of hazardous waste. Resources No. 89 (Fall). Lave, L. B. 1987. Improving quantitative health risk assessment tech- niques. In Environmental Monitoring, Assessment and Management,

OCR for page 291
310 Draggan et al., eds. New York: Praeger Publishers. Lukjanowicz, E., J. R. Paris, P. F. Fuglevand and G. L. Hartman. 1988. Strategies and technologies for dredging and disposal of contami- nated marine sediment. Presented at the National Research Council Workshop on Contaminated Marine Sediments, Tampa, Fla. Landin, M. C., ed. 1988. Beneficial Uses of Dredged Material: Proceed- ings of the North Atlantic Regional Conference, 12-14 May 1987. Baltimore Maryland. Vicksburg, Miss.: U. S. Army Engineer Waterways Experiment Station. National Research Council (NRC). 1979. Polychlorinated Biphenyls. Wash- ington D.C.: National Academy Press. Office of Technology Assessment (OTA). 1987. Managing dredged mater- ials. _ Wastes in the Marine Environment. Washington, D. C.: OTA. Opaluch, J. J. and T. A. Grigalunas. 1984. Controlling stochastic pol- lution events with liability rules: Some evidence from OCS leasing. The Rand J. Econ. 15(Spring). Orchard, I. 1988. Oral statements presented at National Research Coun- cil Workshop on Contaminated Marine Sediments, Tampa, Fla., May. Roberts, M. J. and M. Spence, 1976. Effluent charges and licenses under uncertainty. J. Pub. Econ. Vol. 5. Schwartz, D. G. and I. E. Strand, 1981. Avoidance costs associated with imperfect information: The case of kepone. Land Econ. 57~2~. Tietenburg, T. H. 1978. Spatially differentiated air pollutant emission charges: An economic and legal analysis. Land Econ. 54~3~. U.S. Congress. 1980. Comprehensive Environmental Response, Compensation and Liability Act of 1980. Public Law 96-510, 96 Cong. 94 (Dec. 11, 1980) Stat. 2767-2811. U.S. Congress. 1982. The Clean Water Act (as amended through December, 1981~. Public Law 97-117, 97th Cong. (Feb., 1982~. Washington, D.C., U.S. Government Printing Office. U.S. Congress. 1986. The Superfund Amendments and Reauthorization Act of 1986. Public Law 99-499, 99th Cong. Washington, D.C. U.S. Department of Commerce (DOC), National Oceanographic and Atmospheric Administration, Ocean Assessments Division, 1988. A Summary of Data on Chemical Contaminants Collected During 1984, 1985, 1986, and 1987. Rockville, Md.: NOAA. U.S. Environmental Protection Agency (EPA). 1985. Remedial Action Costing Procedures Manual. Cincinnati, Ohio: Hazardous Waste Engineering Research Laboratory, Office of Research and Development. U.S. Environmental Protection Agency. 1985b. Guidance on Feasibilitv Studies Under CERCLA. Office of Emergency and Remedial Response and Offices of Waste Programs Enforcement and Office of Solid Waste and Emergency Response, Washington, D.C. U.S. Environmental Protection Agency. 1985c. Guidance on Remedial In- vestigation Under CERCLA. Office of Research and Development, Office of Emergency and Remedial Response and Office of Solid Waste and Emergency Response, Washington, D.C. U.S. Environmental Protection Agency. 1986. Superfund Public Health Evaluation Manual. Office of Emergency and Remedial Response, Washington, D.C.