Issue Analysis 70 – Superfund: The Good, the Bad, and the Broken

When President Clinton was first elected, both the new administration and Congress were calling for significant reforms of Superfund, a program created to clean up the nation’s toxic waste dumps. “[W]e all know it doesn’t work — the Superfund has been a disaster,” noted President Clinton.1 Three Congresses later, however, Superfund remains virtually untouched by reform, despite broad consensus that the program is broken and ill-suited for accomplishing its original mission.

At the most fundamental level, Superfund is a flawed program built from a premise that guarantees its failure. Although created with the best of intentions — to identify and clean up hazardous waste sites — the Superfund program was structured around a faulty mechanism that relies on a self-perpetuating “blame game” to raise the funds necessary to clean up waste sites. A bizarre set of liability rules at the heart of the program has diverted resources away from actual cleanups in favor of costly legal proceedings and enforcement actions that do little to clean up communities. To date, more than 100,000 parties have been dragged into the Superfund net, which has been cast so wide that it includes the Catholic Church, hospitals, schools, and a Girl Scout troop.

A Brief History

In 1978, Love Canal, an abandoned waste site near Niagara Falls, brought toxic waste into the media limelight. Concerns over chemicals seeping into the homes of local residents and the subsequent evacuation of almost 1,000 families garnered headlines across the nation and spurred Congress into action.2 The result was the Superfund program, which was created in 1980 when Congress passed the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA). The new federal program enacted an extensive liability system and created a public trust fund — the Superfund — to fund cleanups where responsible parties could not be found.

The EPA and other defenders of the program view the liability system as the core of the Superfund program, claiming that it embodies the virtues of the “polluter pays” principle. Through a web of retroactive, strict, and joint and several liability, the Superfund program searches out “potentially responsible parties” (PRPs) who can then be held liable for the costs of a cleanup. This allows the EPA to cast a wide net when looking for parties to pay for a cleanup. Unfortunately, relying on the legal system for revenue is an unfair and inefficient way to raise money. Resources are focused on litigation, not cleanup, with tremendous amounts of money diverted to legal fees, negotiations, and administrative costs. Studies by the RAND Corporation found that up to 36 cents out of every dollar go toward such costs, and that for firms with annual revenues of more than $100 million, up to 60 cents out of every dollar go towards transaction costs rather that hazardous waste cleanup.3

In the case of “orphan sites,” where no responsible parties are identified, the cleanup is funded through the Superfund trust. The trust fund is financed primarily through excise taxes on petroleum and chemical feedstocks as well as a corporate income tax.4 Since its inception in 1980, the Superfund has grown from $1.6 billion to more than $15 billion, of which $10 billion had been spent by the end of 1994. Private sector expenditures on Superfund clean-ups added in, a total of more than $30 billion has been spent over the last 17 years.

Unfortunately, there is little to show for all the money that has been spent. The cleanup process is slow and tedious. On average, the typical Superfund site takes more than 10 years to cleanup at a price tag of $32 million — excluding litigation and administrative costs. As a result, only around 40 percent of the priority sites identified by the EPA have been cleaned up. At the same time, the program has become a bureaucratic nightmare, consuming more than 20 percent of the EPA’s $6.8 billion budget. Moreover, 47 percent of the Department of Justice’s enforcement actions for major environmental programs are dedicated to the Superfund program. From 1995 to 1997, a total of 839,500 hours were billed to Superfund enforcement, according to the Department of Justice. Without fundamental reforms, the costs will only increase. A study conducted at the University of Tennessee estimates that, under the present system, cleaning up 3,000 Superfund sites will cost between $150 billion and $352 billion.5

Despite the dismal record of the Superfund program, the Clinton administration and EPA Administrator Carol Browner defend the program, claiming that administrative reforms have effectively fine-tuned the program and hastened the pace of cleanups. As Browner recently stated, “By any measure, we are making a great deal of progress in our efforts to improve the nation’s hazardous waste cleanup program — to make it faster, fairer, and more efficient — and to ensure that it does the best possible job of protecting the health of our citizens and returning land to communities for productive use.”6

The Cleanup Process

Superfund is a comprehensive program that starts with the identification of potential hazardous waste sites with continued oversight through the final remediation, or clean-up, of priority sites. To identify sites, the EPA has created an extensive database known as CERCLIS (Comprehensive Environmental Response, Cleanup, and Liability Information System), which catalogs potential hazardous waste sites based on information provided by the public and the EPA’s internal studies.

Once a site is catalogued in CERCLIS, a preliminary assessment determines whether further action is necessary. Where further action is warranted, the EPA will conduct a site inspection to assess the potential risk and what remediation may be required. At this point, about one-third of the sites are listed as requiring no further action.7 For the remaining sites, the EPA uses its Hazardous Ranking System (HRS) to evaluate the severity of the problem. A ranking of 28.5 or higher qualifies a site to be placed on the National Priorities List (NPL), which qualifies a site for federal funding. Although the NPL is supposed to ensure that the most dangerous health risks are addressed most swiftly, critics of Superfund claim that NPL status does not necessarily correlate with health risk. Some sites on the list may pose little or no health threat, while others that are not listed may actually pose greater health risks.8 The HRS ranking of 28.5 is arbitrary, a political artifact of the original legislation that required 400 sites to be listed on the NPL (with at least one in every state). Today, there are 1,192 sites on the NPL, and the EPA has proposed adding another 52 sites.9

Two basic cleanup options are available to the EPA under Superfund: short-term removal actions and long-term remediation. Removal actions spark little debate and are widely acknowledged to be effective in removing immediate health risks. Often, removal actions are emergency measures aimed at cleaning up chemical spills and removing hazardous wastes that threaten a local population.10

Long term remediation efforts are far more controversial, and questions about their funding and procedures lie at the core of the debate over Superfund reform. Once a site is listed on the NPL, the cleanup process involves significant oversight by state and EPA officials. The process begins with a remedial investigation to determine the physical and chemical properties of the site and the exposure levels of the local population. The EPA regional administrator is responsible for evaluating the findings and selecting the appropriate method of remediation, which is referred to as the Record of Decision (ROD).

To comply with the ROD, those responsible for the cleanup develop a remedial design, which will then be used to carry out the final remedial action. At large Superfund sites, this process can be lengthy because the remediation process is sub-divided into a number of individual operating units, each of which requires its own studies and cleanups. Once the remedial action is completed, the EPA, in conjunction with state officials, can remove the site from the NPL. Due to the fact that sites often must be maintained and monitored over time, sites commonly remain on the NPL but are shifted to the “construction completed” category, where many sites remain for years before being removed from the NPL. The EPA currently lists 506 sites as construction completed and 162 sites as deleted from the NPL.11 However, these numbers do not indicate how many cleanups actually have occurred. For example, a study by the General Accounting Office (GAO) of 149 sites found that for 21 percent of the sites no cleanup was necessary.12

Tag, You’re It!

The liability system that serves as the foundation for the Superfund program is as expansive as it is flawed. For the purpose of Superfund cleanups, liability is strict, retroactive, and joint and several. Strict liability means that individuals can be held responsible regardless of intent or negligence. Retroactive liability means that individuals can be held responsible for past actions, even if those actions were legal at the time. Finally, joint and several liability means that any given individual may be held liable for the full cleanup, not just a proportionate share. Taken together, these liability rules allow the EPA to cast a wide net for PRPs and deep pockets that can bear the costs of the cleanup. Due diligence and other standard legal defenses do nothing to minimize the threat of a being identified as a PRP. The original CERCLA legislation identifies four specific groups to whom the liability standards apply: present owners and operators of a site, past owners, individuals who generated the waste at a site, and transporters who hauled the waste to the site.

Only orphaned sites are cleaned up using the trust fund.13 In all other instances, the EPA relies on the Superfund liability scheme to identify PRPs that will pay for the necessary remediation. The EPA can either begin the remediation process itself, using money from the trust fund, or the agency can issue an administrative order demanding a PRP to begin the cleanup process under the threat of fines that start at $25,000 per day and can rise to $75,000 per day for ensuing violations. Either way, PRPs will eventually pay; if the EPA conducts the cleanup itself, it will seek reimbursement from PRPs.

When the EPA identifies PRPs and demands their participation in the remediation process, the cogs of the legal system begin churning. Existing PRPs begin filing “contribution suits” to expand the number of PRPs involved in order to reduce their share of the cleanup costs. The list of PRPs quickly expands from the handful of “corporate polluters” identified by the EPA to a long list of smaller and smaller businesses such as “mom and pop” businesses on Main Street, hospitals, the Girl Scouts, and even an Elks Club in one instance.

To understand the reach and tangle of Superfund, consider Petroleum Products in Pembroke, Florida. This was a recycling site where people sent waste oil to be recycled. The site was placed on the NPL in 1987, and somewhere between 2,000 to 4,000 PRPs have been identified. The PRPs include cities, the University of Miami, the Miami Fire Department, and thousands of others who thought they were acting responsibly by sending their waste oil to a recycling facility. Although many have been dragged into the liability maze, no long-term cleanup has been done in the ten years that the site has been on the NPL.

In another instance, the original PRPs at a site in Ludlow, New York, called a meeting at the local auditorium, inviting all the local businesses to attend. At the meeting, the small businesses were told that they should offer a payment to cover part of the cleanup or else they would be sued. The small businesses decided to pay $1,500 each to get out of the Superfund web, because it would be cheaper to pay than go to court. These payments were made under the threat of litigation, rather than as a result of any evidence that found them liable for the site. As Rep. Sherwood Boehlert (R-NY) stated: “They were operating under the current Superfund law. We should never let this happen again.”14

Clearly, leaving PRPs to sort out their differences in court can be a costly venture, as expenses are compounded by enforcement actions and lengthy negotiations with the EPA at every step of the remediation process. Moreover, the legal and other transaction costs incurred contribute nothing to the remediation process. Numerous studies have examined the burden of transaction costs on the Superfund program. As noted earlier, RAND estimated that up to 36 cents out of every dollar goes to transaction costs instead of cleanup. And the greater the number of PRPs, the higher the transaction costs. Another RAND study found that at single party sites, transaction costs were 5 percent of total costs with costs rising to between 34 percent and 36 percent of total costs at sites with more than 16 PRPs.15

The liability system’s legacy is also evident when examining the length of time required to clean up Superfund sites. Fifty-three percent of the orphan sites, where no PRPs are present and cleanup is funded by the EPA, have been cleaned up. At sites with more than 500 PRPs, however, only 17 percent have been cleaned up.16

Despite mounting evidence that the liability system diverts resources away from cleanups and delays their completion, the EPA continues to defend the liability system on the grounds that it embodies the “polluter pays” principle. Any changes to the current liability system, they claim, will “let polluters off the hook.” A closer examination of the Superfund, however, demonstrates the weakness of this claim. First, the simple scope of PRPs should suggest that the liability system hews too broadly and indiscriminately to be seen as punishing polluters. In fact, an argument could be made that the liability system provides polluters with incentives to shift the burden to those who bear little or no responsibility for a Superfund site. To force a small restauranteur to pay tens of thousands of dollars in cleanup costs for sending mashed potatoes to the local dump is hardly taking a stance against polluters.17

Also, the Superfund itself is funded by a special tax on those identified by the EPA as corporate polluters — oil and chemical companies. Such companies are forced to pay into the trust fund, irrespective of their records as environmental stewards. Even in the absence of identifiable PRPs, “polluters” are paying. Unfortunately, this tax does little to provide incentives for favorable environmental practices; the tax makes no distinction between good and bad actors.18

Another claim made by supporters of the status quo is that without the threat of the Superfund program, bad actors will continue to pollute. Yet, it is important to remember that Superfund was created to address existing hazardous waste sites, not to regulate current waste management practices. There are numerous federal, state, and local laws and regulations that govern the disposal and management of hazardous waste. Most significant is the Resource Conservation and Recovery Act (RCRA), which was passed in 1978 to regulate treatment, storage, and disposal facilities. Also, the legal system may be used where parties have disputes over land use or dangerous activities.

What then, is the real reason for the EPA’s blind support of Superfund’s liability system? Perhaps it lies with the fact that it provides access to an extra-budgetary source of funding. Without the limitations of the traditional federal budgetary process, the EPA has been able to pursue costly cleanups with a significant amount of discretion and no dollar limit on the amount that can be spent at one site. As Richard Stroup of PERC notes: “So long as EPA follows the procedures it has written for itself, so that those forced to pay cannot prove that the EPA acted in an arbitrary or capricious manner, the Superfund site manager’s decisions will have the force of law. Those forced to pay have no recourse to substantive review.”19

The EPA may see the existing liability program as an unlimited source of resources, but this clearly ignores the transaction costs of this funding mechanism. One estimate suggests that the private sector spends $700 million annually raising the money necessary to fund the EPA’s program.20 And the system is horribly inefficient for raising money. A study by National Strategies, Inc., found that in 1995 the current liability system cost roughly $1 billion to raise $1.4 billion in funds for cleanups; in other words, “this means that 72 cents is spent in quantifiable public and private sector transaction costs to raise one dollar from PRPs.”21 In recent years, the return may have actually worsened, because both the amount that PRPs agree to pay for future cleanups and the amount that PRPs paid to the EPA as reimbursement for cleanup expenses fell dramatically. From Fiscal Year 1996 to Fiscal Year 1997, payment for future cleanups dropped by almost 50 percent and reimbursements dropped by 65 percent.22

Even a brief examination of the workings of the current liability system suggests reform is necessary. Superfund reform is liability reform. In fact, most reform options under discussion do contain some degree of liability reform. Proposals favored by the EPA and the administration talk of exempting small (de minimis) and very small (de micromis) parties from the liability web. Others favor removing classes of individuals, such as transporters and generators at multi-party sites, from liability. Still others are working to exempt types of sites, such as landfills and recycling sites, from liability.

The empirical data suggest that reducing the number of PRPs will accelerate cleanups. Proposals to eliminate certain sites or classes of PRPs can significantly streamline the process. One study found that if generators and transporters (who did not control disposal at sites) were exempted and the EPA focused only on owners or operators of sites, over 53,000 parties would be eliminated from the liability maze. As the number of parties involved decreases, legal fees and transaction costs decline as well.

Granting exemptions to different PRPs may be a step in the right direction, but it does not alter the fundamental structure of the program, which would still rely on CERCLA’s liability rules to raise money. Ideally, the liability system itself must change. Retroactive liability, which holds PRPs liable for actions that may have been perfectly legal (and even may have been considered environmentally sound at the time), should be repealed. Unfortunately, this approach was stymied in the 104th Congress when the EPA and environmentalists mobilized opposition against the repeal of retroactive liability.

The current debate over Superfund reform has bogged down on the question of just how much liability reform is politically feasible. Proponents of reform continue to float new ways to contain the scope of Superfund’s liability system. The EPA and others remain skeptical of such attempts and appear unwilling to move beyond exemptions for small parties. As a compromise position, some support exemptions, but only with a reach back provision to allow any party to be drawn back into the fray. Although this may offer the EPA the largest pool of potential funding, it does little to alleviate the uncertainties faced by small businesses and others.

Given the opposition facing full retroactive liability repeal, Superfund reform should examine additional options as well. Another important reform would focus on joint and several liability. The courts have interpreted CERCLA in a way that holds PRPs liable for cleanup costs well beyond their contribution. The use of joint and several liability fuels much of the litigation surrounding Superfund. Large PRPs, who may actually bear only a small portion of the responsibility at a given site, have strong incentives to file contribution suits any time they are identified as PRPs. Superfund reform should clearly indicate that liability for a cleanup cannot exceed a PRP’s contribution to the problem. This would eliminate much of the incentive to litigate while holding true to the EPA’s “polluter pays” principle. If, in fact, fairness is the reasoning behind the “polluter pays” principle, then reforming joint and several liability is an option that should be examined.

What Are We Cleaning?

In addition to the liability quagmire, the remediation process itself has been a constant source of criticism for the Superfund program. Not only are there concerns over excessive administrative costs and poor management practices, but oftentimes the cleanups are costly and bear no relation to eliminating identifiable health risks for individuals in the local community.

The process used by the EPA to determine the level of remediation at a given Superfund site begins with a risk assessment in which the agency uses a number of assumptions to determine the level of risk and the exposure of the local population. However, the EPA has typically relied on what many claim are unrealistic assumptions and conjectures about the future use of the site in question. Excessively conservative assumptions tend to overstate the risk and raise remediation costs significantly. Currently, the average cleanup costs more than $30 million.23

For example, consider the following EPA standard for cleanup, “For 350 days per year a child is assumed to eat 200 milligrams of dirt while playing in the soil at a Superfund site that may be surrounded by fences with few residences nearby.”24 Such demands for high-quality, edible dirt are made in instances where there are few residential areas and where a fence to keep trespassers out surrounds the Superfund site. Highly speculative conjectures about future use allow the EPA to demand expensive remediation programs to eliminate any potential pathways of exposure. It is often the assumption that in the future a residential neighborhood may take over an industrial site that raises the EPA’s assessment of potential health threats. A study of 77 Superfund sites found that 91 percent of the cancer risk claimed by the EPA was future risk that depended on the EPA’s assumption of future land use and behavior.25

Clearly, any meaningful reform of Superfund must address risk assessments and remediation technologies required by the EPA. Costly decisions have been made in the past because Superfund has been viewed as a cleanup program, not a risk reduction program. Decisions about which sites move forward have been made based on administrative requirements (such as the length of time a site has been on the NPL) rather than risk assessments that identify real hazards to public health.26 Sound science and more accurate risk assessments can significantly reduce the costs of remediation, while reducing real health risks when they are found. Steven J. Milloy of the National Environmental Policy Institute suggests that the costs of cleanups would fall by 60 percent if the program focused more directly on risk when identifying the appropriate remedies.27 At the same time, resources would be allocated to those sites posing the greatest health risks.

Is the Program Getting Better?

In response to proponents of a major overhaul of the Superfund program, the EPA moved forward with a number of administrative reforms to increase the pace of cleanups. The agency points to the fact that 268 sites have been cleaned up in the past four years, which is more than the previous 12 years combined. In conjunction with the administrative reforms, the EPA proposes moderate Superfund legislation, claiming that the EPA will “hold firm to the ‘polluter pays’ principle and to ensuring that money is going to cleanups, not to lawyers.”28 These goals appear to contradict each other, because it is difficult to keep the money away from lawyers when it is a liability-based system. And a closer look at the numbers suggests more than moderate reforms are necessary.

In fact, there is little data to suggest that the program is working better today than it has in the past. A GAO study of the EPA’s administrative reforms found that out of 45 reforms only 6 could be said to have quantifiable accomplishments, and only three had documented accomplishments.29 As David Aylward, president of National Strategies, Inc. (NSI), testified before Congress: ” [T]here is no evidence that the program has been significantly improved, and current data indicates that the program has worsened, at least in outputs.”30 Mr. Aylward bases this assertion on analysis of the EPA’s data conducted by NSI that provides a careful assessment of Superfund’s progress. As for the EPA’s claim that more sites are being categorized as construction complete, the NSI research finds that this is simply a function of time. Given that most sites were listed in the mid-1980s and that cleanups have been averaging 10 to 12 years to complete, it follows that they would be finished, or close to finished, in recent years. Consequently, the current pace of completion simply confirms earlier estimates that cleanups require at least 10 years.

The NSI analysis goes on to show that three other measures of accomplishment also declined from 1996 to 1997: remedial cleanups completed, remedial designs completed, and remedial actions started.31

It is not surprising that the administrative reforms have not altered the program significantly. Administrative changes at the margin do not alter the underlying principles that drive the Superfund machine. Raising funds through the liability system continues and parties drawn into a Superfund site continue to file contribution suits and drag others into the maelstrom. Legal maneuvering diverts resources away from questions of public health while delaying cleanups. Until these core issues are addressed, reform can only achieve minor successes at best.

Meaningful Superfund Reform

Superfund is in clear need of fundamental reform. Good intentions have wrought a bad program that is broken and has not served the public well. Even EPA Administrator Carol Browner admits: “[W]e’ve got to have legislation. And we are 100 percent committed to enacting responsible Superfund legislation.”32 Unfortunately, for the EPA this means focusing almost exclusively on “getting the little guys out.” Real reform must go further. Both the scope of the program and the remediation process must be recast to promote swifter and more efficient cleanups that focus on real threats to public health.

Hazardous waste sites, in fact, are local problems and should be addressed at the state and local levels. The federal National Priorities List should be capped, with future site identification and remediation conducted by state and local governments. Already, a number of states are moving forward with their own cleanup programs. Michigan, for example, employs over 400 people and has spent close to $1 billion on cleanups.33 State level programs need the flexibility to be innovative in their approach to remediation. Simply transferring the federal program — with its burdensome remediation process and costly liability system — down to the state level will do little to expedite the cleanup process. Allowing states to develop their own approach to site identification and remediation will foster cleanups that are suited to the specific circumstances of different states. In addition, states could work to privatize Superfund sites, whereby states could auction off waste sites. The price could be positive (if cleanup costs are lower than the value of the site) or negative (if cleanup costs are greater than the value of the site). With a negative price, the state would pay the bidder to clean the site. In exchange for the rights to the property, the winning bidder would post a bond that the government would hold until the health risks have been mitigated.

This would allow the Superfund program to focus exclusively on remediation at the NPL sites. Improved, risk-based remedy selection and a streamlined administrative process would reduce the costs of the remaining NPL cleanups. These savings and the finite scope of the program would eliminate the need for new taxes. This is especially true if the EPA meets its stated goal of cleaning another 400 sites by the end of the year 2000.34 If the EPA required additional resources to complete the remediation of NPLS sites, funding should be allocated from the annual budgetary appropriation to the EPA.

Remedy reform alone, however, will not eliminate the inherent inefficiencies of Superfund. Reform must address the liability system, which has proven to be a stifling, inefficient way to raise funding for Superfund cleanups. The current liability system bears little semblance to the “polluter pays” principle, nor does it provide incentives for environmental responsibility. Repealing retroactive liability would provide the greatest return in terms of eliminating transaction costs and promoting cleanups; identifying ways to minimize the applicability of retroactivity may be the second-best solution. Examples include exemptions for generators and transporters who were not bad actors and did not have responsibility over site management, exemptions for co-disposal sites (which typically have numerous PRPs and poor record keeping, making it difficult to allocate responsibility), or exemptions for small businesses. What is clear, however, is that reforms must be broader than simply exempting the “little guys.” There are too many sites where the PRPs are so numerous that even if the de micromis and de minimis were exempted, the legal wranglings of the remaining PRPs would still delay the program.

Reforming joint and several liability would also significantly reduce the legal proceedings that delay the cleanup process. Making a PRP potentially responsible for the full cleanup, no matter how slight the PRP’s involvement, generates incentives to file contribution suits and expand the pool of PRPs. Clarifying that liability means liability for a PRP’s proportionate responsibility ensures fairness while reducing the need to litigate.

Without significant reform, hazardous waste cleanups will remain mired in legal battles. Changes at the margin or new layers added to the existing Superfund program offer little hope. Superfund must be reformed substantively and structurally. States should play a greater role in cleaning up waste sites, remedy selection must be based on risk assessment and sound science, and excessive oversight costs must be reduced. The original structure of the program must be revisited, with an eye toward extricating the program from the courtroom. Until then, Superfund will remain as an example of failed environmental policy.

1 White House Briefing, February 11, 1993.

2 It should be noted that much of the concern over Love Canal might have been excessive. Studies have not conclusively linked the original chemicals in Love Canal to increased health risks. Today, after remediation and habitability studies, Love Canal has been resettled. See Chapter 4 of Aaron Wildavsky, But Is It True? A Citizens Guide to Environmental Health and Safety Issues, Harvard University Press, Cambridge, MA (1995).

3 See, “Fixing Superfund: Getting the Formula Right,” Research Brief, Institute for Civil Justice, RAND, July 1994; and Lloyd Dixon, Deborah Drezner, and James K. Hammitt, “Private-Sector Cleanup Expenditures and Transaction Costs at 18 Superfund Sites,” RAND, 1993.

4 Funding is also received from general revenues, repayments for advances on the trust fund, interest on the trust fund and penalties assessed against responsible parties at Superfund sites.

5 E. W. Coglazier, T. Cox, and K. Davis, Estimating Resource Requirments for NPL Sites, Waste Management Research and Education Institute, University of Tennessee, 1991.

6 Carol M. Browner, “Remarks Prepared for Delivery, HazWaste World Superfund XVIII Conference,” Washington, D.C., December 2, 1997.

7 See John Hird, Superfund: The Political Economy of Environmental Risk, Johns Hopkins University Press, 1994, for a more detailed overview of the Superfund program.

8 Jerry Taylor, “Salting the Earth: The Case for Repealing Superfund,” Regulation, 1995, no. 2, p. 55.

9 EPA data listed at http://www.epa.gov/superfund/oerr/siteinfo/index.htm.

10 By law, such removal actions are limited to a $2 million budget and a one-year duration.

11 EPA data listed at http://www.epa.gov/superfund/oerr/siteinfo/index.htm.

12 “Superfund: Cleanups Nearing Completion Indicate Future Challenges,” General Accounting Office, GAO/RCED-93-188, September 1993.

13 It should be noted that the EPA, since 1989, has sought to minimize the number of trust fund-financed cleanups in order to preserve the trust fund. Where possible the agency pursues PRPs through enforcement actions to fund cleanups.

14 Question and Answer Session of Superfund Reauthorization hearing before the U.S. House of Representatives, Transportation and Infrastructure Committee, Subcommittee on Water Resources and the Environment, March 12, 1997.

15 Lloyd Dixon, Fixing Superfund: The Effect of the Proposed Superfund Reform Act of 1994, Institute of Justice, RAND, Santa Monica, 1994.

16 Superfund in 1997: Costs, Results, and the Impact of Potential Reforms, National Strategies, Inc., July 1997.

17 “Slants and Trends,” Hazardous Waste News, no. 22, vol 19, November 3, 1997.

18 See, Richard L. Stroup, “Superfund: The Shortcut that Failed,” PERC Policy Series, issue no. PS-5, PERC, May 1996, for a discussion of Superfund and the polluter pays principle.

19 Ibid., p. 9.

20 National Strategies, Inc., p. 11, July 1997.

21 Ibid..

22 Testimony of David K. Aylward before the U.S. House of Representatives, Commerce Committee, Subcommittee on Finance and Hazardous Materials, February 4, 1998.

23 Lloyd Dixon, Deborah Drezner, and James K. Hammitt, “Private-Sector Cleanup Expenditures and Transaction Costs at 18 Superfund Sites,” RAND, 1993. This figure represents actual remediation costs; it does not include additional transaction costs such as legal fees and arbitration.

24 A Historical Perspective on Risk Assessment in the Federal Government, Center for Risk Analysis, Harvard School of Public Health, p. 38, March 1994.

25 W. Kip Viscusi and James T. Hamilton, “Superfund and Real Risks,” The American Enterprise, pp. 38-45, March/April 1994.

26 General Accounting Office, “Superfund Program Management,” GAO/HR 95-12, p. 11, February 1995.

27 Steven J. Milloy, Science-Based Risk Assessment: A Key to the Superfund Puzzle, National Environmental Policy Institute.

28 Testimony of Carol M. Browner before the U.S. House of Representatives, Transportation and Infrastructure Committee, Subcommittee on Water Resources and Environment, March 12, 1997. Summary available at http://www.epa.gov/superfund/oerr/whatsnew/ summo312.htm.

29 General Accounting Office, “Superfund: Information on EPA’s Administrative Reforms,” GAO/RCED-97-174R, May 30, 1997.

30 Testimony of David K. Aylward before the U.S. House of Representatives, Commerce Committee, Subcommittee on Finance and Hazardous Materials, February 4, 1998.

31 “1998 Superfund Facts: The Program Does Not Work,” National Strategies, Inc., 1998.

32 Carol M. Browner, “Remarks Prepared for Delivery, HazWaste World Superfund XVIII Conference,” Washington, D.C., December 2, 1997.

33 Testimony of Charles McIntosh, Deputy Director, Michigan Department of Environmental Quality, before the House Committee of Commerce, Subcommittee on Commerce, Trade, and Hazardous Materials, October 18, 1995.

34 Carol M. Browner, “Remarks Prepared for Delivery, HazWaste World Superfund XVIII Conference,” Washington, D.C., December 2, 1997.