The Hudson River:

A Case Study of the Superfund Policy Process

 

by Richard J. Benware

 

 

I. Objective

 

 

 

This study looks at the real life function of the Superfund policy process by examining its application to the Hudson River. First, it defines the theoretical framework. Second, the study locates the places where stakeholders have access to this process. Third, after locating these access points, the study addresses the transparency of involved parties. Fourth, it identifies any precedents that can be set within this policy framework. Fifth, it describes the advantages and disadvantages of federal cleanups compared to state cleanups. Finally, after discussing all of these results, the study prescribes amendments to the current laws and policy process which it argues would facilitate environmental remediation.

 

 

II. Introduction to Superfund and to the Hudson River

 

 

 

In 1980, Congress passed the Comprehensive Environmental Response, Compensation, and Liability Act (CERCLA or Superfund for short) into law. With a few amendments in 1986, Superfund gave the federal government new powers. First, the government was given authority to respond to releases of hazardous substances. For these responses, permanent solutions were favored over temporary solutions. Second, a National Priorities List (NPL) was established. This list contained the most hazardous sites in need of cleanup. Third, a trust fund was created which financed cleanup activities. Funds came from oil taxes, chemical taxes, and corporate environmental income taxes. Finally, when they could be identified, potentially responsible parties (PRPs) were required to pay for site cleanups. For acquiring these funds, the government was given the right to negotiate legal agreements with PRPs. With these guidelines, the government began addressing sites and adding them to the NPL. (Reisch, 1-13)

The Hudson River was one of these sites addressed under the new legislation. During the mid 1900’s, General Electric (GE) dumped polychlorinated biphenyls (PCBs) into the Hudson River. The Environmental Protection Agency (EPA) first examined the Hudson River in 1983. In 1984, after studying the river, the EPA decided to take no action. In 1989 the EPA decided to redo their assessment and the new Feasibility Study began in 1990. In this study, it was determined that PCBs are probable carcinogens which can become harmful in higher accumulations. The EPA then released a Record of Decision (ROD) where it determined that dredging the Hudson River was the optimal course of action. However, no cleaning has occurred since the ROD was released in 2002. As of 2005, GE released an Intermediate Design Report (IDR) but has not yet begun the cleanup of PCBs (Shiaffo).

Table 1.  List of Common Acronyms in this Case Study

 

CERCLA	Comprehensive Environmental Response, Compensation, and Liability Act
EPA	Environmental Protection Agency
GE	General Electric
IDR	Intermediate Design Report
NGO	Non-Government Organization
NPL	National Priorities List
PRP	Potentially Responsible Party
ROD	Record of Decision

 

 

III. Theoretical Framework for the Superfund Policy Process

 

 

In order to examine the policy processes of Superfund and the Hudson River, it is necessary to define the framework which CERCLA creates. First, before any site is placed on the NPL it must be screened. While this initial assessment is not conducted in great depth, it does allow a Hazard Ranking System Score to be given to the site. Based on this rating, a decision is made whether or not to add the site to the NPL (Introduction, 1-2).

Once a site is listed on the NPL, the EPA conducts a Feasibility Study. This study contains both government and non-government research that deals with the scientific and economic issues of the particular site. It describes all of the options that the EPA considers for handling the site. In addition, it presents all of the benefits and costs associated with each option before they makes a decision (Remedial, 1-2). Upon release of the study, the EPA accepts public comments concerning the site. After reviewing all relevant documents, the EPA may alter their decision (Public, 1).

Once a decision is made, either the EPA or the PRP begins the process of hiring contractors. Sometimes, the Department of Justice assists by conducting negotiations between involved parties and the EPA. Once contractors and subcontractors are hired, the remedial action is begun. Throughout remediation, the EPA and PRP closely monitor the site (Rosales). Data collected throughout the process is analyzed to determine if people and the environment can be protected by further remedial action. If the EPA and the state concur that further remedial action is not appropriate, then the site is deleted from the NPL (How Sites are Deleted, 1).

IV. Access Points within Framework

 

Given this framework for the policy process, the many actors and stakeholders of a site can begin to take action. The actors for Superfund are those who are part of the policy process but have nothing to gain or lose as a result. Actors typically include congress, the EPA, and NGOs. On the other hand, stakeholders’ well-being is directly changed as a result of the policy process. Typical stakeholders are the towns and residents affected by the site, the PRPs, and any species that are injured as a result of the site’s hazardous substances. Some stakeholders may also be actors. For instance, a congressman may have a large number of constituents living at or near a Superfund site. In this case, the congressman would have much to gain if the site is well handled. However, since most stakeholders are not actors, these individuals and organizations seek access to the policy process.

The most obvious stakeholder in the Superfund policy process, the public, is provided with comment periods as their primary method of access. NGOs and other groups may also participate in this process. During comment periods, the public is given access to EPA documentation of a site’s studies and reports (Public, 1). The comment periods last for a designated period of time, usually multiples of 30 days, and occur:

·        Before placing a site on the NPL

·        Between the release of the feasibility study and the ROD

·        Before the EPA accepts any agreements from the PRP

·        Before a site can be deleted from the NPL

·        Any time the EPA or PRP makes significant alterations to the proposed plan

While these comment periods give the public a great deal of access to the policy process, the Hudson River has shown that a well aggregated public can also exert pressure on the actors directly. When the prospect of dredging loomed over the Hudson Valley, the populous replied with loud disapproval. Even though there are no provisions in the theoretical process presented earlier, the EPA responded to this public backlash by establishing a local office in Hudson Falls (Rosales).

At the same time, individual communities were joining together to voice their objections. The most vocal communities were the potential hosts of dewatering facilities. Joined together behind Town Supervisor Mary Pulver and Congressman John Sweeney, Fort Edward was able to communicate with the EPA and other Federal agencies. First, monthly meetings were formed that reported on the progress and latest information available about the Hudson River (Rayball). Second, the Inter-Agency Work Group was established. This entity combines members of state and Federal agencies to help Fort Edward apply for grant money (Rayball).

While the public was the most heard stakeholder in the case of the Hudson River, the largest and most influential stakeholder has been General Electric Company. GE is the potentially responsible party for the Hudson River Superfund Site. As a private company, GE has less direct access to the policy process than the public. Still, even though they cannot make decisions directly, they have found many ways of gaining representation in the government. Therefore, by combining their political capital and financial assets, GE built sufficient political power to alter the course of the CERCLA policy process.

First, “GE spent $313.6 million on issues related to the cleanup of PCB’s in the Hudson from 1990-2005.” (Clary, 1) Looking at the payroll of GE, one possible explanation of these expenses is lobbying. And who lobbied for GE?

 

As one can see from this table (The Politics, 1), GE has employed many individuals with connections and political capital from working in Washington D.C. in the past. This expenditure on government relations, however, supposedly only cost GE at least $1.8 million (Clary, 1).

Another large chunk of change, $31.9 million, was spent on press coverage and advertising (Clary, 1). General Electric basically created an access point through media. By purchasing millions of dollars worth of television commercials, billboard ads, and full page newspaper ads, GE was able to create and alter public opinion. These ads contained pictures of backhoes scooping up contaminated river sediment and splashing it back into the water. Even though the EPA’s proposed plan called for hydraulic dredges and silt screens in the river, many residents were convinced that the dredging of the Upper Hudson River would be as messy as GE’s ads (Rosales). Once a large number of these residents were swayed, GE had effectively used the public as a tool for accessing the policy process in their favor.

A final access point that General Electric had was the court system. While CERCLA put the burden of proof on GE, there were other ways for the company to benefit by legal action. GE was able to delay cleaning for years by using their lawyers to negotiate with the EPA. While the legal fees totaled $32.2 million (Clary, 1), the profits GE earned by not cleaning the Hudson, and instead investing that savings, surely outweighed these costs.

One final access point to recognize is that of individual Congressmen and Senators. These stakeholders are still actors, but do not have the individual influence needed to alter the course of a given Superfund site’s progress. However, in many cases, they can use committees to add riders to bills. One instance, was when Congressman Charles Taylor of North Carolina added an appropriations rider requiring an additional year of studies relating to the Hudson River dredging project (Wojnar). This rider was defeated, but goes to show that individuals in Congress can make a difference.

V. Transparency Under Superfund

 

While the Administration Procedures Act (APA) legally binds the EPA to disclose their actions to the public, General Electric and other PRPs are not under any obligations at all. These companies get protected from the public’s eye while negotiating with the EPA and the site’s Trustees.

In the past, this has lead to very difficult analyses of how efficiently Superfund is working (Probst, interview). One conclusion that has been discovered is that private cleanups tend to cost less than EPA funded cleanups (Probst, Fullerton, Litan, Portney, 39-53). A consequence of this is the tradeoff between transparency and efficiency. If the government organizes and runs the cleanup of a given Superfund site, the process will be very transparent to the public, but will cost significantly more. Conversely, if the PRP runs the cleanup, there will be virtually no disclosure of information, but in an effort to keep profits high, the PRP will take as many actions as possible to cut costs. Thus, we discover that society faces the tradeoff of whether to have an open cleanup or a cheap cleanup.

Offering an opposing view, Resources For the Future (RFF) finds that the amount of transactions costs saved by having federally funded cleanups would easily outweigh the lost savings that private firms would have created by conducting the cleanups (Probst, Fullerton, Litan, Portney, 47). If this is the case, then this solution may be effective at increasing both efficiency and transparency. One thing to note is that for the United States Government to implement more cleanings, the EPA would need to spend more money. This is highly unlikely because the CERCLA taxes expired in 1995. When they weren’t reinstated during subsequent years, the balance in the fund itself quickly diminished to nothing, and the amount of money allocated to the EPA’s Superfund budget decreased substantially (McCarthy, 3-10).

 

VI. Precedents for Future Sites

 

The Hudson River Superfund Site has created, and will create, many precedents for future Superfund sites. These precedents occur for two reasons. First, there are several other PCB-laden rivers, like the Fox River, which are not as far along in the policy process. Their feasibility studies, PRP negotiations, and remediations will reflect the Hudson River successes as well as failures (Georges). Second, in an incrementalist government such as the United States, there is a strong tendency to retain the status quo. Due to the national attention which the Hudson River has received, many people consider the Hudson River the posterchild of the EPA (Rosales). Therefore, there will be little tendency to deviate from the actions taken during the Hudson River project. For these reasons, any standards that are set will more than likely be repeated in future EPA projects.

One of the most likely precedents set by the Hudson River will be the higher quality of life standards. Due to the extent of the dredging, the resuspension risks, and the number of people affected, there was a lot of unrest over the proposed cleanup. To assuage these fears, the EPA presented the residents of the Hudson Valley with a list of items which they will require throughout the dredging project. This list included resuspension rates, noise pollution levels, light pollution levels, and various other restrictions (Rosales).

An additional precedent that could affect communities near other Superfund sites is the compensation received in Fort Edward. According to Friends Of a Clean Hudson, KLIOS, Inc. discovered the following when analyzing the Hudson River site:

“ * Of the $460 million in cleanup projected expenses (which are the responsibility of the polluter, the General Electric Company), nearly half - $225 million - would be expended locally.

 

* 3,543 jobs, with $88 million in wages, would be added to the local economy by the initial project – i.e. the construction, operation, and monitoring components.

 

* 1,028 indirect jobs, paying $53 million in wages, would be created by the multiplier effect, the ripple of economic activity generated by the exchange of project-related goods and services.

 

* Between 3,700 and 8,900 jobs, bringing with them payrolls totaling between $144 and $346 million, would be created in the long term. These projected employment numbers were reached by studying the twelve affected counties for the potential in waterfront activities that would resume because of a cleaner river.

 

* 300 direct and 150 indirect jobs, worth $18 million in wages, would likely result from the restoration of commercial fisheries wiped out by PCB contamination 25 years ago.

 

* The larges beneficiaries would be Washington and Saratoga Counties, the locales contiguous to the actual cleanup project. The economies of these two upriver communities would realize a windfall of an estimated $800 million in near and long term.” (Trieste, 1).

 

Given all of these benefits, it would seem that the Town of Fort Edward doesn’t have a need for compensation at all. Nevertheless, Bill Rayball of Congressman Sweeney’s staff claimed, “…Rep. Sweeney has worked hard to provide federal funds to certain high-priority projects in the Town of Fort Edward.” He stated that these, “… will help offset some of the financial burden the Superfund site places on the local economy.” (Rayball)

So what is Fort Edward gaining for hosting the EPA’s dewatering facility? They are receiving $4.73 million. This includes funds for the Fort Edward Historic Train Station, Water District #2, the restoration of the Route 4 corridor, and additional study of PCB’s effects in groundwater (Rayball). Whether these funds are necessary or not may be debatable, but Sweeney was crucial in helping his constituents receive them. The fact that constituents can access their representatives in this way will no doubt lead other municipalities to request similar compensation in the future.

Finally, legal precedents may also be set. For instance, the agreement of GE to pay for the first portion of remediation gives the EPA something to cite in any future trials. The sequence of events that occurs during the second portion of remediation, however, will prove more interesting. First, there is a chance that GE may choose to do the second portion of the clean up. Due to the size and expense of the project, merely agreeing to clean can itself set a precedent for any smaller companies to follow suit. On the other hand, GE may choose not to immediately begin the cleanup described in the second phase. The EPA’s decision of whether or not to pursue legal action, and how hard they are going to fight in such an action, could set legal precedents that will encourage PRPs to agree to remediation in the future.

 

VII. Successes and Failures of Federal Cleanups

 

The first reason for preferring federal cleanups to state cleanups is that states weren’t acting in many cases. On example is that of TCE contamination in New York State. In Endicott, NY there was a large amount of TCE contamination. The state government did nothing about the site until the EPA got involved and studied the effects of TCE on the residents (Georges). However, there are thousands of TCE sites in the United States (Trichloroethylene, 1). With so many sites, it is probably just not possible for them all to make it on state agendas. Congresswoman Sue Kelly said that her constituents, “… want a clear national standard for addressing TCE contamination and they want it now.” Given the states’ limitations, it makes sense for the EPA to step in and take charge.

One success of the EPA Superfund program is that large sites like the Hudson River can procure enough funding for remediation.  If this and other megasites were left to the states, they might never be addressed.  New YorkState remediation of the Hudson River can be used as an example.  While the estimated cost of remediation for the Hudson River is over half a billion dollars, the New York State Department of Environmental Conservation had a 2004-2005 fiscal year budget of only $564,405,000 (Cape, 78).  That means that if the only New YorkState site in need of environmental remediation was the Hudson River, New York would still find it too difficult to fund.  However, there are 86 Superfund sites in the state, not just one (NPL Sites in New York, 1).  Thus, for residents to have any hope of environmental remediation, the EPA and the Federal Government must maintain an active role with the Superfund program.

Another success of Superfund is that the Federal Government has historically not been as susceptible to industry as states (Probst, interview).  That is not to say that it is completely unsusceptible, just that on average states are more easily manipulated in their environmental decisions.  Assuming that all states can afford environmental remediation and choose to partake in the maximum amount, one state could improve their economic situation by lowering their environmental standards.  Why is this true?  If businesses can move from state to state without incurring any travel costs, then by one state offering looser environmental restrictions than another, disequilibrium is created.  Since businesses are in search of higher profits, they would prefer to spend less money on environmental protection, and would move to a state with looser restrictions to cut costs.  Thus, with respect to environmental controls, it is constructive for the Federal Government to be less responsive to businesses.

However, the opposite can be said of their responsiveness to constituents.  At the state level, employees and representatives respond very quickly to unrest from their constituents because those living near a particular Superfund site make up a large portion of voters in that election district.  However, at the level of federal organizations, there is slightly less voter sensitivity for the simple fact that those living near a particular Superfund site do not make up a significant portion of the voters in the country (Georges).  While this insensitivity is more characteristic of the Federal Government than state governments, the ease of access to the EPA’s Superfund process allows the EPA to partially overcome this trend.

A third success that the EPA has with Superfund is their ability to negotiate with PRPs.  While states have attorney generals and other legal staff, the burden of proof often rests in their hands.  Thus, PRPs would have a significant amount of bargaining power.  However, for federal Superfund sites, the DOJ does not have the burden of proof (Reisch, 7).  This is due to the wording of CERCLA itself.  By putting the burden of proof into the suing party’s hands, it encourages PRPs to negotiate before attempting legal action.

In addition, the DOJ has arrived at settlements that would otherwise not have been possible.  In the case of the Montrose Settlements Restoration Program, the EPA was able to sue the PRPs jointly with the site’s trustees (Montrose, 1).  Since the Superfund process usually has separate settlements for the EPA and the trustees, this action must have significantly cut back on transactions costs. 

While Resources for the Future does make a good case that there are transactions costs to negotiating the cleanup of Superfund sites (Probst, Fullerton, Litan, Portney, 37-38), these costs are only a fraction of the costs resulting from these negotiations. While most economists make the argument for the option value of delaying a decision, consider the option cost of delaying that decision. One must keep in mind that lost benefits are also a cost. Consider a yearly amount of benefit βi which people will enjoy from the cleanup of a Superfund site in year i. The total present value of benefits would be:

β0 + ρ*β1 + ρ^2*β2 + …

In this equation, ρ is the discount factor equal to 1/(1+δ), where δ is the discount rate. However, if preferences and relative prices do not change, then one can assume that β0 = β1 = β2 = … = β. Thus the total present value of benefits becomes:

β + ρ*β + ρ^2*β + …

Now let us assume that the cost of cleaning the same Superfund site is a one-time cost in year zero designated by C. Since the benefits will not be felt until cleanup is complete, the net present value will most likely take the form:

–C + ρ*β + ρ^2*β + ρ^3*β + …

or

(1/ δ)*β - C

Now assume that the cleanup of the Superfund site is delayed for one year by the PRP.  For a single year in the United States it is reasonable to assume that the interest rate will be close to the rate of inflation.  If the money saved by not cleaning up in year zero is invested at interest rate i then the PRP will have a sum of money C*i to spend on cleanup the following year.  Similarly, at a rate of inflation π, the goods and services that would have been purchased in time zero for cleanup will cost C*π the following year.  Since it is assumed that i ~ π, then C*i ~ C*π.  From the perspective of people in year zero, the result is that the societal cost of cleaning the Superfund site would not differ if the cleaning occurred now or later.

While the costs associated with cleaning do not change significantly due to delays, the same cannot be said of benefits.  Consider the initial net benefit stream:

–C + ρ*β + ρ^2*β + ρ^3*β + …

Because benefits cannot be conveyed until cleanup is complete, and the cleanup is delayed for one year, the benefits that would have occurred the year after the site was cleaned are now lost. Thus, this stream becomes:

–C+ 0 + ρ^2*β + ρ^3*β + …

or

ρ*(1/ δ)*β - C

Even at a small discount rate of .05, this means that delaying cleanup of a Superfund site by just one year would decrease the present value of benefits by about 5%.  Using the Hudson River Superfund Site as an example, these lost benefits would be approximately 5% of the benefits in the contingent valuation.  Since access to the Hudson River contingent valuation has not been extended to the public, the only thing that is known is that the projected benefits are greater than the projected costs. For simplification, assume that the benefits and the costs were equal at the time the decision to dredge was made. Thus:

(1/ δ)*β > C

becomes

(1/ δ)*β = C

Using the dredging cost C as the approximation for the total benefits of dredging, (1/ δ)*β, one can delay cleaning by a year and see what happens to society’s estimated benefits:

(1/ δ)*β – C à C - C

becomes

ρ*(1/ δ)*β – C  àρ*C - C

Since the Hudson River is estimated to cost over half a billion dollars, this would imply a loss of at least $25,000,000. One must keep in mind that this is with a very low discount rate. If this rate were high, say .10, then the loss in present value benefits would exceed a staggering $50,000,000.

Now consider the case where cleanup was delayed by more than one year. The Hudson River Record of Decision was released in February 2002 and at the time this paper is being written, the spring of 2006, cleanup has not yet begun. The cleaning is currently scheduled to begin in the spring of 2007. If remediation is not delayed beyond this date, it will have been just over 5 years that the project was delayed. Using the same math as above over a five-year period, the unrealized benefits of the Hudson River Superfund site add up rapidly. Even assuming a minimal discount rate of .05, society has lost benefits exceeding $100,000,000 from the Hudson River Superfund Site alone. Thus, delays are probably the most significant failure in the EPA’s entire Superfund program.

 

VIII. Discussion of Results

There are five important results discovered in this case study. The first result is that many parties have access to the Superfund Policy Process. There are many access points for each of these parties, and efforts are made by the EPA and other government agencies to make the process as transparent and participatory as possible.

Not surprisingly, transparency costs money. If PRPs no longer had to clean Superfund sites, then the public would have much better documentation and involvement, however, they would also have to pay more. This increased cost is likely to be born by the tax payers because the original Superfund taxes have not been renewed since their expiration in 1995.

The third result is the ability to set precedents under the current framework. Once an action is taken at a given Superfund site, it is very likely going to be duplicated for future sites. This happens because the stakeholders of a particular site do not want anything less than their counterparts received at another site.

One of the more significant results is that the United States Government can handle these sites better through the EPA than states can through their own agencies. The federal government is less susceptible to businesses, has more funds to draw from, has more legal power at its back, and can fit more onto its agenda than the sum of the individual states.

Finally, and most importantly, the policy process has lead to a significant deadweight loss. This loss occurs by the delays caused by the government, the PRPs, and a number of other factors. While some of these delays may be unavoidable, they could be costing taxpayers billions of dollars a year in lost benefits.

 

IX. Prescription of Future Changes

Given the results discussed above, I believe a few changes could significantly improve the current Superfund process. The first and easiest change is requiring PRPs to disclose information on their cleanup costs, legal costs, cleanup progress, and any other related information. The EPA could easily force this by requiring such disclosure in the consent decree, or any other legally binding contract made with a PRP. Since businesses try to operate at the least possible cost, it is feasible that they would be able to clean and report for less than it would cost the EPA. Thus, this would encourage the innovation of cost saving methods that increase the efficiency of remediation.

Another potential improvement for the program would be to create incentives that expedite cleanups. Since the interest rate that companies can earn from investing their money is usually higher than the rate of inflation, PRPs can actually earn money by delaying the date at which they begin remediation. However, while these companies are being made better off, society as a whole is being made much worse off. To reach the socially efficient outcome, the EPA could use either positive incentives or negative incentives. One possible negative incentive in this setting might be taxing PRPs that do not begin cleaning by a date given in the ROD. However, this would likely create great unrest in the private sector because businesses hate new taxes. In addition, it could increase legal costs because companies may attempt to fight the tax. However, by taxing, a double dividend might result. This is because any tax revenue collected could be used to replace funds obtained through the government’s general revenues. This extra income could in turn be used to lower taxes on other goods or to fund some other project.

Alternatively, a positive incentive option might be to actually give these companies money if they begin remediation by the given date. This could be done through tax credits, direct subsidy of the project, or other waived costs and fees. For this alternative, there is unlikely to be any angst among businesses because they would be saving money, not spending it. It is also unlikely that companies would devote as much money towards fighting the remedial projects because by doing so they would not only be spending money to fight the projects, but they would also be surrendering their potential monetary incentives. Thus, such a program could lessen or even eliminate the gaps between the release of the ROD and launch of remediation.

The only problem with this type of positive incentive is that it requires federal funding. Thus, my final prescription is that Congress and the President either vote to resume some sort of tax for Superfund or increase the appropriations that are given to the EPA for Superfund. When choosing the type of tax to implement, the effects on consumers should be examined. The old Superfund taxes were collected in such a way that companies could transfer the additional costs to consumers (Probst, Fullerton, Litan, Portney, 6-8). Since the main idea behind CERCLA was that the polluter pays, it only makes sense that a tax collected for its implementation is garnered from the polluters, and not from the citizens. Because the current administration values reducing taxes, I highly doubt that the either current Congress or the current President will tax any polluters, even if Exxon Mobile recently became “the first U.S. company ever to ring up quarterly sales of $100 billion” (Quinn, 1). So for the time being, it appears that neither incentives program will be feasible. However, in order to maximize the welfare of our country in the future, the framework that CERCLA has created will have to be altered and somehow improved.