33 ELR 10559 | Environmental Law Reporter | copyright © 2003 | All rights reserved

Improving State Environmental Enforcement Performance Through Enhanced Government Accountability and Other Strategies

Clifford Rechtschaffen is a professor at Golden Gate University School of Law. He is also co-director of Golden Gate's Environmental Law and Justice Clinic. David L. Markell is the Steven M. Goldstein Professor of Law at Florida State University College of Law. This Article is derived from CLIFFORD RECHTSCHAFFEN & DAVID L. MARKELL, REINVENTING ENVIRONMENTAL ENFORCEMENT & THE STATE/FEDERAL RELATIONSHIP (Envtl. L. Inst. 2003). The book may be ordered or additional information obtained by visiting http://www.elistore.org. Portions of this Article first appeared in David L. Markell, The Role of Deterrence-Based Enforcement in a "Reinvented" State/Federal Relationship: The Divide Between Theory and Reality, 24 HARV. ENVTL. L. REV. 1 (2000), and are reprinted with permission.

[33 ELR 10559]

Modern federal environmental law has a built-in dynamic of tension that since its inception has led to challenging relations between the U.S. Environmental Protection Agency (EPA) and the states. On the one hand, the U.S. Congress enacted numerous environmental statutes in the early 1970s because of widespread dissatisfaction with state efforts to protect the environment. Congress made EPA ultimately accountable for achieving the mandates of federal law. On the other hand, Congress contemplated that states would actually implement and enforce the requirements of these statutes, subject to federal oversight. To a large extent, its vision has been realized; states now administer over 75% of the major federal delegable environmental programs,¹ and conduct the great majority of environmental enforcement in this country. But as Prof. David Hodas notes, "heavy reliance on state enforcement is a double-edged sword. When we 'deputize' the states to implement national environmental laws, we shift the government's discretionary enforcement power to state and local officials, who may not be interested in, or able to carry out, federal goals."²

The tension characterizing the federal/state enforcement relationship has, if anything, intensified in recent years, particularly during the 1990s. Many state programs have grown and become more professional over the past two or three decades. With this maturation, state agencies are increasingly feeling their oats. They have become dissatisfied with a role that they perceive as EPA "dependents" or "subordinates," and increasingly insist on changing the terms of this relationship to one of "partners." With the emergence in the 1990s of the Environmental Council of States (ECOS), the influential lobbying group representing the environmental commissioners of 51 of the 55 states and territories, in particular, state environmental agencies enjoy a new level of influence in Washington, D.C. State calls for devolving more environmental authority have met with sympathetic receptions in both Congress and the executive branch.

One of the central sources of tension between EPA and the states derives from their competing visions of enforcement. Many states disagree with EPA's traditional vision that deterrence-based enforcement must be a central part of a government enforcement and compliance program. They contend, instead, that more flexible and regulated party-friendly approaches are superior. They argue that carrots work better than sticks, and that states should be given more flexibility to explore various strategies to promote compliance, rather than be locked into EPA's approach. There has been an especially lively debate over the past 5 to 10 years over which set of strategies works better to achieve compliance. (This debate extends well beyond simply EPA and the states.) To some extent, the distinctions between pure "deterrence-based" and "cooperation-based" models have been blurring in recent years. EPA, and some states, have been increasingly utilizing tools from both schools of thought in a more integrated and strategic fashion, experimenting with combinations of targeted enforcement, compliance assistance, and compliance incentive programs. Nonetheless, pronounced differences in enforcement philosophy continue to exist.

As a host of studies indicate, the performance of many state enforcement programs falls substantially short of EPA's expectations, which traditionally have been based on a deterrence-oriented model of enforcement. Many states do not adhere to EPA guidelines for how an enforcement program should be run, e.g., guidelines governing when violations are sufficiently serious to justify an enforcement action, and the types of sanctions that should be imposed on significant violators. The significant gap between state performance and EPA expectations is troubling for at least three important and related reasons. One, rates of noncompliance with environmental laws in many states remain significant.³ [33 ELR 10560] Two, the gap is symptomatic of a lack of accountability that underlies the actual administration of the environmental protection regulatory infrastructure. This is particularly a cause for concern when a number of empirical studies and experience have found that deterrence-based enforcement works to promote compliance and reduce pollution, but some states have essentially opted not to conduct deterrence-based enforcement and either invested a lower level of resources in promoting compliance than EPA believes are needed, or are emphasizing different approaches without documenting or, in some cases even seriously trying to document, that these are producing desired results. Finally, while some variation among state enforcement performance and approach is inevitable under the system of cooperative federalism that characterizes federal environmental law, when state capabilities and performance vary too dramatically, they threaten to undermine the balance between the interests of consistency and state autonomy that Congress struck in enacting the environmental statutes. Among other things, the differences may create an unequal playing field for regulated entities, enabling some facilities to gain a competitive advantage through noncompliance or less rigorous compliance in states with weak enforcement practices.⁴

Numerous factors contribute to the substantial divide between state performance and EPA expectations. They include the differences between EPA and the states in their enforcement philosophies, limitations in state legal authorities and resources, and weaknesses in the data systems of EPA and the states that have complicated efforts to monitor state performance. EPA resource constraints have limited EPA's practical ability to influence states. EPA also has delivered an ambiguous message to the states—and to its own regions—about federal expectations. In addition, the gap signals flaws in EPA's system of oversight, either in theory or how it has been implemented in practice. Among other things, the threat of EPA withdrawing authority for poorly performing states, the ultimate sanction that in theory should motivate state conformance to EPA policy, has proven to be largely hollow, for political and resource reasons. As Prof. William Rodgers notes, "authorization is very much a one-way street that is granted but almost never withdrawn."⁵ Other oversight tools, such as overfiling, also have not been particularly effective. There also has been significant criticism from a number of quarters about the traditional criteria that EPA has used to evaluate state enforcement programs, largely focused on agency enforcement activities. Thus, the nature of EPA oversight of state performance warrants special attention. Invigoration, in the form of new approaches to oversight, and reworking of some existing approaches, is needed.

The creation of the National Environmental Performance Partnership System (NEPPS) in 1995 represented an effort to abate the tension between EPA and the states, and also, in theory, an approach that can end the gap between state performance and EPA expectations by making the states more accountable. At its core, NEPPS promised to let states be judged by bottomline performance—how well the states improve environmental conditions and achieve compliance. As originally conceived, NEPPS envisioned a system of differential oversight, rewarding better performing states with greater autonomy and flexibility, and even conferring a "leadership" status for exceptionally performing states.⁶

Thus far, NEPPS has not lived up to its promise, including in the enforcement/compliance area.⁷ Relatively few states have developed and implemented the core performance measures that would allow EPA to judge them by bottomline performance.⁸ There has been relatively little reduction in EPA oversight of states, and EPA has dropped the concept of differential oversight, despite its potential for holding states accountable for bottomline performance and strengthening incentives for states to improve their performance.⁹

NEPPS may still prove to be an effective vehicle for encouraging state performance and improving the state/federal relationship, and it is likely to enjoy strong support from [33 ELR 10561] the current Administration. But NEPPS also may very well complicate EPA's efforts to have states meet its enforcement expectations, if it results in EPA granting greater enforcement autonomy to the states before they have demonstrated bottomline success in achieving compliance and improving the environment or, even, a commitment and the capacity to show how well they are doing. And it also could undermine rather than enhance state accountability if it reduces EPA's ability to monitor state performance.

This Article discusses a number of options for EPA to strengthen state performance and bring it more in line with EPA's expectations. First, EPA must play a stricter gate-keeping function in initially authorizing state programs, and more regularly reassess and report the adequacy of state enforcement authorities and state capacity. Second, EPA must stop delivering a mixed message to the states about the enforcement practices it expects the states to follow. Instead, it must establish clear expectations for performance. Third, in terms of the substance of those expectations, EPA should revise its criteria for evaluating whether state enforcement programs work. Because this question is central to EPA's oversight practices, we discuss it at length in this Article. Among other things, we recommend that EPA place greater emphasis on measures based on how well states are doing in actually achieving compliance by regulated entities, and, to the extent feasible, improved environmental outcomes, although we also believe that continued reliance on activity-based measures is appropriate. Fourth, EPA should explore different administrative approaches for conducting program evaluations. Fifth, and related, EPA should improve its data management programs to enhance its ability to evaluate state programs. Sixth, EPA should implement a "differential oversight" scheme that rewards in a meaningful fashion better performing states. Seventh, EPA should publicly report its evaluation of state programs, and also improve public accessibility to the underlying data about state performance, as a way of using the power of an external spotlight to spur improved government performance. Eighth, EPA should make the prospect of program withdrawal a more credible threat and should also develop more graduated sanctions short of outright program withdrawal. Finally, EPA needs to continue to hone its capacity for strategic intervention—to help states to build capacity, and to take direct action in selected cases when needed. These options are not mutually exclusive; instead, they can and must be integrated in a variety of ways.

EPA Should Play a Stricter Gatekeeper Role in Authorizing State Programs

In enacting the federal environmental regulatory infrastructure, Congress reserved the role of primary program implementers for the states, but only on the condition that the states demonstrate to EPA the capacity to fill this role. In performing its authorization responsibilities, EPA has sought to strike a balance between national consistency and deference to state autonomy interests by requiring states to show certain types of capacity but insisting on less than functional equivalence with federal authority or capacity. In fact, EPA has authorized states to operate programs even though the states' legal authorities created significant doubts about their ability to meet EPA expectations, such as their ability to recover economic benefit realized by noncompliance.¹⁰ Critics have pointed out that as part of EPA's authorization process, "staffing levels, the implications of weak administrative structures, the enforcement impacts of review boards and other institutions, and the effects of state procedures are not always thoroughly understood."¹¹ Compounding the present situation is that once authorization has been granted, EPA does not routinely reevaluate the adequacy of state legal authorities or the level of state resources committed to implementing the state program.

In part, EPA's failure to take a harder look at state programs reflects political reality; as Professor Rodgers argues, "there are psychological inhibitions among EPA staffers who have a limited number of 'no's' [sic] in their repertoires…. Refusals over time to give the reins to states willing to assume them also tend to become uncomfortable liabilities."¹² While mindful of Professor Rodgers' assessment, we recommend that EPA pay greater attention to determining enforcement capabilities in making initial decisions about whether to approve state program authorization. It is far harder, politically and otherwise, to withdraw a program once authorization has been approved. Rather, EPA should insist that its existing authorization requirements are honestly met by the states in the first instance.

Thus, in our view, at a minimum, states that lack the legal authority to recoup the economic benefit of significant violations should not be authorized in the first place. Recovery of economic benefit has long been a cornerstone element of EPA's enforcement policies,¹³ and indeed, is a central component of deterrence-based enforcement. If regulated entities believe that because of limits in a state's enforcement authority they will realize savings, or competitive advantages, from violating the law, they have little incentive to comply.

Likewise, EPA should not authorize states that lack administrative penalty authority. Such authority allows agencies to impose penalties quickly and with a minimum of resources, and, as a result, Congress and many others have recognized it is a key ingredient of an effective enforcement program. States which lack such authority are often hindered in their ability to collect penalties from violators. Similarly, EPA should not—as it has done in the past—approve state programs that have significant preconditions to the initiation of enforcement actions (such as elaborate civil referral processes), or other cumbersome requirements that reduce the effectiveness of enforcement activity.¹⁴ (It should [33 ELR 10562] be noted that under EPA's current authorization regulations, recovery of economic benefit and administrative penalty authority are not expressly required as a condition of authorization.¹⁵ EPA, however, has the authority to require that authorized state programs include these elements. The environmental statutes speak in general terms about adequate enforcement authority being a condition of authorization,¹⁶ and, the courts have accorded EPA substantial discretion in defining what this means.¹⁷ The Clean Air Act (CAA), moreover, specifically requires that penalties assessed by state programs against stationary sources include the economic benefit of noncompliance.)¹⁸

Moreover—especially since at this point over 75% of the major environmental programs that can be delegated are administered by the states—frequent reassessment of the extent to which individual states meet the threshold authorization criteria should be a priority. Thus, EPA annually should reassess the state legal authorities that it believes are necessary to implement a federal program to determine if they have been eroded by judicial decisions, Attorney General opinions, legislative changes, etc. EPA's Safe Drinking Water Act (SDWA) regulations require this type of review, mandating an annual assessment of state compliance with "primacy standards,"¹⁹ but other statutes apparently do not. EPA's review of state audit privilege/immunity legislation in the mid-1990s is an important example of the type of ongoing review of state authorities that we have in mind, but by itself is too limited, focused just on this one aspect of state laws. We recommend, moreover, that Congress require EPA to prepare an annual public report tracking state environmental authorities in all states with authorized programs, including enforcement/compliance authorities.²⁰

Similarly, EPA annually should satisfy itself that authorized states retain the other types of capacity needed to perform their enforcement responsibilities adequately, such as financial support, personnel resources, and the like. EPA also should publicly report its findings on these matters. While states generally have made significant strides in recent years in upgrading their capacity, the inevitable ebbs and flows in economic conditions, legislative priorities, gubernatorial politics, and other developments makes ongoing evaluation important. For example, during the 1990s a number of states shifted personnel and resources from traditional enforcement activities to compliance assistance and other programs. Does this leave them without adequate enforcement capacity to carry out the core functions deemed important by EPA? Likewise, the downturn in the economy beginning in 2000 has left many states with significant budgetary shortfalls. Has this resulted in programs operating with inadequate capacity? An official with ECOS estimated in early 2002 that the nation's economic slowdown had forced states to cut environmental spending by $ 200 million nationwide.²¹ Indeed, in an extreme (and unusual) effort to draw attention to resource shortfalls within his state, which had dropped it to last in the United States in both per capita and per industry spending on environmental issues, in 1993 the governor of Indiana commenced the process of voluntarily returning to EPA its federally delegated Resource Conservation and Recovery Act (RCRA) and Clean Water Act (CWA) permitting programs because of the Indiana Legislature's failure to fund those programs adequately.²² Likewise, in 1993, Connecticut and Massachusetts almost returned primacy for implementing the SDWA to EPA after their state budgets eliminated funding for the state-authorized programs, and a recent report indicates that Missouri officials have asked EPA Region 7 to take back control of the state's delegated CWA § 401 certification program because of a lack of resources.²³

EPA Region 5's 2001 review of Ohio's environmental programs is instructive in this regard. The agency found that

one of U.S. EPA's biggest concerns about Ohio's air enforcement program is the adequacy of the resources provided to the agency. When OEPA [Ohio Environmental Protection Agency] first submitted its Title V program for approval in 1992, it estimated that DAPC [Division of Air Pollution Control] would need 399 employees to run all of its air programs once the Title V program was in place. As of December 4, 2000, DAPC only had 222 employees on staff…. In other words, OEPA may lack sufficient resources to perform all of its permitting and enforcement duties.²⁴

Notably, EPA's review of Ohio's program, discussed below, was undertaken in response to citizen withdrawal petitions filed with EPA, not on EPA's own initiative. EPA might find similarly stark resource shortfalls if it periodically reviewed agency capacity.

EPA, on occasion, has demonstrated the backbone to take a tougher approach to authorization. In recent years, for instance, it has delayed authorization of new statutory programs, such as the CAA's Title V permitting program, to a [33 ELR 10563] number of states because of unresolved concerns about state audit privilege and immunity laws or other deficiencies in the state programs. Thus, for example, in 1996 EPA granted Maryland only interim, rather than final, approval to implement the Title V program, because Maryland law did not give nonstate residents or businesses the right to challenge agency permit decisions judicially, which contravenes the public participation requirements of the Title V program.²⁵ After several extensions, and after a lawsuit by environmental groups forced EPA's hand, in December 2001, EPA withdrew its interim authorization and assumed control of the Title V program in Maryland.²⁶ Even this action is less than it appears, because EPA contracted with Maryland's Department of the Environment to continue to actually administer the program (on the assumption—which ultimately proved correct—that Maryland's Legislature would act quickly to remedy the flaw in its standing law and resolve EPA's concerns). However, federal, rather than state, permitting regulations provided the applicable law, and EPA, not Maryland, had authority over permit decisions.²⁷ In early 2003, after the Maryland Legislature had changed its standing law, EPA granted final full approval of the state's Title V program, and returned control to the state.²⁸

Unfortunately, this level of close EPA scrutiny of state programs' legal authority and capacity does not occur on a systematic basis, but we recommend that it should.

EPA Needs to Deliver a Clearer Message About Its Expectations

EPA's oversight of the states from the mid-1990s to the present has been characterized by conflicting themes. On the one hand, EPA has sought to give the states more leeway to pursue cooperation-based enforcement strategies, and embarked on a new, more flexible oversight system, the NEPPS. On the other hand, EPA has maintained that states should continue to follow all of its deterrence-based enforcement policies, and that the Revised Policy Framework for State/EPA Enforcement Agreements (Policy Framework), described below, still governs oversight of state programs.²⁹ Moreover, parts of EPA have appeared to encourage states to experiment with alternative compliance techniques, while other parts have criticized the states for declining enforcement activities and other departures from EPA's enforcement policies. In short, EPA has tried to have it both ways, and, unsurprisingly, states, as well as EPA regional offices, report receiving a mixed message and confusing signals. The lack of central guidance from EPA in 1995 regarding implementation of the NEPPS has exacerbated the mixed message received by states and regions. A recent evaluation notes that

no common understanding exists regarding the role of federal oversight—and there is consequently uncertain scope for flexibility and the bounds of acceptable over-sight. This point is perhaps the most serious. EPA intentionally implemented NEPPS without detailed, prescriptive program structure or guidance. This decision on EPA's part was intended to allow states and regional offices the opportunity to tailor their NEPPS efforts to their needs. There is a clear downside, however: we observed tremendous uncertainty in EPA regional offices and in state agencies over the nature and degree of flexibility that EPA can grant to states under NEPPS and over what level and type of accountability EPA must impose on states and programs.³⁰

EPA's ambivalent approach is driven by clear political imperatives. Many observers believe that permitting state deviations from EPA's deterrence-based policies is the price it must pay for state cooperation. Prof. John Dwyer points out, for example, that "the federal government cannot implement its air pollution program without the substantial resources, expertise, information, and political support of state and local officials."³¹ Moreover, in the 1990s, EPA granted states more flexibility (and indeed, embarked on a number of "reinvention" efforts) as a way to deflect political pressure for more radical devolution or even deregulation of its authority. But it will be very difficult to secure significant improvements in state performance until EPA adopts a less schizophrenic and more internally consistent message, and delivers this message more clearly to the states and regional offices. Consistent with the analysis and recommendations of the book from which this Article is excerpted,³² we argue that this message should be that the states should follow an enforcement strategy firmly grounded in deterrence-based theory, but that also incorporates cooperation-based elements such as compliance incentives and compliance assistance programs. In addition, as discussed below, newer performance measures should be utilized as a basis for evaluating state enforcement programs, so long as they provide for sufficient state accountability. In our view, this enforcement approach will best achieve more widespread compliance with our environmental laws.

[33 ELR 10564]

EPA Should Revise Its Criteria for Evaluating the Efficacy of State Enforcement and Compliance Programs

As noted above, an issue at the heart of the federal/state environmental enforcement relationship is how to evaluate whether enforcement programs work. Broadly speaking, there are three types of performance measures used to evaluate environmental programs: environmental outcome indicators (sometimes called "terminal outcomes"³³), such as measures of air and water quality; behavioral outcomes (sometimes referred to as "intermediate outcomes," or "program outcomes"), which reflect changes in the behavior of regulated entities in response to government actions, such as compliance rates; and outputs, or measures of agency activity, such as the number of inspections carried out or enforcement actions filed. Factors such as agency expenditures, staffing levels, and enforcement authorities, sometimes referred to as "inputs," are also relevant. EPA's traditional approach for judging state performance, embodied in the Agency's Policy Framework, focuses largely on the last broad category of performance measures, agency outputs.³⁴ Critics argue that these evaluation measures result in a pre-occupation with numerical outputs rather than the bottom-line success of enforcement efforts. State officials have contended that newer, more results-oriented performance measures are necessary to fully capture the impacts of cooperative-based approaches, which do not rely on the traditional tools measured by EPA. In recent years, EPA and the states have been trying to fashion new evaluative criteria for judging enforcement programs. However, many of these measures have not yet been implemented, and many states have not been willing to spend the resources required to do so, such as more extensively monitoring ambient environmental conditions. Moreover, EPA and the states disagree about the extent to which these newer measures should replace, or merely supplement, traditional measures.

This section discusses the evaluative criteria that EPA should use to evaluate the efficacy of state enforcement efforts. While environmental indicators most closely reflect the bottomline issue of the condition of the ambient environment and thus in theory are the most desirable measures, they generally are expensive to implement. Moreover, by and large it is very difficult to link ambient environmental conditions to enforcement activity or compliance assistance programs (although there are some recent pilot projects in which this has been done successfully, as we discuss below). Thus, we do not think it is currently realistic to shift primarily toward an indicator-based system, although efforts should certainly be made to increase the capacity to do so. We think that greater reliance should be placed on behavioral outcome measures, particularly various indices of compliance by regulated entities, although calculating compliance rates also can be quite expensive and raise challenging methodological issues. Finally, even with a shift to outcome-focused performance measures, we believe that the more traditional activity-based indices remain a useful and necessary tool, especially if enhanced to provide additional information about enforcement and compliance assistance activities.

Environmental Indicators

There has been widespread discussion in recent years about the desirability of using environmental indicators to evaluate the effectiveness of agency programs, including enforcement programs, and NEPPS envisions that EPA and the states will shift toward such measures. Environmental indicators are measures of environmental and public health conditions and how they change over time. These include ambient air concentrations, water quality conditions, drinking water quality, acres of damaged wildlife habitat, and so forth. Some environmental statutes have long required agencies to monitor certain environmental conditions. For example, under the CAA, states must regularly monitor ambient concentrations of criteria air pollutants. Under the CWA, at least in theory, states must regularly assess the condition of water bodies within their jurisdiction (we say "in theory" because in practice most states have fallen well short of this requirement, as noted below). Other indicators might be geared to gauging the human health consequences of noncompliance, such as the number of reported illnesses from water systems out of compliance with drinking water standards, reported illnesses from consuming fish caught in waters that do not meet water quality standards, or the population affected by noncompliance with air quality standards.

Although environmental indicators represent the most desirable performance metric, they should not be the principal means for evaluating the effectiveness of enforcement programs in the immediate future, for several reasons.

First, it generally is very hard to link enforcement or compliance activities to improvements in the environment. Although some water quality and air quality changes can happen quickly, other changes in environmental conditions, particularly those impacting entire ecosystems or more distant locales, may take years to manifest themselves. The focus of enforcement activity, by contrast, is short term. Moreover, a range of external variables other than how well an enforcement program is operating—factors such as weather, the condition of the economy, demographic trends, etc.—may influence ambient conditions. The environmental indicators used may not capture the full range of changes triggered by enforcement actions.³⁵

To date, few studies have sought to isolate the impacts of individual enforcement activities on environmental conditions. For example, in its survey of state environmental [33 ELR 10565] agencies, ECOS found that none were able to provide measures of the effectiveness of compliance assistance programs.³⁶ Scholars such as Prof. William Gormley argue that the impacts of agency enforcement activities have been successfully measured in other fields, such as education and health care. He notes, however, that "genuine measures of environmental impact [from agency activities] are practically nonexistent," and that it is particularly difficult to measure the impacts of decisions made by environmental agencies.³⁷ Even states that have been leaders in seeking to develop performance measures, such as Florida and Oregon, he points out, still rely very heavily on traditional output measures, and, to a lesser extent, outcome measures. Gormley elaborates:

The production of authentic impact measures would be both complex and expensive. The failure of the four state agencies [studied by Gormley—Florida, Oregon, New Hampshire, and Virginia] to produce such measures reflects resource constraints, as well as the state of the art. Even the U.S. EPA, with much larger research and planning staffs, has not developed a battery of good, reliable impact measures. We have much better statistical models for isolating the effects of government interventions in education and health care on desired outcomes in those areas.³⁸

One notable exception is EPA's Clean Charles 2005 initiative, a multi-pronged effort launched in 1995 by EPA's Region 1 to make the Charles River swimmable in 10 years. In that instance EPA was able, to a considerable extent, to gauge the effectiveness of various enforcement and compliance assistance actions it undertook, and to isolate these effects from other background variables. EPA accomplished this by monitoring water quality in the river very frequently (monthly), using geographically close measuring stations, focusing on a limited water body (the Charles River is only 80 miles long), and separately monitoring water quality during wet weather and dry weather conditions, in an attempt to isolate the impacts of wet weather on pollution levels (since stormwater runoff during heavy rains tends to significantly affect water quality).³⁹ Researchers Shelley Metzenbaum and Tiffin Shewmake argue that other "place-linked" efforts such as this can help overcome some of the difficulties in linking ambient conditions to enforcement activity.⁴⁰

Second, the data necessary to meaningfully implement such performance measures is woefully lacking. The state of data about water quality in the United States is an oft-cited example, and it is telling. For instance, according to EPA, as of 1998 states had assessed water quality for only 23% of the nation's rivers and streams, 42% of its lakes, ponds, and reservoirs, and 32% of its estuaries.⁴¹ Even for those water bodies that have been assessed, the data is often unreliable and inconsistent across states (or even over time within the same state). Many states report attaining water quality standards based on measurements of dissolved oxygen or suspended solids, not using other biological, physical, or chemical parameters deemed important by EPA.⁴² Some states report that water bodies are "assessed" based on a single sample anywhere on the body, no matter how large it is; others assume that rivers for which no information exists are supporting their uses; and many water bodies are "evaluated" rather than "assessed," meaning there is no current site-specific data available for them.⁴³ The U.S. General Accounting Office (GAO) found in 2000 that only six states reported that they had a majority of the data needed to assess whether their waters meet water quality standards. Another recent report on environmental conditions in eight Great Lake states concluded that "the quality and consistency of state environmental data needs dramatic improvement. States collect and report even the most basic air and water quality data in ways that make comparisons difficult and frustrate the public's ability to understand their own state's problems and progress."⁴⁴

It will take tens of millions of dollars, if not more—and years—to remedy this informational shortfall. Critics have called for the establishment of a bureau of environmental statistics, led by a professional appointed by the president and confirmed by the U.S. Senate for a fixed term, and with sufficient budget and independence to be insulated from the political process, and for strengthening EPA's Office of Environmental Information, giving it authority to build far better information about environmental conditions and results.⁴⁵ We agree with these recommendations, but we are a long way from having these federal resources in place. [33 ELR 10566] Moreover, it is hard to imagine that states will have the political will to develop such measures.⁴⁶ Nor does effective pressure from EPA to force the states' hand seem likely. As the National Academy of Public Administration (NAPA) recently concluded, EPA approved, indeed invited, the participation in NEPPS of some states that had poor track records of performance and little or no history of developing performance measures.⁴⁷ Likewise, while some states have included performance measures in their Performance Partnership Agreements (PPAs) with EPA,⁴⁸ many have failed to implement these measures, with no apparent consequences from EPA.⁴⁹ Even during the economic prosperity of the late 1990s, states apparently did not increase their spending on environmental protection; in fact, according to ECOS, overall state environmental spending declined slightly from 1996 to 2000.⁵⁰

An alternative indicator measure is trends in statewide or regional emission or discharge levels from regulated facilities, particularly trends in such levels in excess of permit limits. For example, a state agency could examine the collective level of discharges of a set of pollutants from regulated entities over a period of time, compare it to the level of legally permitted discharges, and see whether the trend is toward compliance or away from compliance with discharge limits. Calculating this type of measure is more feasible than obtaining comprehensive ambient data, since many facilities are required to monitor their routine air and water discharges, although this measure is less reflective of "bottom-line" environmental health.⁵¹ It should be noted, however, that such measures, even when feasible, have their limits as well. For example, an enforcement program may be successful in prompting compliance with important regulatory requirements (training, spill prevention, emergency planning, etc.) that do not translate into demonstrable emission reductions (a point we also discuss below).

Thus, we recommend that at present, EPA not place principal emphasis on environmental indicators measures in its evaluative criteria, although it should make a concerted effort to develop the capacity to use such measures in the future.

Rates of Compliance and Other Outcome Data

The second type of performance measure is behavioral outcomes measures, which reflect the actions taken by regulated entities or the public in response to enforcement actions. In our view, one outcome measure, levels of compliance, should be very significant in judging the efficacy of enforcement programs. After all, the underlying goal of enforcement efforts is to ensure that regulated entities comply with the law. If a state can show that high percentages of its regulated facilities, especially major facilities whose non-compliance can pose significant adverse impacts, consistently comply with environmental requirements, there generally is little basis for EPA to question the mix of its enforcement and compliance assistance strategies. Indeed, the states have identified compliance rates as the performance measure they would most prefer to use to evaluate their enforcement programs (although relatively few do so).⁵² At the same time, it is important to recognize that compliance rates do not tell the full story about the efficacy of a state's environmental program. If a state's underlying permit requirements are weak, or outdated, for instance (and recent studies indicate that a distressingly high number of CWA permits are outdated),⁵³ a high rate of compliance may not be particularly meaningful or informative about the condition of the state's environment. This is obviously an issue that extends well beyond the efficacy of an enforcement program.

The Policy Framework identifies rates of compliance as one criterion for evaluating state program performance, and compliance measures are now included in the core accountability measures identified by the Office of Enforcement and Compliance Assurance (OECA) for evaluating state enforcement and compliance activities.⁵⁴ But historically, compliance measures have not played a central role in EPA's evaluation of state programs. A major reason for this is that there are significant difficulties in trying to calculate compliance rates (only a few of which are highlighted here). For example, to obtain a valid measure of compliance rates by a particular industry, agencies need to survey all regulated facilities, or chose a statistically reliable subgroup at random. But in many cases agencies have traditionally targeted facilities to inspect based on factors such as which facilities pose known or suspected risks, a facility's volume of emissions, past compliance history, etc. Likewise, in attempting to measure progress in rates of compliance, an important pre-requisite is baseline data with which to compare current conditions. But both EPA and state agencies have limited baseline information. For instance, the agencies have background compliance information typically only for larger, more significant facilities; data are especially scarce for small businesses that historically received few inspections.

Another thorny set of issues concerns how to measure what compliance is, including the appropriate time period in which to judge compliance. Does compliance mean total continuous compliance, or compliance in excess of a certain [33 ELR 10567] percentage of the time? EPA assesses compliance quarterly; if a firm is out of compliance once each quarter, it may be considered out of compliance for an entire year. Looked at differently, the same data would show a facility in compliance most of the time.⁵⁵

A related issue pertains to the time it takes a firm to return to compliance, after noncompliance is discovered. After a violation is uncovered, how quickly must a firm act in order to be considered in compliance?⁵⁶ Moreover, how does the government verify compliance? Is it based on an on-site inspection and sampling? A physical inspection without sampling? A phone call? A facility's self-monitoring records?

And related, but separate, and of central importance, what is the definition of significant noncompliance? EPA and the states have often disagreed in recent years over what the definition of this term is, leading to considerable acrimony. NAPA recently concluded that no standard methodology exists for calculating reliable compliance rates.⁵⁷

To facilitate use of compliance data as a central measuring tool, EPA should work with the states to develop standard methodologies for calculating compliance rates. This should include an agreed-upon (and consistently interpreted) definition of significant noncompliance for purposes of federally authorized programs. There are no easy answers to some of the questions posed above, but using a standard methodology will greatly enhance EPA's ability to evaluate the progress of state programs and compare them to each other, and also allow states to learn from one another. (The Environmental Compliance Consortium—a project of approximately 15 states and academics working to develop better ways to measure enforcement and compliance programs—also has suggested that compliance rate information can be very useful even if compliance rate methodologies are not fully standardized, provided that state agencies clearly report all of the variables that affect the determination of the compliance rates.⁵⁸ That project is seeking to develop ways for states to report compliance rates in a way that makes the underlying assumptions more transparent.)

Once standard compliance rate methodologies are adopted, we recommend that several types of compliance data, among others, be used to judge state performance:

1. Obviously, of central importance is the overall rate of compliance: What percentage of regulated facilities are found in noncompliance, and what percentage in significant noncompliance? Secondarily, what is the rate of compliance and significant noncompliance in sectors identified by EPA or the states as high priority, because of the high volume or high toxicity of facility emissions, significant potential for harm from activities, proximity to population, contribution to air or water quality violations, or other factors?

2. Of facilities found in significant noncompliance, how many are repeat violators from the prior five years? How quickly are the facilities found in violation returned to compliance? These questions illuminate how well an enforcement program is working to deal with problematic facilities.

3. What is the percentage of facilities found in compliance during unannounced inspections? (Compliance rates based on pre-announced inspections may not be as illuminating as those found based on surprise visits. While it may not be practicable to make all or even most visits unannounced, it would be helpful to obtain data from at least some subset of regulated facilities based on surprise visits.)

4. What is the severity of noncompliance that occurs, and its potential environmental and health impacts? For example, how much additional pollutant loadings or emissions have resulted from facility noncompliance? How far in excess of permitted levels are noncompliant facilities' releases? How many unregulated waste streams, newly created waste sites, or environmental threats are found at facilities where noncompliance occurred?

5. What is the rate of noncompliance among facilities that received compliance assistance, voluntarily disclosed violations under state audit laws or policies, or participated in other cooperative enforcement programs, as compared to historic rates of noncompliance, or rates of compliance by the entire universe of regulated facilities? This can help states (and EPA) assess the efficacy of these alternative enforcement approaches.

Finally, another useful behavioral outcome measure (although not strictly a compliance index) is environmental and health improvements resulting from enforcement actions, compliance incentive programs, or compliance assistance programs. This can include reductions in pollutants discharged; institution of new compliance management procedures; or other behavioral changes by regulated entities. These measures can help provide a fuller picture of the efficacy and impacts of agency enforcement activities.

Many states do not currently report the measures outlined above, including rates of compliance, the performance metric the states indicate that they prefer to be judged by. For its part, EPA in its own enforcement program, now reports many of these measures, including the percentage of recurrent significant nonviolators, the duration of significant noncompliance, and the environmental and health improvements [33 ELR 10568] resulting from enforcement actions.⁵⁹ EPA also has started determining compliance rates for selected regulated sectors.⁶⁰

Agency Enforcement Activities

Outputs are measures of program activity, the traditional "beans" used by EPA to evaluate state enforcement programs. These include the number of inspections, number of notices issued, civil and criminal actions filed, and, more recently, the number of facilities that are reached through various compliance assistance efforts and that voluntarily disclose violations under audit policies or laws.

Critics charge that EPA should pay relatively little attention to these enforcement indices as metrics of performance and instead focus on results-based measures. (Some commentators, such as NAPA, also suggest that activities-based data should not be used to evaluate state performance because of inadequacies in the data and disputes between the states and EPA over its accuracy.)⁶¹ We conclude, however, that it would be inappropriate—indeed irresponsible—for EPA to discount traditional measures of performance, especially until better performance measures are developed and implemented, and data improve. Despite the advent of NEPPS and its promise of performance-based management, many states have not developed new performance measures, or even gathered the data required by their PPAs (although to be fair, some have made some significant breakthroughs). As noted above, moreover, huge gaps remain in our data about ambient environmental conditions, and about compliance rates.⁶²

Interestingly, moreover, despite the chorus of public calls by states and other critics for newer, results-oriented performance measures, many states still appear to be comfortable with the traditional evaluative criteria. For example, in Region 8, which has recently implemented a new uniform oversight system that we profile below, the oversight criteria, all agreed to by the states in the region, focus very heavily on traditional measures of performance, including compliance/monitoring strategies, timely and appropriate enforcement action, and accurate and complete data management. These are, according to Region 8 officials, the evaluation criteria the states were interested in using. The Region broached the idea of using more outcome-based measures. Several states declined. Other states expressed interest in using such measures in addition to tracking the base program performance, and the Region added an additional criterion, entitled "Alternative Projects Evaluation," that is intended to encompass alternative approaches to promoting compliance. The Region also indicated that several states expressed little interest in this criterion. Their primary interest lay in obtaining feedback on performance of the core enforcement functions. (With respect to NEPPS, Region 8 advised that all of the NEPPS accountability measures that are "outcome" measures are only "pilot" measures; states only have been asked to report regarding these measures, they are not obligated to do so. At least in Region 8's experience, none of the states has done so.)⁶³ Likewise, in Region 1, the Region and states have agreed to a system of comprehensive, multimedia reviews of state environmental and compliance assistance programs that do not contain information on outcomes or environmental results.⁶⁴

Other arguments support continued reliance on activitybased measures. One of the central findings of the empirical research about enforcement is that enforcement presence matters. Repeated studies show, among other things, that more frequent inspections promote compliance (as well as reduced emissions).⁶⁵ (We know less about the impacts of compliance assistance visits or other monitoring efforts.)⁶⁶ Thus, EPA should continue to require states to report data about the number of inspections and other compliance monitoring efforts carried out by their enforcement programs, since this is one strategy that the empirical literature indicates works. Ideally, states should report the coverage achieved by their programs, i.e., the percentage of regulated facilities reached by inspections and other activities.⁶⁷ Similar reasons counsel retaining other deterrencebased indices, such as the degree to which states take timely [33 ELR 10569] and appropriate enforcement action in response to noncompliance, and impose escalating and meaningful penalties in response to repeated violations, since enforcement research and experience suggests that these actions also promote compliance, although the evidence about this is less well-developed.

Moreover, as Professor Gormley explains, states need to continue to collect data on the level of program activity in order to accurately compare the relative efficacy of various enforcement and compliance tools. Despite recent research in this area, there still is relatively very little empirical data examining which enforcement and compliance tools work best.⁶⁸ Gormley points out:

If states that opt for more technical assistance and fewer civil penalties against polluters experience greater environmental improvements than other states, that would be an extremely important finding. But we will never know that—or the reverse—unless states continue to collect output data on a regular basis. The ideal set of performance measures is a mix of outputs and outcomes, not outcomes alone.⁶⁹

Likewise, even once systematic data are in place for performance measures, it is important to continue to measure output measures, because outcomes are so significantly influenced by factors beyond the control of agencies.⁷⁰ For the reasons detailed above, ambient environmental conditions are affected by a range of considerations apart from the efficacy of enforcement efforts. Compliance rates as well may be influenced by economic and industry trends, prevailing industry beliefs about the reasonableness of environmental regulation, and other factors. Likewise, a facility's compliance with certain regulatory requirements, such as the many standards that are preventative, may not translate into quantifiable emissions reductions or environmental benefits. As EPA has correctly noted, "causality between program activities and outcomes is usually impossible to prove. Outcomes generally cannot be attributed to individual functions of an agency or program. 'Prevention' or deterrence of undesired outcomes is difficult to measure."⁷¹

Accordingly, we believe that EPA should continue to require states to report many of the traditional activitybased measures currently required by the agency. (As explained below, we advocate a system of differential oversight in which better performing states will be relieved from various reporting requirements.) Thus, states should report the percentage of the regulated community that is reached by enforcement activities, including traditional inspections, as well as, to the extent feasible, compliance assistance visits or other regulatory interactions. (We recognize that accurately tracking the coverage of compliance assistance programs can be problematic, since some companies prefer to participate in such programs anonymously.)⁷² Likewise, states should report on their responses to instances of significant noncompliance. In counting enforcement responses, EPA should consider informal enforcement actions, which the states argue can effectively return facilities to compliance, such as oral warnings, warning letters, and notices of violations that require facilities to correct violations without assessing penalties.

We discuss below two suggestions for supplementing the traditional measures used by EPA.

Response to Citizen Complaints

ECOS reports that in 1999, states received 88,000 complaints from citizens.⁷³ While it is uncertain how many of those involved compliance- and enforcement-related issues, it is fair to presume that a substantial portion did. States vary in their procedures for receiving and handling such complaints. In its recent audit of Ohio's environmental programs, for example, EPA expressed concern that the rigors of the state's complaint process may dissuade citizens from sending in complaints.⁷⁴ Likewise, in a 2001 audit, EPA's Inspector General concluded that the states examined did not maintain adequate or consistent tracking systems for citizen complaints, even though such complaints are an important source of violation information.⁷⁵

We think that EPA should consider, as a component of state performance, how well a state facilitates and responds to citizen complaints. Admittedly, this may be difficult to measure in any type of meaningful qualitative way, other than perhaps to survey representative samples of citizens. One useful index might be how quickly complaints are responded to and resolved. Another is whether and how well the agency has publicized its complaint process. Another is how many complaints led to some type of contact with a regulated entity, whether the provision of technical advice, an informal visit, or enforcement activity. Our suggestion, in short, is that a state should be judged in part on how well it serves the public, and measures should be developed and implemented to facilitate evaluations on this basis.

Reporting Information to the Public

As we argue later in this Article, EPA has limited leverage in influencing state enforcement practices, and, as a result, public disclosure of agency performance is likely to generate some of the most effective pressure for improvement. In order to generate such pressure, agency performance data must be disclosed to the public in clear, meaningful, easily accessible, i.e., on-line searchable, and standard, i.e., consistent from year to year, formats. States should be evaluated on how well they discharge their obligation to report performance data to the public in this fashion. Environmental groups can use this data to compile organizational report cards on the efficacy of state enforcement efforts. At least 24 [33 ELR 10570] states have published "state of the environment" reports or analogous environmental indicator reports, with the number having increased significantly in recent years.⁷⁶ We agree with the recommendations of a recent NAPA report that

each state should publicly report its progress in achieving environmental results as part of an annual state-of-the-environment report. These reports can strengthen the accountability of government agencies to each other and to the public. They also can facilitate learning across agencies.⁷⁷

In a similar vein, the Michigan Environmental Council, in its recent review of the environmental performance of eight Great Lake states, included as an evaluative criterion the quality of "state of the environment reports." The council explained that the most important factor in reviewing the reports "is whether they serve as a useful 'one-stop' digest for the information citizens are entitled to have on the full gamut of a state's environmental programs."⁷⁸

In recommending that reporting information to the public be a basis for judging state performance, we believe that the recent failure of most states to encourage public involvement under NEPPS is instructive. Although one of the basic tenets of NEPPS is to "facilitate and encourage public understanding of environmental conditions and government activities,"⁷⁹ an outside review in 2000 found that most states had made little progress in involving the public in the NEPPS process. The review also concluded that the states' poor performance was attributable in part to EPA's failure to insist upon such efforts as a condition for approving PPAs.⁸⁰ If, in contrast to this early experience under NEPPS, states know they will be judged by how well they provide enforcement data to the public, they are likely to respond with better information.

Agency "Inputs"—Spending and Staffing

For a state to receive authorization from EPA, it must demonstrate adequate resources to implement the federal programs. We believe that an ongoing measure of state performance should be the state's level of environmental spending and staffing, measured against the level of commitments promised to EPA as part of receiving authorization to administer a program. Money is certainly not a panacea to achieving compliance, but it does matter. Studies by Prof. Eric Helland report the intuitive conclusion that states with less resources focus their efforts on more routine inspections rather than comprehensive sampling inspections, which may be up to 10 times more likely to detect a major violation.⁸¹ Likewise, in its 2001 audit of Ohio's air enforcement program, EPA found serious resource gaps, noting that the state appeared to lack sufficient resources to perform all of its permitting and enforcement duties.⁸² Another recent EPA audit found that the Los Angeles water quality control board had only 2.5 staff years to monitor more than 3,300 stormwater permit holders and conduct searches for as many as 10,000 unpermitted facilities that were subject to regulation.⁸³ On its face, a program like this is grossly underfunded and cannot hope to achieve meaningful enforcement coverage.

Better Methods of Program Evaluation

What is the best administrative approach for EPA regional offices to use in conducting oversight of state enforcement and compliance efforts? EPA affords regional offices considerable latitude, and in practice, they use different approaches.⁸⁴ For example, in the early 1990s Region 1 adopted an unusually comprehensive approach to oversight in which it conducts one review at a time; as of late 2001, it had completed reviews of four of the six states, with a fifth virtually done and the sixth to be completed within the year.⁸⁵ EPA's Inspector General found in 2001 that other regional offices were also performing in-depth evaluations of state programs, although the scope and frequency of these evaluations varied. For example, Region 4 performed indepth evaluations every eight years or so, while Region 6 did so on a semiannual basis.⁸⁶ In short, regional oversight has been inconsistent and not always very effective. EPA's Inspector General, writing about the CWA, concluded in 2001 that

to ensure fair and effective enforcement … EPA regions need to improve their in-depth program evaluations and annual performance evaluations of state performance. These evaluations need to be consistent, continue toward a goal of measuring the effectiveness of performance, and be made easily accessible to the public.⁸⁷

We recommend that EPA systematically review the oversight methods of the regional offices, pick out some of the best features of the various approaches, and make a concerted effort to improve the quality and consistency of oversight evaluations nationwide. In particular, EPA should recommend approaches that allow for consistent comparisons among states and programs, and that are transparent to the public. There are a number of examples that EPA can look to. One is the Region 1 strategy, mentioned above. Another example comes from Region 4, which developed a series of charts showing how each of the states in its area is meeting the core performance measures identified by NEPPS as the basis for judging state performance (including measures showing the performance of state waste, water, air, as well [33 ELR 10571] as enforcement and compliance assurance programs).⁸⁸ The charts also indicate performance measures for which the states do not have any data. Another related example (outside of EPA) is the evaluation of the environmental performance of eight Great Lake states that was recently completed by the Michigan Environmental Council. The council evaluated the states on a set of consistent criteria, including the strength of their statutory authorities, and implementation of policies, for addressing certain key environmental problems; the rights they afforded citizens to participate in agency decisions; and how well they reported information to the public in "state of the environment" reports.⁸⁹

We describe two additional possibilities below. One is a "unified" approach recently initiated by Region 8, which has been identified by some observers as a leader in innovative oversight practices. This approach is simple and understandable, and because of its common format across program areas and across states, allows for easy comparisons of performance and prioritization of resources. An alternative method is to have regular external evaluations of state performance, to ensure the type of hard look review that has been lacking from some regional evaluations.

A Unified Enforcement Oversight Approach

Region 8 efforts to revamp its oversight efforts were precipitated in part by a critical EPA Inspector General audit of the Pennsylvania enforcement program in 1997, and in part by the request of state environmental directors to have Region 8 move away from case-by-case reviews of enforcement activities and instead provide after-the-fact reviews of the overall performance of their enforcement program and activities.⁹⁰

Region 8 brought its programs together and, working with state counterparts, developed a uniform enforcement oversight system (UEOS) intended to bring a greater level of consistency to EPA oversight (across programs) and also to establish in a straightforward way what the expectations should be.⁹¹ Region 8 first used the UEOS in 1998, refined the version for fiscal year (FY) 1999, and has refined it still further for FY 2002, based on "consensus recommendations" from regional office personnel and personnel from the six Region 8 states (Colorado, Montana, North Dakota, South Dakota, Utah, and Wyoming). The UEOS covers enforcement programs in the air, hazardous waste, drinking water, and water pollution control programs.⁹² It focuses on "core" program activities, divided into four categories: compliance monitoring, i.e., inspections-related performance, enforcement activities, e.g., pursuit of timely and appropriate enforcement, database management, e.g., timely submission of accurate and complete data, and performance under annual agreements, i.e., the PPA or state/EPA agreement. There are 14 criteria listed for UEOS 2002.

There are several notable features of this approach. First, it is manageable for Region 8 and states alike. The 14 criteria take up all of one page. (The program-specific "legends" that provide background on what each program is interested in learning for each criterion are set up so that each is quite accessible, and is only eight pages long. For each criterion, the review provides for three types of information: (1) findings; (2) the information reviewed; and (3) recommendations if corrective action is needed.) Thus, the basic expectations are laid out clearly and succinctly, eliminating the need to search through reams of guidance, policies, directives, agreements, and the like to try to determine the precise measures or expectations against which performance will be evaluated.⁹³ Further, from its structure and content, the UEOS' purpose—problem diagnosis and problem solving, rather than finger pointing—comes through clearly. Region 8 first lays out its findings about a state's performance. It then spells out the sources of information that it reviewed in reaching those findings. Finally, it concludes with recommendations for fixing any deficiencies.

The UEOS makes it much easier for EPA managers to identify patterns in performance within and across states. The 14 performance criteria are identical for all programs⁹⁴; program-specific legends are based on a generic legend Region 8 prepared, ensuring additional consistency in the format and content of the evaluations. Previously, Region 8 did not use consistent criteria across programs or use a common format for evaluation, so it was much more difficult to spot patterns or to get an overview of program strengths and weaknesses.⁹⁵ Indeed, commentators more generally have argued that the varying criteria traditionally used to evaluate state programs "virtually precludes comparison either between states or between programs," and that the information is "not presented in a way which encourages [its] use by the public and legislators."⁹⁶

Regional staff believe that the new oversight system has led to improvements in state performance, by facilitating better communication and joint problem solving.⁹⁷ Specifically, they report: (1) one state's hazardous waste program went from having 50% of penalty orders include both gravity and economic benefit calculations, to 100%; (2) a [33 ELR 10572] state CWA permit program is finally drafting an Enforcement Management Strategy after discussions and a clear understanding that it was not collecting the correct economic benefit and gravity portions of its enforcement penalties; (3) a state air program made great improvements in its inspection reports; and (4) one state's drinking water program has reduced the number of lead and copper significant noncomplying facilities by 83%. The new oversight system also has facilitated EPA's implementation of an (informal) differential oversight scheme for the states. Region 8 has the program directors review the evaluations for all six states together. Using the same forms, applying the same criteria, and reviewing the evaluations at the same time has enabled Region 8 to systematically identify gaps in state performance and allocate its resources accordingly. (This informal differential oversight typically takes one of two forms—the provision of additional technical assistance or a decision to conduct additional oversight, such as reviewing more inspection reports, or conducting more oversight inspections.)⁹⁸

Region 8's approach also can easily be used to facilitate public comparisons of relative state performance—an idea we endorse in more detail below—though one which Region 8 has not pursued. Indeed, initially, Region 8 calculated numerical scores for the states so that it would be possible to compare state scores and rank them. It subsequently eliminated the scores because of its sense that they were distracting and that too much time was spent debating the details that affected the score.⁹⁹ Probably Region 8's concern about impairing its working relationship with the states was also a factor. The trade off, however, is a loss of easily comparable evaluations. Moreover, the regional evaluations of state performance are not public; again, as we argue below, these types of evaluations should be disclosed to enhance public accountability of state agencies.

"External" Evaluations of State Performance

One reason that EPA oversight of state programs has not been more effective is the tight link between EPA regional offices and the states they supervise. Many regional administrators have close political ties with state officials and may be more responsive to pressures from state officials than to EPA headquarters. Indeed, a great deal of the publicly available evidence about the quality of state performance has come not from EPA's regional offices but from "outside" auditors, particularly EPA's Inspector General and the GAO. In a series of audits and reviews in the mid- and late 1990s, these offices documented widespread inadequacies in state enforcement efforts.¹⁰⁰

To ensure a "harder look" at state performance practices, we believe that EPA should institute a system of regular, "external" (that is, outside of the regional offices) review of state performance. This could come in a number of forms. For example, EPA could ask the Inspector General's office to regularly audit state enforcement programs. NAPA opined in its 2000 report that the Inspector General has the resources and potential to perform effective program evaluation of EPA programs, but that it would have to transform its staff expertise to do so.¹⁰¹ Another worthwhile idea is offered by Lee Paddock and Suellen Keiner. They suggest that specially trained, experienced EPA staff, or perhaps senior staff from other states, be used to evaluate state performance. They note that this is similar to what the Dutch government does to evaluate provincial and municipal environmental programs.¹⁰² It also is comparable to how the U.S. Department of Justice evaluates the performance of its 94 U.S. Attorneys' Offices; evaluations of U.S. Attorneys' Offices are conducted by teams of experienced prosecutors and administrative personnel from other U.S. Attorneys' Offices.¹⁰³ Such an approach can have multiple benefits: those being evaluated are likely to take the findings of the evaluators seriously because of the evaluators' closely related backgrounds; at the same time, evaluators are likely to be exposed to different approaches through the evaluation process that will help to inform their own performance in their "home offices."

EPA and the States Should Improve Their Data Management Systems

An issue closely related to that of developing new evaluative criteria for state enforcement programs is the need for EPA and the states to improve the quality of their data management systems, and in particular to improve their coordination in managing such information. Numerous observers have pointed out that existing enforcement and compliance data is of relatively poor quality. The lack of better quality data considerably complicates efforts to evaluate state performance. It has also led to contentious disputes between EPA and the states over the reliability of enforcement data and what conclusions can be drawn from it, and makes it hard for the public to access enforcement and compliance information.

Thus, for example, NAPA found that "EPA's current data systems are seriously limited by data inaccuracies, [the] failure of some states to submit complete data, separate single-medium systems, and data inaccessibility because information resides on antiquated main-frame machines."¹⁰⁴ There are also significant discrepancies between data maintained by the states and EPA, and EPA and the states' databases are generally not technically compatible.¹⁰⁵ ECOS, for instance, found that about 80% of states had experienced significant data discrepancies between their state data and the data as it appears in EPA's databases.¹⁰⁶ Likewise, a 2002 report by EPA's Inspector General found that the National Enforcement Docket System (DOCKET), the official EPA [33 ELR 10573] database for tracking and reporting on civil judicial and administrative enforcement actions, "contained significant instances of inaccurate and incomplete data."¹⁰⁷ An audit of CWA enforcement by the Inspector General similarly found that "EPA's Permit Compliance System—its national permitting and enforcement system—was incomplete, inaccurate and obsolete…. Hundreds of thousands of dischargers were not monitored by the system."¹⁰⁸ And, tellingly, when ECOS attempted to prepare its comprehensive report of state enforcement activity to Congress in 2001, only 38 states were able to provide data about enforcement and compliance assistance activities. ECOS had to resort to EPA's databases for data on the state's own enforcement activities (even though states have been required to keep such data under an EPA-ECOS agreement at least since 1995).¹⁰⁹

Clearly it is desirable to fix these shortcomings and coordination problems, and EPA has undertaken a number of efforts in recent years to do so. The GAO reported, for example, that the OECA has organized a series of workgroups to address data quality issues.¹¹⁰ EPA is in the process of implementing a system to modernize its enforcement and compliance systems, called the Integrated Compliance Information System (ICIS). EPA's creation of an Office of Environmental Information in 1999 is likely to lead to easier access to information in EPA's database.¹¹¹ Another effort under way to improve the accuracy and flow of enforcement data is the National Environmental Information Exchange Network, an information network that is creating common data standards and standardized methods of data exchange between EPA and the states.¹¹² In 2001, Congress authorized $ 25 million to implement the Exchange Network, and it approved nearly $ 20 million a year later.¹¹³ As NAPA correctly has concluded, moreover, "it is not sufficient merely to fix the data systems and improve the flow and accuracy of existing data activities. The systems must also be modernized, integrated, and expanded to include more environmental data, greater ability to analyze them, and increased utility for setting priorities, targeting resources, and providing better information to the public."¹¹⁴ We recommend that EPA and the states vigorously pursue these efforts, and, in particular, that they devote sufficient resources to ensure that these data quality problems are remedied.

EPA Should Implement a Formal System of Differential Oversight Based on State Performance

We believe that a key to improving state performance is making states more accountable by establishing a clear "differential oversight" approach that encourages good performance and discourages undesirable behavior. EPA should provide relatively substantial oversight and less flexible funding for states with poor performance, including increased EPA review of state permits and inspections, and reduced oversight and increased flexibility in terms of use of federal funds for better performing states.

The idea of differential oversight is quite straightforward: EPA should have a strategy for overseeing states that provides maximum encouragement and incentives for desired behavior. Since states strongly prefer less oversight, a differential oversight scheme should be a strong incentive to encourage desired state behavior. Moreover, such an approach would allow for the most efficient deployment of limited federal enforcement resources.

There has long been widespread agreement that EPA has the flexibility to tailor the exercise of its oversight responsibilities based on the performance and capabilities of particular states and that, conceptually, differential oversight is a good idea. NAPA, one of the strongest proponents of devolving authority to the states, argued in 1995 that

the paired ideas of "differential oversight" and "accountable devolution" that are beginning to percolate within the agency hold particular promise as a model for encouraging less prescriptive environmental management approaches without compromising environmental quality…. Differential oversight would match EPA's intervention in the state to the state's success in achieving desired results. In some states, EPA's involvement would be limited to constructive program review and public praise; in others, EPA would have to work more aggressively to ensure a basic level of environmental protection.¹¹⁵

Moreover, NEPPS reinforces, and indeed embraces, the notion of differential oversight as one of its key principles.

Unfortunately, EPA and the states have retreated from the differential oversight system envisioned in NEPPS—both the concept of "leadership" status for the absolutely best performing states, as well as the idea of a formal differential system that distinguishes, and rewards, better performing states. The major reason appears to be state resistance to such a scheme and EPA's unwillingness to press the idea in the face of such resistance. The GAO reported that a formal system would be "both controversial and difficult to implement."¹¹⁶ A 2000 report concluded that differential oversight has stalled because

states resisted any formal ranking of their performance, so EPA backed down. EPA and ECOS did agree [33 ELR 10574] on core measures that might allow comparison of state performance, but many states refused to report all measures, and EPA has decided not to assemble—even at the regional-office level—the data that could make such comparisons possible.¹¹⁷

NAPA itself, which championed the idea as noted above, now favors a less structured approach.¹¹⁸

If EPA is serious about embracing the rhetoric of performance-based management, differential oversight is a concept on which it should stand firm. In fact, EPA regional offices have traditionally carried out some type of differential oversight informally, and continue to do so, with buy-in from their states. Both Region 1 and Region 8, for example, reported in interviews that they tailor the number of federal inspections or conduct more intensive reviews of files or inspection reports based on the performance of their states. A 1995 study by the Environmental Law Institute reported that Region 5 had implemented a differential oversight scheme for RCRA and the CWA in which reporting requirements and real-time permit reviews were reduced based on the quality of state performance.¹¹⁹

What incentives would be appropriate for superior performing states? One way for EPA to reduce its oversight role would be to conduct less frequent audits of state performance, or to only conduct limited, programwide, after-the-fact reviews, rather than "real-time" or case-by-case review of permits or enforcement actions.¹²⁰ Thus, for example, EPA would not review state-drafted CWA permits before they are issued (a practice the CWA authorizes but that states find irritating),¹²¹ or conduct oversight inspections of facilities that the state has inspected (or require joint EPA/state inspections), or review individual enforcement cases to determine if the penalties assessed are adequate. This type of programmatic rather than case-specific review is specifically contemplated by NEPPS, and various PPAs negotiated to date sound the same theme (although these have not been limited to better performing states). In carrying this out, EPA should take care to reduce its oversight only in those specific areas in which the state is performing well. EPA could also agree to refrain from any overfilings in the state for a given period—a source of continual friction with the states (see discussion below). EPA could also relieve states of some of their traditional reporting requirements.

In exceptionally performing states, EPA should take steps to liberate the states from some of the strictures of EPA's traditional enforcement policies and the Policy Framework. Thus, for instance, regional offices would no longer insist on a given frequency of inspections, i.e., all major air permit holders be inspected every year, or a given stringency of inspections, i.e., that level 1 or level 2 inspections—the most comprehensive type—be carried out at the facilities. Likewise, EPA could reduce or make more flexible state reporting obligations about adherence to EPA's "timely and appropriate" enforcement policy, or its penalty policies (in practice many states don't adhere to these policies, but not with EPA's blessing). States bridle at many of EPA's reporting requirements and what they view as the inflexibility of EPA's enforcement policies, so they should welcome the prospect of these carrots.

Moreover, EPA should use the carrot of performance partnership grants (PPGs)—the more flexible funding approach created by NEPPS that consolidates up to 16 categorical grants into a single grant—to reward better performing states. PPGs weaken EPA's influence over state enforcement activities because they deprive EPA of the ability to track program-specific expenditures; some EPA program managers believe that the flexibility inherent in the PPGs allows states to deviate from EPA's national media-specific priorities.¹²² Presently, EPA offers PPGs to all states, despite the level of state performance; 33 states currently receive PPGs, EPA instead should award PPGs only to better performing states, and limit states with poor performance to traditional grants.¹²³

Conversely, poorer performing states should receive relatively more intensive oversight, particularly in those areas in which they are not doing well. This can include more EPA capacity building for state agency staff, such as training and other assistance. It should also include case-by-case review of activity, such as individual reviews of permits, less flexibility for state priority setting, and less leeway for departure from EPA enforcement policies. It should also include less latitude in federal funding. Of course, these sanctions are likely to be ardently resisted by states. This is especially true since in the view of most (if not all) states, NEPPS was supposed to lead to more lax, rather than heightened, oversight of their activities.

Perhaps the most potent driver in a system of differential oversight would be the public attention it draws to poorer performing states (as we argue below, the results of EPA's performance reviews should be made available to the public). Beyond simply publicizing its evaluations, EPA also should provide some special positive recognition of strong performers.

EPA Evaluation of State Performance Should Be Publicly Disclosed

A central lesson of the past quarter century plus is that while EPA has made some progress in promoting better state adherence to EPA enforcement policies, its leverage with [33 ELR 10575] states is limited. Thus, key to EPA's efforts to ensure adequate state performance is establishing a public spotlight to facilitate public scrutiny of state performance. Although this would be a marked departure from EPA's traditional strategies for addressing deficiencies in enforcement performance, it clearly would be consistent with rhetoric from public officials of all political allegiances.¹²⁴

A spotlight approach is desirable for numerous reasons. First, spotlighting and other information disclosure strategies enjoy broad support across the political spectrum, in part because of the diverse rationales which underlie them. Consumer and environmental advocates favor more information as a right, as well as a tool, that enables individuals to protest and protect their interests. Economists prefer such laws over direct government regulation because they rely on the efficiency and power of market forces.¹²⁵ Prof. Cass R. Sunstein has stated that "informational regulation, or regulation through disclosure, has become one of the most striking developments in the last generation of American law."¹²⁶ EPA has identified expanding public access to environmental information as one of the Agency's 10 strategic long-term goals.¹²⁷

Equally important, information disclosure approaches have proven to be very effective. The most prominent example is the federal toxics release inventory (TRI) program of the Emergency Planning and Community Right-To-Know Act (EPCRA). TRI requires manufacturing and certain other industrial facilities to annually disclose their releases and transfers of 654 specified toxic chemicals, subject to reporting thresholds. The information is provided on standardized reporting forms that are submitted to EPA and state officials. EPA is required to make the information available to the public through a national computerized database accessible through personal computers. From 1988 to 2000, reported toxic releases dropped by 48%.¹²⁸ Moreover, releases of some chemicals that may pose particularly significant human threats, i.e., carcinogens, declined at a faster rate than other releases. EPA officials, as well as environmentalists and regulated entities, regularly tout TRI as one of the nation's most effective environmental laws. As Prof. Bradley Karkkainen explains, TRI works because "[firms] are subjected to the scrutiny of a variety of external parties, including investors, community residents, and regulators, any of whom may desire improved environmental performance and exert powerful pressures on poor performers to upgrade their performance as measured by the TRI yardstick."¹²⁹ He further notes the motivating influence of comparative performance data: "Because TRI data allow easy comparisons among facilities, firms, and industries, a poor environmental performance record as reflected in TRI data can cause reputational damage, potentially affecting relations with customers, suppliers, employees, or investors."

Likewise, California's Proposition 65, which requires warnings prior to exposures to listed carcinogens and reproductive toxins, has led to the elimination or reduction of toxics in numerous consumer products, such as ceramicware, nail polish removers, lead foil caps on wine bottles, submersible well water pumps, brass faucets, calcium supplements, and hair dyes.¹³⁰ It also has prompted reductions in air emissions of certain toxic chemicals of approximately 85%.¹³¹

EPA also has reported positive results from its Sector Facility Indexing Project (SFIP), which provides public access to detailed environmental profiles, including compliance histories, on approximately 650 facilities in 5 major industrial sectors. In addition to finding that there was a significant public interest and demand for compliance-related information, EPA noted that public disclosure created healthy incentives for improving data quality.¹³² Among other things, disclosure led to greater consistency in reporting of particular types of information and correction of data processing errors.¹³³

Also related to the enforcement context, research has shown that public disclosure of environmental performance can create strong incentives for pollution control and compliance. For example, several studies have demonstrated reductions in the stock price of poor environmental performers when information about their performance is released, including information about their compliance status.¹³⁴ Likewise, a program in which the Indonesian Environmental Ministry evaluated and publicly reported the environmental performance of regulated parties prompted substantial improvements in facility compliance. A study of compliance and emission levels by pulp and paper firms in British Columbia from 1990 through 1996 found significant improvements in compliance and emission reductions from public disclosure of firms' environmental performance. One recent evaluation of disclosure strategies used throughout the world concludes that such strategies "can be effective in motivating environmental improvement."¹³⁵

The same forces that have made public disclosure effective at stimulating improvements by regulated entities can work to trigger better performance by government agencies. As Prof. Shelley Metzenbaum has persuasively argued, public comparison and evaluation has great power to embarrass and motivate environmental agencies, just as comparison [33 ELR 10576] shopping by the public drives companies to offer better products. She writes:

The potential for embarrassment (and its converse, the possibility of earning honor and distinction) can be a powerful performance incentive. When Florida's environmental agency started to publish comparative compliance rates for each of its regional offices, the compliance rates for one region was [sic] going to be significantly worse than that of the others. The region housed very few regulated facilities and one of those facilities had been out of compliance for years. The facility's environmental staff had unsuccessfully petitioned their management in the past to secure the funds they needed to correct the compliance problem. As soon as the facility managers learned their compliance problem was going to be apparent to the public, they fixed the problem.¹³⁶

In a similar vein, EPA's Inspector General concluded in a 2001 audit that evaluations of state programs by EPA Regions should be made easily accessible to the public. The audit noted that

publicity apparently caused North Carolina to improve its storm water program. Although EPA's Region 4's evaluation of North Carolina's enforcement program in 1999 found serious deficiencies in management of the storm water program, the state did not agree to make investments in the program until we reported these same problems a year later.¹³⁷

Likewise, in the Clean Charles Initiative, described above, a public spotlight on performance also helped lead to improvements in compliance and water quality. At the start of the initiative, EPA's regional administrator presented the Region's first "report card" on the health of the river (it received a "D," since it met only met swimming standards 19% of the time and boating standards 39% of the time). The Region issued report cards each of the next several years thereafter, and EPA and state officials report that the report card "may have been the single most important factor in the initiative's success."¹³⁸ By 1999, the river met boating standards 90% of the time and swimming standards 65% of the time. (The initiative, it should be noted, also relied on a combination of targeted enforcement, targeted compliance assistance, state funding for infrastructure improvement, and voluntary agreements with local communities.)

Public scrutiny also led Florida in the mid-1990s to invest considerable resources in developing performance measures:

Florida officials … told us [the GAO] that the number of penalties assessed, and dollar value of penalties collected, under its federally delegated programs decreased from 1994 to 1996, and that questions were raised as to whether these decreases resulted, at least in part, from a greater emphasis on the use of assistance to achieve compliance. In fact, newspapers in the state subsequently published articles questioning whether the state was letting violators continue to pollute without fear of punishment. Florida officials told us that their major investment in measuring the results of their enforcement and compliance assistance efforts was undertaken, in part, to determine whether these concerns were well-founded.¹³⁹

Indeed, a spotlight approach underlies the citizen submission process established in 1994 by the environmental side agreement of the North American Free Trade Agreement (NAFTA). Under Article 14 of the side agreement, any citizen or nongovernmental organization in Canada, Mexico and the United States (the treaty's signatories) can file a submission alleging that a Member country is failing to enforce its environmental laws effectively. If certain criteria are satisfied, the agency implementing the agreement known as the North American Commission for Environmental Cooperation (NACEC) will prepare a "factual record" about the country's environmental practices at issue, which can include information submitted by interested environmental groups, the country involved, and experts, and which will be made public upon a vote by the NACEC. That is the end of the process—it does not provide for sanctions or directions to the government involved to change its enforcement practices. The underlying idea is that domestic environmental enforcement in the Member states will be enhanced by an international spotlight on such practices, that public scrutiny will motivate the three countries involved to maintain or enhance their enforcement efforts.¹⁴⁰ As one distinguished group of reviewers has put it, the process makes it possible for "some 350 million pairs of eyes to alert the [NACEC] of any 'race to the bottom' through lax environmental enforcement."¹⁴¹

Nonetheless, spotlighting is likely to be resisted by states. Government officials sensitive about their performance are not likely to jump at the prospect of having such performance made easily accessible to the public, or to being "graded" based on such performance. The concerns many states expressed regarding EPA's SFIP initiative, and their urging a go-slow approach for this initiative on a variety of grounds, likely is the harbinger of their reaction to any effort to spotlight enforcement and compliance practices.¹⁴² It is a natural human tendency to prefer not to be judged and have one's "warts" placed in the public spotlight. Certainly efforts to report on performance to date have not been marked by their self-critical nature. (EPA's annual Enforcement and Compliance Assurance Accomplishments Reports, to take one example, have often painted a glowing picture of EPA and state enforcement efforts, overlooking evidence of persistent deficiencies in state performance.)¹⁴³

Likewise, states will contend that efforts to grade and compare their performance are oversimplified and misleading—that current performance measures do not accurately capture the effectiveness of their enforcement and compliance programs; success is determined by matters outside of [33 ELR 10577] their control; and that evaluative criteria are not comparable across states. There is some justification to these arguments, but the solution is not to keep data about state performance private. Rather, greater government accountability can only come with public disclosure of how well states are performing, based on the state of the art of performance measures, delivered with the appropriate caveats. Some oversimplification is inherent, but to withhold the information presumes the public is less capable of carefully and appropriately evaluating information than state or EPA officials.

Moreover, the enhanced public attention on evaluation measures which the states believe are inadequate will itself generate pressure to enhance performance measures, as the example from Florida discussed above illustrates. At bottom, we believe, opponents of spotlighting are fighting a movement for public accountability that is pervasive in society.

One question that arises is whether or not state performance should be publicly graded or ranked. There certainly is a strong argument for this. As noted by Professor Metzenbaum, comparison can be a very powerful motivator. Indonesia's rating program was very successful, especially concerning plants that were rated as below average, and it has led several other countries to similar programs, including Columbia, Mexico, and the Philippines.¹⁴⁴ The TRI program has been quite effective without any governmental gloss on the data, although the nature of the data disclosed—volume of releases by facility—lends itself to ready comparisons.

On the other hand, some may argue that numerical rankings or grades are inevitably misleading or overly simplistic, and there is some truth to that. Moreover, some EPA officials believe that publicly ranking states could impair their ability to maintain a cooperative working relationship with the states. Thus, EPA may simply want to report or provide information and leave it to public interest organizations to create scorecards or variations on such a theme. This is similar to the charts prepared by Region 4, discussed above, which provide comparative information about how states are meeting the core performance measures developed under NEPPS. As another alternative, EPA could simply list the best performing and worst performing states in a Region in a given year, without ranking the remainder.

EPA Should Maintain Withdrawal of State Program Authorizations as a Credible Option

EPA always has the option of using its "ultimate sanction" of withdrawing a state's authorization if it determines that a state's performance falls too far below acceptable levels. This threat is more theoretical than real, however, since EPA has never fully withdrawn a state environmental program, and probably is unlikely to do so in the near future. Nonetheless, the possibility of withdrawal remains a potent tool which EPA has wielded with some limited success in recent years, and it can be strengthened by providing for enforceable time frames for EPA to act on citizen withdrawal petitions.

The substantive criteria governing withdrawal are laid out in each statute and its implementing regulations, as well as in EPA/state operating agreements or delegation documents. They generally provide that EPA can withdraw approval of state programs if any of various aspects of the state programs "do not meet" federal requirements, including the state's legal authorities, permitting program, public participation requirements, or enforcement program. With respect to enforcement, for example, RCRA provides that grounds for withdrawal include failure to act on violations, failure to seek adequate enforcement penalties or collect fines that have been imposed, or failure to inspect and monitor regulated facilities.¹⁴⁵ EPA seemingly has considerable discretion in determining the adequacy of state performance.¹⁴⁶

The procedures for withdrawal are quite involved. For example, under RCRA, EPA is required to prepare written findings, hold an adjudicatory hearing and provide a 90-day grace period for correcting identified deficiencies.¹⁴⁷ The procedures under the CWA are comparable.¹⁴⁸ Indeed, EPA's recent review of a set of withdrawal petitions filed by citizen groups in Ohio illustrates the length and complexity of the process. The petitions were filed and then amended over the course of a three-year period, and raise a host of alleged flaws in programs under several statutes administered by Ohio EPA. After dismissing the earlier claims relating to Ohio's audit privilege and immunity law, Region 5 spent close to two years investigating the other allegations, and issued an over 200-page draft report in August 2001. Region 5 then convened a series of public hearings, and provided for a public comment period of close to six months. It then took the matter under submission for another year, before denying the petitions in February 2003. All of this activity precedes any formal hearing, if EPA decides to initiate withdrawal proceedings. Professor Rodgers has opined that "the procedures for withdrawal of state programs would be suitable for the Nuremberg trials, and will be invoked only upon epochal occasions."¹⁴⁹

EPA has rarely if ever actually withdrawn a state's authorization (although as noted above, it did quasi-revoke Maryland's Title V authority in 2001). It has initiated withdrawal proceedings on a number of occasions. For example, in the early 1990s, it started formal actions to withdraw primacy for eight state SDWA programs (but never carried through with any of them).¹⁵⁰ Similarly, as discussed above, in 1987, EPA sought to revoke authority for North Carolina's RCRA program. The administrative law judge hearing the matter determined that withdrawal was not warranted, a conclusion adopted by EPA's Regional Administrator (and subsequently [33 ELR 10578] upheld by the court of appeals).¹⁵¹ EPA has also threatened withdrawal of state programs in a number of additional instances. Probably its most serious threats came in the mid-1990s, when it initiated withdrawal proceedings in several states because of concerns about audit privilege and immunity laws.¹⁵² During this period EPA also held up granting some states authorization to implement new environmental programs (such as the permitting program under Title V of the CAA) pending changes in their audit laws, and also warned some states that it might withhold federal enforcement grants.¹⁵³ EPA also threatened to withdraw Rhode Island's water pollution permitting authority in 1997 unless the state restored cuts to its environmental budget.¹⁵⁴

EPA's reluctance to withdraw state programs is explainable by two primary factors. The first is money: EPA has stated that it lacks the resources to take over previously authorized state programs. For example, former EPA Administrator Carol Browner testified in 1993 that "there are some States that have seriously considered returning primacy to the Federal government. I will be very honest with you, we don't have the resources to manage even one major State if primacy were to be returned."¹⁵⁵ Another EPA official opined that "even if only a small number of delegated states returned their programs to EPA, [EPA's] enforcement program would not be able to cope with the new responsibilities. Ultimately, there would be less enforcement, not more."¹⁵⁶ Other commentators agree, one noting that "when there is friction between the federal EPA and a state over the administration of an approved program, the federal government's threat to revoke the state's approval may be more than balanced by the state's threat to return the program to the federal government."¹⁵⁷

On the other hand, not all observers think this is true. For instance, after talking with numerous officials in all of EPA's programs, NAPA staff concluded that EPA has sufficient national capacity to operate nondelegated programs centrally. It suggested that EPA could consolidate the expertise of its regional staff and create specialized teams capable of taking over poorly performing state programs.¹⁵⁸ It may be that as some states continue to mature and perform better, EPA could reduce its resources devoted to oversight and free up other resources. Moreover, in our view a key question is not whether EPA can administer a significant number of state programs, but rather whether it has the resources to take over even one or two states, since even a very occasional program withdrawal could have a notable impact on state performance.

The second roadblock to withdrawal is politics; withdrawal is a highly charged, politically difficult endeavor. The move toward devolution has made the political calculus of withdrawal even more improbable. Indeed, as noted above, the only apparent instance in which EPA went through the formal administrative hearing process for withdrawing an authorized state program is when it sought to revoke authority for North Carolina's RCRA program. In that case, notably, the basis of withdrawal was not deficiencies in state performance, but rather that the state program imposed unduly stringent requirements for the siting of new hazardous waste facilities. The withdrawal petition was filed by a hazardous waste facility owner and a hazardous waste trade group. EPA's Regional Administrator ultimately decided not to withdraw North Carolina's RCRA authorization.¹⁵⁹ Professor Rodgers argues that "the case confirms that program withdrawal is a remedy strictly confined to unusual legal pathologies."¹⁶⁰

While recognizing the political limitations of withdrawal, we think it still can be employed as an effective tool. EPA need only act in a few states to have a significant deterrent effect; moreover, EPA can chose to withdraw only part of a state program that it has authorized. Notably, one of EPA's sister agencies, the Office of Surface Mining, did muster the political will to reassume control of the surface mining programs in two states because of poor state performance.¹⁶¹ EPA's threats over state audit privilege and immunity laws in the 1990s were perceived as credible and, despite protests, triggered changes in virtually all of the state laws that EPA found objectionable. As noted above, in 2001 EPA withdrew (at least on paper) Maryland's air permitting program, an action that led the state in 2002 to revise its problematic law on standing to challenge permits.

Even an announcement of a formal policy of revisiting states' legal authorities, capacity, and enforcement performance may prompt improved performance. A withdrawal proceeding is embarrassing to state officials. It represents a public statement by EPA that the state's program is entirely inadequate. The prospect of being branded in this way might well serve as strong motivation to state officials. Further, a withdrawal threat or proceeding might enlist strange bedfellows within at-risk states in the common cause of improving state capacity and performance. Such a proceeding is likely to send an alarm signal to members of the regulated community, who are likely to prefer the "devil they know." State environmental [33 ELR 10579] organizations are likely to recognize that such an EPA initiative is an opportunity to make the case that something is seriously amiss in their state and to win political support for invigorating the state infrastructure. Possibly this sort of strong medicine from an outsider, the federal government, may serve as a focal point for disparate statebased forces to coalesce in favor of enhancing state capacity, even if their motivations for supporting such an upgrade may differ.

The improbability of program withdrawal in most instances, however, highlights the need for more graduated sanctions for poor performing states. As Professor Metzenbaum notes, "if EPA can only threaten withdrawal of delegation authority for poor state performance, a state is likely to treat the threat with some skepticism."¹⁶²

One option slightly less onerous than program withdrawal is for EPA to take over a state's entire enforcement function. Such federal "assumption of enforcement" is provided for by both the CWA and the CAA, whereby violations of permits or requirements of CAA state implementation plans (SIPs) are so widespread that they appear to result "from a failure of the State to enforce" effectively.¹⁶³ After giving a state 30 days' notice of such failure (90 days in the case of Title V permit violations under the CAA), EPA is authorized to enforce any permit condition or SIP requirement with respect to any person by issuing a compliance order or bringing an enforcement action. While in theory less intrusive than complete program withdrawal, in practice the same political and resource constraints limiting program withdrawal also make significant use of this option unrealistic.¹⁶⁴

More realistic options include those discussed in the section above on differential oversight. EPA should audit more permits, inspections, and enforcement actions in weaker states. Poor performing states also should have enhanced real-time permit reviews, closer examination of priority setting, and increases in federal inspections and direct enforcement activity. They should also have less flexibility in obtaining federal funding. (As an aside, we note that the OECA's ability to encourage desired state performance through financial incentives is hampered by the fact that it generally does not have control of the purse strings for grants to states. Rather, funding is controlled by the various, media-specific program offices at EPA (air, water, solid waste pesticides and toxic substances, among others), from which the OECA is organizationally distinct. It would be helpful if EPA program officials responsible for funding receive significant input from enforcement staff, and pertinent recommendations for improving state enforcement performance are made part of a state's grant conditions.)

EPA also can work more intensively with states that are deficient to help them meet necessary requirements. One option that EPA does not currently take full advantage of involves upgrading the Agency's contacts with state attorneys general, or in some states, state auditors. EPA contacts with states tend to rely on the state environmental agency as the hub, at least based on our experience. In most states, the attorney general is an independently elected official. An attorney general in a state that EPA believes is lax on enforcement potentially has much to gain from "being part of the solution" of enhancing such enforcement. Such an attorney general can point to EPA concerns about the lack of enforcement and take the position that his/her office is taking action to ensure that pollution does not pay. Further, such an attorney general can minimize criticism associated with taking an aggressive stance on enforcement by suggesting that his/her office will keep the federal government out and thereby produce better, home-grown solutions to the state's environmental problems. On the other hand, because of the potential political sensitivity of such contacts (especially where the attorney general and governor are from different political parties), EPA may be reluctant to pursue this approach very often.

Finally, we believe that reshaping the withdrawal process could result in additional pressure for improved state performance. In particular, current rules set forth time frames for the resolution of withdrawal actions once commenced by EPA, but not for a determination by EPA about whether to accept or reject a citizen withdrawal petition.¹⁶⁵ We believe that EPA should be required to investigate and issue a determination about whether it will take action based on citizen petitions within a specified time frame. Such final agency actions (including failures to act) then would be subject to judicial review. This is different from the current situation, in which petitions can, and often do, languish indefinitely within the agency, with the petitioners being powerless to do anything to expedite action. We also think that it is worth exploring the feasibility of allowing greater citizen participation in the investigative process that EPA initiates in response to citizen withdrawal petitions. For example, it is worth considering whether, assuming certain thresholds for a citizen petition are met, e.g., EPA determines that a petition raises meritorious allegations of inadequate state performance, the citizen petitioners should be allowed to supplement EPA's investigation by requesting information from, interviewing, or otherwise engaging state agency officials, whose practices are at issue. The practicalities of this proposal need to be further examined, but just as citizen enforcement of environmental laws has proven to be a useful adjunct to government enforcement, we believe that greater citizen involvement has the potential to invigorate the withdrawal process. In particular, citizen groups may be more forceful in investigating poor state performance than EPA, especially since withdrawal petitions are investigated by EPA's regional offices, with whom states often enjoy close relationships.

EPA Should Enhance Enforcement in Authorized States

EPA's own resources, albeit limited, give it another tool to influence state behavior. First, EPA can take steps to enhance [33 ELR 10580] the capacity of state programs. Second, EPA should strategically use its own enforcement resources to bolster weak state programs. With regard to the latter option, EPA can expand its own direct enforcement role by more systematically dividing enforcement responsibilities with the various states. Alternatively, EPA can increase the number of "overfilings" and other enforcement actions in states that are not performing to EPA expectations. These are discussed in turn.

EPA Should Take Steps to Enhance the Capacity of State Programs

One important carrot EPA can offer to improve state performance is assistance in building the capacity of state programs. If a state, for example, needs instruction in conducting certain types of inspections, EPA can arrange for state inspectors to be trained in the proper techniques. In FY 2000, the OECA reported that it conducted 463 training courses, made over 6,700 regulatory determinations, responded to 1,527 requests for assistance, and conducted 1,132 assisted inspections, in order to help states, tribes, and localities build capacity. Likewise, EPA can provide analytic support to the states to identify patterns of noncompliance and suggest the best means for targeting their enforcement efforts. As one illustration, EPA reported that in FY 2001, it responded to 10,000 queries per month from state (and tribal) agencies seeking to use the Agency's specialized tracking system for identifying patterns of noncompliance by industry sector, geographic area, and statute.¹⁶⁶ As EPA continues to transfer more of the primary authority for implementing its programs to the states, it should be able to shift more of its program staff from direct implementation to state capacity building. The capacity-building activities initiated by EPA's Region 8, discussed above, are the type that should be part of EPA's efforts systematically to identify state needs and tailor EPA support accordingly. They should be an integral part of the differential oversight scheme discussed above.

EPA Should Divide the Enforcement/Compliance Universe More Systematically With States

Even in states that have received program authorization from EPA, EPA files some of its own, direct enforcement actions. Critics charge, however, that these actions are not as strategic or as well-coordinated with the states as they should be. NEPPS contemplates that EPA and individual states will engage in joint planning and priority setting.¹⁶⁷ It further appears to contemplate the possibility of an active EPA enforcement presence.¹⁶⁸ The idea is that the Regions and states will mutually determine their priorities and the resources available, and then divide responsibilities accordingly. The 1994 Joint Policy Statement on State/EPA Relations likewise reflects an intent to "utilize the comparative advantages and inherent strengths that each party brings to the relationship."¹⁶⁹ As part of joint planning efforts, EPA Regions and states could agree to divide enforcement responsibilities in various ways. EPA, for instance, might take on some or all of the regulated parties in a state under a particular program.¹⁷⁰

Coordinated action based on mutually agreed-upon priorities is likely to work better in advancing environmental protection than is uncoordinated, ad hoc action that fails to target priority concerns. Further, each sovereign could choose those responsibilities it is most committed to fulfilling and for which it is best suited.

In deciding how best to divide the enforcement pie between them, states and EPA Regions would need to determine, among other things: (1) the key priority areas, e.g., industries in which levels of noncompliance are unacceptably high and in which violating sources are causing significant, adverse environmental impacts; (2) the interest and capacity of EPA and the relevant state in addressing the particular areas of concern, e.g., the number of inspectors compared to the size of the regulated party universe, the number of lawyers available to bring cases, etc.; (3) strategic considerations that might favor federal or state primacy in particular situations, e.g., the value of reserving EPA for a "gorilla in the closet"—type role that the state could use to leverage appropriate settlements; and (4) the need for federal resources to make up for shortfalls in state performance (if EPA believes that it is important to physically inspect 100 facilities in a particular state one year, and the state only is comfortable committing to inspect 75, EPA can use its own staff to inspect the remaining 25). As a general matter, certain types of cases seem particularly appropriate for direct EPA enforcement—cases where it is important to establish a national precedent; against an entire industrial sector; cases against a single defendant with facilities in multiple states; or cases that are especially complex or resource-intensive.¹⁷¹

As with the option of program withdrawal, there are real practical and political constraints to the above strategy. For one, EPA may lack the resources to do direct enforcement on a wide scale. Second, while some states may embrace an enhanced EPA presence as a means of alleviating enforcement resource constraints, many others may view any direct federal enforcement presence as undercutting the sacrosanct [33 ELR 10581] notion of state primacy. One state official testified that "when EPA brings a direct enforcement action in a state, there is often concern that the principle setting forth the primary role of the state has been violated."¹⁷² Prof. Robert Kuehn captured the more recent mood of many states when he wrote that

environmental enforcement's gorilla is facing extinction or, perhaps more accurately, execution or starvation …. Federal environmental agencies find themselves subject to increasing calls to reduce dramatically, or even eliminate, their enforcement roles. Once seen as a gorilla in the closet whose threatened release could persuade violators to comply with the law, many politicians and commentators now view federal enforcement as an unwarranted intrusion into the efforts of capable state enforcement agencies.¹⁷³

Nonetheless, EPA needs to maintain a direct enforcement capacity in authorized states, and it needs systematically to use it to address what it considers to be the most significant enforcement needs that otherwise will not be handled effectively by the states.

EPA Should Continue to Overfile Strategically

Overfiling refers to the filing of a suit by EPA against an alleged violator even though the state already has initiated its own enforcement action against the party alleging the same violations. EPA initiates such actions when in its view the state action constitutes an inadequate enforcement response, such as "where the state response to a violator or environmental condition fails to protect human health or the environment, fails to deter future violations by a major repeat violator, or fails to protect law abiding facilities from competitive disadvantage."¹⁷⁴ We believe that the strategic use of overfilings is an important oversight tool that EPA should continue to employ. As one detailed analysis of the practice concludes, "overfiling provides the EPA with a realistic way to nip at the heels of the states to properly implement federal prerogatives."¹⁷⁵

While the record of state performance indicates that EPA has had ample opportunity for overfilings,¹⁷⁶ EPA only rarely overfiles, as former EPA Assistant Administrator Steven Herman made clear in his June 1997, congressional testimony:

Statistics show that overfiling is in fact a rare event. As reported by a state-by-state survey conducted by ECOS, the Agency overfiled on about 30 cases or 0.3% of all federal enforcement action during [FY] 1992 through [FY] 1994. During [FY] 1994 and [FY] 1995, the Agency overfiled on a total of 18 cases or about 0.1% of state enforcement cases. From October 1995 through September 1996, there was a total of four overfiling cases.¹⁷⁷

In approximately the first two years of the Bush Administration, EPA overfiled in three instances.¹⁷⁸

There are many reasons for the reality of few overfilings. "Double-teaming" violators is problematic from a government resources perspective. Further, overfilings create a basic "fairness" issue from the perspective of the regulated community. Moreover, overfilings, by their very nature, place considerable stress on the notion of a strong EPA/state partnership. They represent a clear federal rebuke of state performance and therefore are embarrassing to the states. Oklahoma Executive Director Mark Coleman's June 1997, congressional testimony captures states' distaste for federal overfiling practices:

Although the instances of EPA overfiling are relatively few, the possibility of overfiling and the use of overfiling comes at a great cost. The potential for overfiling leads to mutual wariness and if not done with extreme care it can rapidly damage the enforcement relationship.¹⁷⁹

Indeed, even a very few overfilings provoke strong state opposition. In Utah, as a result of three overfilings within a two-year period, a nonbinding resolution was introduced in 1999 in the legislature calling on Congress to investigate EPA overfiling because of its adverse impact on "working relationships between the EPA and the states."¹⁸⁰ A resolution adopted by ECOS in October 1998, calls for overfiling to be limited to a narrow set of circumstances that do not include a state obtaining an insufficient penalty, no matter how inadequate.¹⁸¹ A final factor potentially impacting [33 ELR 10582] EPA's use of overfilings is the preference evinced by NEPPS for programmatic, rather than case-specific, reviews. If this type of review becomes more of the norm in all states, even poorer performing states, it will be harder for EPA to learn of specific enforcement actions in which it should overfile.

However politically challenging, we believe that there are advantages to greater use of overfilings. First, overfilings are a far more strategic and less intrusive and resource-intensive form of intervention than wholesale program withdrawal. Indeed, before overfiling EPA must give the state and the violator adequate notice, a requirement that two state officials argue has "the practical effect of [providing] more state control over enforcement actions."¹⁸² (Under the CAA and the CWA, EPA is required to notify both the violator and the state at least 30 days before overfiling. RCRA requires advance notice but does not specify the number of days.)¹⁸³

Second, overfilings can be quite effective. Being sued once for a set of alleged violations is not particularly pleasant. Ending up on the receiving end of a second lawsuit aimed at the same set of violations is likely to raise the anxiety level of the alleged violator. Overfilings thus are likely to get the attention of the regulated community very quickly and have a ripple effect extending well beyond the particular alleged violators targeted, particularly since EPA's penalty demands are likely to be much higher than those of the state. State enforcers may be able to more credibly pursue higher sanctions to avoid the prospect of multiple suits. Overfilings can be a particularly effective check on local political (or other) pressures that lead to weak state enforcement practices or sweetheart settlements. One example is the well-publicized EPA overfiling in 1996 against Smithfield Foods in which EPA acted after Virginia failed to assess meaningful penalties for over 5,000 violations by Smithfield of its permit. In that case, Smithfield was contributing to the governor's political action committee, and also repeatedly threatened to leave Virginia if state regulators were too strict.¹⁸⁴ EPA audits have cited other instances in which state and local agencies acknowledged that they were reluctant to take strong enforcement actions because of fear that companies might close or chose to relocate elsewhere.¹⁸⁵

Finally, because overfilings are anathema to states, an EPA policy to overfile strategically may lead states to upgrade their efforts in order to limit EPA "interference." Some states may reassess their own practices to respond to inevitable pressure from regulated parties concerned about the emergence from the closet of an EPA gorilla that is much more fearsome than the state enforcement arm, or their concern about embarrassment due to public attention, including from environmental nongovernmental organizations galvanized to focus on states' performance in this area. An EPA overfiling, and even more likely a public EPA decision to overfile in specified states on a regular basis because of purported deficiencies in state enforcement practices, is likely to place a spotlight on such practices.

Conclusion

Because EPA's traditional methods of oversight have left a sizeable gap between state enforcement performance and EPA expectations, new oversight approaches are needed. This Article outlines a number of ideas for improving state performance and making the federal/state relationship more effective. These recommendations can work regardless of whether the "governing document" for the EPA/state enforcement relationship is NEPPS, the traditional Policy Framework, or some combination of the two. Some of the recommendations explicitly build on the themes embraced by NEPPS; others draw on more traditional oversight tools, including ones that have not been used with great effectiveness in the past. The underlying theme to many of these is greater state accountability, both to EPA and to the public, starting at the initial authorization or gatekeeping stage, extending through regular reviews of state performance and a system of differential oversight, to enhanced oversight and engagement by the public through a spotlighting approach. In our view, the lack of accountability has been a significant shortcoming in the state/EPA relationship to date, and we strongly endorse greater reliance on a "spotlight" strategy.¹⁸⁶ In this age of increasing accountability, it is time for EPA and the states to make themselves publicly accountable for their enforcement performance.

Many of these options are likely to face considerable political resistance, particularly in the current political climate, because some might be inclined to characterize them as casting EPA in more of the traditional role of "overseer" of state programs than as a true "partner" of the states. We do not minimize the concerns that this criticism raises or the importance of EPA maintaining a cooperative relationship with the states to achieve the goals set forth by our environmental statutes. But the approach we advocate is more likely to achieve better state performance because it conditions the reward of full partnership upon honest achievement of EPA expectations. Indeed, well performing states would be rewarded with even fuller partnership than is available under the current system. The privileges and benefits of partnership should come only with increased accountability to EPA and its expectations.

None of the approaches outlined in this Article is likely to be a silver bullet. In the absence of experimentation with them, however, the gap between expectation and performance is likely to widen, particularly in a political landscape heavily tilted toward devolution of authority to the states.

1. Press Release, Environmental Council of the States (ECOS), States Administer Most Environmental Problems (Feb. 1, 1999), available at http://www.sso.org/ecos/news.htm (last visited Feb. 1, 1999).

2. David Hodas, Enforcement of Environmental Law in a Triangular Federal System: Can Three Not Be a Crowd?, 54 MD. L. REV. 1552, 1574 (1995). Prof. Richard Stewart makes the same observation in an insightful early article. Richard B. Stewart, Pyramids of Sacrifice? Problems of Federalism in Mandating State Implementation of National Environmental Policy, 86 YALE L.J. 1196 (1977).

3. A number of estimates suggest that the rate of noncompliance among regulated entities is in the range of 20% to 40%. See David Spence, The Shadow of the Rational Polluter: Rethinking the Role of Rational Actor Models in Environmental Law, 89 CAL. L. REV. 917, 966-67 (2001); Daniel A. Farber, Taking Slippage Seriously: Noncompliance and Creative Compliance in Environmental Law, 23 HARV. ENVTL. L. REV. 297, 304-05 (1999). As Professor Farber and others have observed, the "optimum" level of compliance, and the relationship between compliance and the contents of underlying standards, are matters of some debate. Id. at 316-17.

4. See Robert R. Kuehn, The Limits of Devolving Enforcement of Federal Environmental Laws, 70 TUL. L. REV. 2373, 2376-78 (1996). For example, Prof. Victor Flatt showed in a case study that regulated entities in one state were allowed to remain out of compliance with Clean Water Act (CWA) requirements for about 50% longer than those in another state. Victor Flatt, A Dirty River Runs Through It (The Failure of Enforcement in the Clean Water Act), 25 B.C. ENVTL. AFF. L. REV. 1, 26 (1997). State officials recognize the value of consistent state enforcement. One New York official told Congress in 1998 that "EPA must act to guard against disparities in enforcement among delegated states," noting "the potential for lost revenues, relocation of businesses, and competitive advantages" from the raft of audit privilege and immunity laws that had been enacted by other states. Testimony of Linda Spahr, Chair, Environmental Subcomm. of New York State District Attorneys Association, Hearing on Environmental Audits and the Federal-State Relationship, House Commerce Comm., Subcomm. on Oversight and Investigation (Mar. 17, 1998). Prof. Richard Revesz and others are not necessarily troubled by the theory or reality of an "unlevel playing field" and indeed suggest that it may be a good thing in some cases. See, e.g., Richard L. Revesz, Rehabilitating Interstate Competition: Rethinking the "Race-to-the-Bottom" Rationale for Federal Environmental Regulation, 67 N.Y.U. L. REV. 1210, 1253 (1992).

5. Rodgers argues that EPA has been strongly constrained from withdrawal "out of a reluctance to provoke political retaliation or to burn bridges if there is any hope of salvaging the relationship." WILLIAM H. RODGERS JR., 4 ENVIRONMENTAL LAW: HAZARDOUS WASTES AND SUBSTANCES § 7.22 (1992).

6. See U.S. EPA, Joint Commitment to Reform Oversight and Create a National Environmental Performance Partnership System, at http://www.epa.gov/ocirpage/nepps/ovrsight.htm (last visited May 15, 2003) [hereinafter Joint Commitment to Reform Oversight and Create a National Performance Partnership System].

7. For detailed evaluations of the impact of NEPPS thus far, see LEROY PADDOCK & SUELLEN KEINER, MIXING MANAGEMENT METAPHORS: THE COMPLEXITIES OF INTRODUCING A PERFORMANCE-BASED STATE/EPA PARTNERSHIP SYSTEM INTO AN ACTIVITY-BASED MANAGEMENT CULTURE (National Academy of Pub. Admin. (NAPA) 2000), available at http://www.napawash.org/pc_economy_environment/learning_texts.html (last visited May 15, 2003); see also JEANNE HERB ET AL., THE NATIONAL ENVIRONMENTAL PARTNERSHIP SYSTEM: MAKING GOOD ON ITS PROMISE? (NAPA 2000), available at http://www.napawash.org/pc_economy_environment/learning_texts.html (last visited May 15, 2003).

8. NAPA, ENVIRONMENT.GOV: TRANSFORMING ENVIRONMENTAL PROTECTION FOR THE 21ST CENTURY 140, 145 (2000) [hereinafter ENVIRONMENT.GOV]; OFFICE OF INSPECTOR GENERAL (OIG), U.S. EPA, WATER ENFORCEMENT: STATE ENFORCEMENT OF CLEAN WATER ACT DISCHARGERS CAN BE MORE EFFECTIVE (2001) (2001-P-00013, 58-59) [hereinafter U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT].

9. PADDOCK & KEINER, supra note 7, at 26.

10. To cite one example, EPA's OIG reported that "officials from Louisiana and Illinois believed that legal decisions and case law prevented them from calculating or assessing the economic benefit," yet EPA has granted authorizations to programs in those states. OIG, U.S. EPA, CONSOLIDATED REVIEW OF THE AIR ENFORCEMENT AND COMPLIANCE ASSURANCE PROGRAMS 25, 28 Audit No. EIGAE5-05-0169-7100306 (1997) [hereinafter OIG, CONSOLIDATED REVIEW]. See generally U.S. GENERAL ACCOUNTING OFFICE (GAO), ENVIRONMENTAL ENFORCEMENT: PENALTIES MAY NOT RECOVER ECONOMIC BENEFITS GAINED BY VIOLATORS 15 (1991) (GAO/RCED-91-166).

11. ENVIRONMENTAL LAW INST., REPORT OF THE COLLOQUIUM ON FEDERAL-STATE RELATIONS IN ENVIRONMENTAL ENFORCEMENT 7 (1991).

12. WILLIAM H. RODGERS JR., ENVIRONMENTAL LAW 367 (Hornbook Series 2d ed. 1994) [hereinafter RODGERS, HORNBOOK SERIES].

13. See, e.g., U.S. EPA, POLICY ON CIVIL PENALTIES (1984), ADMIN. MAT. 35083.

14. ENVIRONMENTAL LAW INST. (ELI), LAW OF ENVIRONMENTAL PROTECTION § 6.03[2] (Sheldon M. Novick ed., Clark Boardman Callaghan 1999) [hereinafter ELI, LAW OF ENVIRONMENTAL PROTECTION].

15. One could certainly argue, however, that the requirement in the current authorization regulations that state civil penalties be "appropriate" to the violation, allows EPA to withhold authorization from states which do not recover economic benefit. See, e.g., 40 C.F.R. § 123.27(c) (CWA national pollution discharge elimination system (NPDES) permit program); id. § 145.13(c) (Safe Drinking Water Act (SDWA) underground injection control program); § 271.16(c) (Resource Conservation and Recovery Act (RCRA) hazardous waste program).

16. They provide, for example that authorized state programs "provide adequate enforcement of compliance with the [law's] requirements," (RCRA); have adequate authority to "assure compliance by all sources required to have a permit" (Title V of the Clean Air Act (CAA)); or have adequate authority "to abate violations of the permit or permit program" (CWA). See 42 U.S.C. § 6926(b). ELR STAT. RCRA § 3006(b); id. §§ 7661a(B)(5)(A), ELR STAT. CAA § 502a(b)(5)(A); 33 U.S.C. § 1342(b)(2)(B), (b)(7), ELR STAT. FWPCA § 402(b)(2)(B), (b)(7).

17. See, e.g., Natural Resources Defense Council v. EPA, 859 F.2d 1565, 19 ELR 20016 (D.C. Cir. 1988).

18. 42 U.S.C. § 7420(a)(2)(A)(ii), (iv)(d)(2), ELR STAT. CAA § 120(a)(2)(A)(ii), (iv)(d)(2).

19. 40 C.F.R. § 142.17(a)(1).

20. At present one likely would have to obtain this information from multiple EPA offices.

21. Steve Cook, EPA Budget Request Includes Staff Cut at Enforcement Office, Grants to States, 33 Env't Rep. (BNA) 286 (Feb. 8, 2002).

22. Robert F. Blomquist, "Turning Point": The Foundering of Environmental Law and Policy in Indiana?, 27 IND. L. REV. 1033, 1034 (1994).

23. DENISE SCHEBERLE, FEDERALISM AND ENVIRONMENTAL POLICY: TRUST AND THE POLITICS OF IMPLEMENTATION 7, 54 (1997); Due to Fiscal Woes, States Warn They May Be Unable to Meet New EPA Goals for 2004, INSIDE EPA, Jan. 24, 2003, at 1.

24. U.S. EPA, DRAFT REPORT ON U.S. EPA REVIEW OF OHIO ENVIRONMENTAL PROGRAMS CAA-20 (2001) [hereinafter DRAFT EPA REVIEW OF OHIO ENVIRONMENTAL PROGRAMS]. Under the CAA, states are required to impose permit fees on regulated sources sufficient to fund the costs of administering and enforcing their Title V permit programs. 42 U.S.C. § 7661a(b)(3)(A), ELR STAT. CAA § 502(b)(3)(A).

25. See 42 U.S.C. § 7661a(b)(6), ELR STAT. CAA § 502(b)(6) (requiring that state permit programs include an opportunity for judicial review of permit decisions by any person who participates in the public comment process). Specifically, under the Maryland Environmental Standing Act, MD. CODE ANN., NAT. RES. I §§ 1-501 to 1-508 (2001), "persons" eligible to seek judicial review of the Department of Environment's issuance of an operating permit under Title V did not include nonstate residents or organizations doing business outside of Maryland. Rather, these parties were required to establish standing for judicial review under the Maryland common law of standing, which requires a party to demonstrate that it has a "specific interest or property right" such that the party will suffer harm that is different in kind from that suffered by the general public, and which EPA concluded was more restrictive than federal constitutional standing. See 60 Fed. Reg. 55231, 55233 (Oct. 30, 1995).

26. 66 Fed. Reg. 63236 (Dec. 5, 2001).

27. Steve Cook & Martha Kessler, EPA Assumes Control of Air Permitting Because of Inadequate Participation Process, 32 Env't Rep. (BNA) 2368, 2368 (Dec. 7, 2001).

28. See 68 Fed. Reg. 1974 (Jan. 15, 2003). EPA had control of the Title V program in Maryland from December 1, 2001, until February 14, 2003.

29. EPA explicitly spelled this out, for example, in a revised enforcement response policy for high priority (previously called significant) air quality violations issued in 1998. OFFICE OF ENFORCEMENT & COMPLIANCE ASSURANCE (OECA), U.S. EPA, THE TIMELY AND APPROPRIATE (T&A) ENFORCEMENT RESPONSE TO HIGH PRIORITY VIOLATORS (HPVs) (1998); see also U.S. EPA, FINAL FISCAL YEAR (FY) 2000/2001 OFFICE OF COMPLIANCE ASSURANCE MEMORANDUM OF AGREEMENT GUIDANCE attach. 4, at 7 (1999). See infra note 34 and accompanying text.

30. Mark Stoughton & Jennifer Sullivan, Mixed Results, ENVTL. F., May/June 2001, at 44, 45.

31. John Dwyer, The Practice of Federalism Under the Clean Air Act, 54 MD. L. REV. 1183, 1224 (1995).

32. CLIFFORD RECHTSCHAFFEN & DAVID L. MARKELL, REINVENTING ENVIRONMENTAL ENFORCEMENT & THE STATE/FEDERAL RELATIONSHIP ( [ILLEGIBLE WORD]. L. Inst. 2003).

33. See WILLIAM T. GORMLEY JR., NAPA, ENVIRONMENTAL PERFORMANCE MEASURES IN A FEDERAL SYSTEM 14 (2000).

34. U.S. EPA, REVISED POLICY FRAMEWORK FOR STATE/EPA ENFORCEMENT AGREEMENTS (1986).

35. Consider, for example, an attempt to link attainment of the CAA's health-based standards for criteria air pollutants, or national ambient air quality standards (NAAQS)—the bottomline objective of conventional air pollution control regulation—with individual enforcement actions. Under the CAA, states develop state implementation plans (SIPs) that impose controls on thousands of stationary sources and motor vehicles in order to achieve NAAQS in a given air quality region. In addition to an agency's enforcement efforts, any one of a myriad of factors can influence what ambient air conditions are in a region. Some of these include weather conditions, economic growth, population changes, closure of a major facility, pollution from neighboring regions or states, changes in driving patterns due to fluctuating gasoline prices, or even a local mass transit strike. Regulated facilities may be in compliance but discharge more (lawful) emissions than in prior years, due to higher production schedules, equipment breakdown, or less than optimal performance of pollution control equipment. Unregulated facilities or facilities under the regulatory radar may offset any gains from enforcement actions. Existing controls may be insufficiently stringent to lower ambient concentrations; air quality modeling assumptions predicting improvements based on enforcing permitted levels may be incorrect. The list of potential confounding factors is extensive.

36. R. STEVEN BROWN & VALERIE GREEN, STATE ENVIRONMENTAL AGENCY CONTRIBUTIONS TO ENFORCEMENT AND COMPLIANCE 51 (ECOS 2001).

37. GORMLEY, supra note 33, at 12, 15.

38. Id. at 16.

39. Shelley H. Metzenbaum & Tiffin Shewmake, The Search for More Nutritious Enforcement Beans 4-6, 10-11 (Environmental Compliance Consortium, draft 2002), at http://www.complianceconsortium.org [hereinafter Metzenbaum & Shewmake, More Nutritious Enforcement Beans].

40. Id. They maintain that another way to attempt to isolate the effects of agency actions is to measure the "place-specific" trends in ambient conditions for given pollutant(s), and simultaneously look at discharge trend lines for facilities in that geographic area discharging the pollutant(s). This analysis could suggest whether problems in ambient conditions are the result of noncompliance with permits, the need for more protective permits, or the need to control unpermitted discharges. Id. at 16-17.

41. OFFICE OF WATER, U.S. EPA, THE QUALITY OF OUR NATION'S WATER: A SUMMARY OF THE NATIONAL WATER QUALITY INVENTORY: 1998 REPORT TO CONGRESS 5 (2000) (EPA 841-S-0-001), available at http://www.epa.gov/305b/98report/98brochure.pdf.

42. Using biological surveys make assessments more accurate, because they can often detect effects of pollutants that otherwise would not be identified with chemical tests alone. For instance, Ohio's EPA found that 50% of its water bodies assessed with chemical tests alone and identified as not impaired actually were impaired after biological tests were conducted. OIG, U.S. EPA, OHIO'S WATER QUALITY PROGRAM (1999), available at http://www.epa.gov/oigearth/ereading_room/list999/9p00210.htm.

43. PUBLIC EMPLOYEES FOR ENVIRONMENTAL RESPONSIBILITY, MURKY WATERS: OFFICIAL WATER QUALITY REPORTS ARE ALL WET. AN INSIDE LOOK AT EPA'S IMPLEMENTATION OF THE CLEAN WATER ACT (1999) (Public Employees for Environmental Responsibility is an organization of current and former EPA and state environmental agency employees). Other reports have reached similar conclusions. See, e.g., OIG, U.S. EPA, REGION III WATER QUALITY STANDARDS, MONITORING, AND REPORTING 7-8 (1999) (E1HWF7-03-0160-9199118).

44. MICHIGAN ENVIRONMENTAL COUNCIL, GREENING THE GOVERNMENTS: ASSESSING THE ENVIRONMENTAL CONDITIONS AND PERFORMANCE OF THE GREAT LAKES STATES (2002), available at http://www.mecprotects.org/greening.pdf [hereinafter GREENING THE GOVERNMENTS]. The report noted "this is in part due to federal requirements imposed by EPA and also because of a lack of state initiatives to coordinate data collection and reporting."

45. NAPA, EVALUATING ENVIRONMENTAL PROGRESS: HOW EPA AND THE STATES CAN IMPROVE THE QUALITY OF ENFORCEMENT AND COMPLIANCE INFORMATION (2001) [hereinafter NAPA, How EPA AND THE STATES CAN IMPROVE].

46. There have been calls for them to do so, such as a recent report's recommendation that the Great Lakes states develop a "Right to Know Compact." See GREENING THE GOVERNMENTS, supra note 44, at 67.

47. NAPA, ENVIRONMENT.GOV, supra note 8, at 151.

48. Under NEPPS, each state has the option of negotiating a PPA with EPA that covers matters such as which environmental problems within the state will receive priority attention, the respective roles of the state and EPA, and how the state's progress in meeting its goals will be measured. These PPAs are intended to replace EPA's traditional oversight agreements with the states.

49. NAPA, ENVIRONMENT.GOV, supra note 8, at 140, 145; U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 58-59.

50. ECOS reported that in 1996, total state environmental spending was $ 13.82 billion, while in 2000, the figure was $ 13.6 billion. R. Steven Brown, States Put Their Money Where Their Environment Is, at http://www.sso.org/ecos/ECOStatesArticles/rsbrown.pdf (last visited June 28, 2002).

51. Nonetheless, observing trends in discharges from regulated facilities can help pinpoint whether ongoing environmental problems result from regulated or unregulated sources.

52. See BROWN & GREEN, supra note 36, at 42 (expressing state preferences for using compliance rates as performance measure).

53. For example, in 1998, EPA estimated that in 27 states, more than 20% of major sources regulated under the CWA's NPDES were operating under expired permits. See Rena I. Steinzor, Devolution and the Public Health, 24 HARV. ENVTL. L. REV. 351, 384 (2000). By March 2000, the backlog of expired permits had increased in 25 states. Rena I. Steinzor, EPA and Its Sisters at 30: Devolution, Revolution, or Reform?, 31 ELR 11086, 11088-89 (Sept. 2001).

54. OECA, U.S. EPA. FY 2000 ACCOUNTABILITY MEASURES FOR ENFORCEMENT AND COMPLIANCE ASSURANCE (2001), available at http://www.epa.gov/ocir/nepps/enflist.html.

55. This point is made in BROWN & GREEN, supra note 36, at 40.

56. OIG, U.S. EPA, AUDIT OF REGION 9'S ADMINISTRATION OF THE CALIFORNIA AIR COMPLIANCE AND ENFORCEMENT PROGRAM 7 (1997) (EIGA06-06-0023-7100246). For example, an EPA audit found that a local air quality control board in California considered a facility in compliance if it immediately repaired a piece of equipment or submitted a permit application even where the facility had previously violated the same requirements. In EPA's view, the repeat violations indicated that the facility was not achieving continuous compliance.

57. NAPA, HOW EPA AND THE STATES CAN IMPROVE, supra note 45, at 38. One study of compliance with various air regulatory requirements concluded that "it would be misleading or useless to attempt to compare compliance rates across states" because of a series of problems, including "the lack of accurate, reliable, commonly accepted compliance indicators, including concerns over differences in facility size and complexity, levels and quality of inspections, severity of noncompliance, practices for resolving violations informally, etc." OECA, U.S. EPA, COMPLIANCE INFORMATION PROJECT: LITERATURE SUMMARIES 44-47 (1999) (EPA-300-R-99-002) (summarizing Woodward-Clyde International Americas, Colorado Compliance Study (Aug. 1, 1997)).

58. These variables include, for example, the number and percentage of facilities inspected or otherwise monitored, the reasons for inspections or compliance monitoring actions, and the methodology for selecting facilities to monitor, i.e., statistically valid, targeted, etc. See Metzenbaum & Shewmake, More Nutritious Enforcement Beans, supra note 39, at 18-19.

59. In 1997, EPA unveiled its National Performance Measures Strategy (NPMS), an approach for evaluating its own enforcement program that includes environmental indicators, outcome measures, and agency enforcement activities. It has been gradually implementing various measures identified in the NPMS since then. U.S. EPA, MEASURING THE PERFORMANCE OF EPA'S ENFORCEMENT AND COMPLIANCE ASSURANCE PROGRAM (1997).

60. See, e.g., U.S. EPA, FY 2001 ANNUAL REPORT II-85 (2002), available at http://www.epa.gov/budget/finstatement/2001ar/2001ar.htm#2001 [hereinafter U.S. EPA, FY 2001 ANNUAL REPORT].

61. In its July 2001 report, for example, NAPA recommended that EPA and the states "should stop relying primarily on activities data to evaluate the effectiveness of environmental protection programs, including compliance assistance and enforcement initiatives…." NAPA, HOW EPA AND THE STATES CAN IMPROVE, supra note 45, at 2.

62. Metzenbaum and Shewmake point out that

in the near term, many agencies will not have the record-keeping systems (neither the computer systems, record-keeping, or recording procedures) needed to calculate inspection (compliance monitoring action) and compliance rates. This requires the ability to track and tally actions relative to specific facilities. Until agencies build record-keeping facilities that allow them to do that easily, some agencies will only be able to count their activities.

Metzenbaum & Shewmake, More Nutritious Enforcement Beans, supra note 39, at 22.

63. Likewise, according to EPA Region 4, as of 2001, none of the states in its Region were reporting on three of the NEPPS accountability measures that the Region had designated as "pilot" measures, including two outcome-based measures: environmental and public health benefits resulting from enforcement activities; results of using alternative compliance approaches; and the number of facilities reached by compliance assistance activities. See U.S. EPA, REGION IV SUMMARY OF STATE CORE PERFORMANCE MEASURES (2001), available at http://www.complianceconsortium.org (under "Related Links" and "Environmental Performance Measures and Reports").

64. Region 4 anticipates that future protocols for reviews will include some pieces of environmental outcome and/or environmental result data, although Region 4 intends that this type of information will complement the output-oriented data that the states and Region 4 develop and review now.

65. Clifford Rechtschaffen, Competing Visions: EPA and the States Battle for the Future of Environmental Enforcement, 30 ELR 10803, 10815-18 (Oct. 2000) [hereinafter Competing Visions].

66. It may be the case, for example, that careful record review and follow-up calls of self-reported monitoring submitted by regulated entities could have a significant deterrent effect.

67. One weakness in this type of measure is that it does not pick up the percentage of parties that are outside the regulatory universe and have failed to file for permits, the so-called nonfilers. In certain regulatory programs, especially where there are significant numbers of small businesses, pollution from these nonfilers can be a considerable problem.

68. See Competing Visions, supra note 65, at 10813-15.

69. GORMLEY, supra note 33, at 42.

70. SHELLEY METZENBAUM, MAKING MEASUREMENT WORK: THE CHALLENGE AND PROMISE OF BUILDING A PERFORMANCE-FOCUSED ENVIRONMENTAL PROTECTION SYSTEM 47-49 (Brookings Inst. Center for Pub. Mgmt. 1998), available at http://www.brook.edu/dybdocroot/gs/cpm/metzenbaum.pdf [hereinafter METZENBAUM, MAKING MEASUREMENT WORK].

71. U.S. EPA, MEASURING THE PERFORMANCE OF EPA'S ENFORCEMENT AND COMPLIANCE ASSURANCE PROGRAM, DRAFT REPORT OF THE NATIONAL PERFORMANCE STRATEGIES 8 (1997).

72. We thank Prof. Shelley Metzenbaum for bringing this point to our attention.

73. BROWN & GREEN, supra note 36, at 26.

74. DRAFT EPA REVIEW OF OHIO ENVIRONMENTAL PROGRAMS, supra note 24, at CAA-34.

75. U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 38.

76. GREEN MOUNTAIN INST. FOR ENVIRONMENTAL DEMOCRACY, THE RESOURCE GUIDE TO INDICATORS (2d ed. 1998), available at http://www.gmied.org/PUBS/papers/inddocs/irguide.html.

77. NAPA, HOW EPA AND THE STATES CAN IMPROVE, supra note 45, at 4.

78. GREENING THE GOVERNMENTS, supra note 44, at 19.

79. See Joint Commitment to Reform Oversight and Create a National Environmental Performance Partnership System, supra note 6.

80. PADDOCK & KEINER, supra note 7, at 13, 60.

81. Eric Helland, The Revealed Preferences of State EPAs: Stringency, Enforcement, and Substitution, 35 J. ENVTL. ECON. & MGMT. 242, 259-60 (1998).

82. DRAFT EPA REVIEW OF OHIO ENVIRONMENTAL PROGRAMS, supra note 24, at CAA-20.

83. U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 39.

84. ENVIRONMENTAL LAW INST., FEDERAL OVERSIGHT OF AUTHORIZED STATE ENVIRONMENTAL PROGRAMS: REFORMING THE SYSTEM 15 (1995) [hereinafter OVERSIGHT OF AUTHORIZED STATE ENVIRONMENTAL PROGRAMS].

85. Professor Markell worked in Region 1 from 1984-1987.

86. U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 54.

87. Id. at ii.

88. A copy of the information can be found at http://www.compliance consortium.org, under "Related Links" and "Environmental Performance Measures and Reports." (The measures for evaluating state enforcement and compliance assurance programs under NEPPS are actually called "accountability measures.")

89. GREENING THE GOVERNMENTS, supra note 44.

90. E-mail message from Christine Lehnertz, U.S. EPA, Region 8 (Apr. 29, 2002). The Inspector General audit found that Pennsylvania routinely failed to identify significant violators or follow EPA enforcement policies in pursuing follow-up concerning these facilities. OIG, U.S. EPA, VALIDATION OF AIR ENFORCEMENT DATA REPORTED TO EPA BY PENNSYLVANIA (1997) (E1KAF6-03-0082—710015). Following the audit, the OECA directed that each Region conduct an audit of state performance and its oversight of such performance.

91. Telephone Interview with Christine Lehnertz, U.S. EPA, Region 8 (Aug. 27, 2001) [hereinafter Telephone Interview with Christine Lehnertz].

92. EXECUTIVE SUMMARY, STATE OVERSIGHT SUMMIT, UNIFORM ENFORCEMENT OVERSIGHT SYSTEM (UEOS) (Dec. 5-6, 2000) (on file with authors).

93. As one report notes, EPA "oversight has evolved into a web of EPA regulations, guidelines, and policy statements. In addition, EPA regional offices often provide guidance statements on oversight requirements under specific programs. State-EPA agreements (SEAs) … also set out oversight obligations state by state for specific authorized programs." OVERSIGHT OF AUTHORIZED STATE ENVIRONMENTAL PROGRAMS, supra note 84, at 13.

94. Telephone Interview with Christine Lehnertz, supra note 91.

95. Id. Another benefit reported is that it has engendered greater consistency and fairness in regional reviews, since the reviews are not determined by the particular views of employees in individual programs. Id.

96. OVERSIGHT OF AUTHORIZED STATE ENVIRONMENTAL PROGRAMS, supra note 84, at 23.

97. Telephone Interview with Christine Lehnertz, supra note 91.

98. Id.

99. Id.

100. See Competing Visions, supra note 65, at 10807-09; David L. Markell. The Role of Deterrence-Based Enforcement in a "Reinvented" State/Federal Relationship: The Divide Between Theory and Reality, 24 HARV. ENVTL. L. REV. 1 (2000).

101. ENVIRONMENT.GOV, supra note 8, at 175.

102. PADDOCK & KEINER, supra note 7, at 66.

103. See United States Attorney's Manual, Title 3, 3-3.000, Evaluation and Review Program; Executive Office of the United States Attorney's Resource Manual, Title 3, Section 6; E-mail message from Barbara Cottrell, Assistant U.S. Attorney, Northern District of New York (June 7, 2002).

104. NAPA, HOW EPA AND THE STATES CAN IMPROVE, supra note 45, at 22.

105. Id.

106. These stemmed from a variety of causes, including design flaws in the databases, difficulties using the databases, problems in interpretation of EPA guidance for using the data system, and time lags from when the data is submitted. Id.

107. OIG, U.S. EPA, INFORMATION TECHNOLOGY: UNRELIABLE DATA AFFECTS USABILITY OF DOCKET INFORMATION (Executive Summary, 2002) (2002-P-00004).

108. The report estimated that EPA lacks data on an estimated 96% of stormwater discharges, 65% of discharges from concentrated animal feeding operations, and thousands of minor dischargers. U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 19-21. Although many states were developing their own systems, they did not fill the information void. Id. at ii.

109. NATURAL RESOURCES DEFENSE COUNCIL (NRDC), COMMENTS ON ECOS REPORT TO CONGRESS ON STATE ENFORCEMENT AND RECOMMENDATIONS FOR STRENGTHENING STATE ENFORCEMENT PROGRAMS (2001) (copy on file with author).

110. U.S. GAO, ENVIRONMENTAL PROTECTION: MORE CONSISTENCY NEEDED AMONG EPA REGIONS IN APPROACH TO ENFORCEMENT 47 (2000) (GAO/RCED-00-108).

111. See U.S. EPA, Office of Environmental Information, at http://www.epa.gov/oei (last visited June 4, 2002) (giving an overview of the Office of Environmental Information, and listing as one function "providing leadership for improving quality and utility of EPA's data and information …"). Advances in electronic reporting are likely to help as well.

112. BROWN & GREEN, supra note 36, at 71.

113. H.R. CONF. REP. NO. 107-272 (2001); H.R. CONF. REP. NO. 108-110 (2003).

114. NAPA, HOW EPA AND THE STATES CAN IMPROVE, supra note 45, at 19 (emphasis in original).

115. NAPA, SETTING PRIORITIES, GETTING RESULTS: A NEW DIRECTION FOR EPA 77, 81 (1995) [hereinafter NAPA, SETTING PRIORITIES].

116. U.S. GAO, ENVIRONMENTAL PROTECTION: COLLABORATIVE EPA-STATE EFFORT NEEDED TO IMPROVE NEW PERFORMANCE PARTNERSHIP SYSTEM (1999) (GAO/RCED 99-171) [hereinafter U.S. GAO, COLLABORATIVE EPA-STATE EFFORT].

117. ENVIRONMENT.GOV, supra note 8, at 159.

118. Id. at 161.

119. OVERSIGHT OF AUTHORIZED STATE ENVIRONMENTAL PROGRAMS, supra note 84, at 28-29.

120. In this regard, EPA could borrow from evolving approaches to permit renewal. For the past several years, New York, among other states, has prioritized its permit renewal work in an effort to allocate resources where they are most needed. Thus, permit renewal applicants whose discharges are "okay" from a legal and environmental perspective receive relatively little scrutiny at the renewal stage. In contrast, dischargers for whom significant concerns exist receive far more intense review as part of the permit renewal process. See, e.g., David L. Markell, States as Innovators: It's Time for a New Look to Our "Laboratories of Democracy" in the Effort to Improve Our Approach to Environmental Regulation, 58 ALB. L. REV. 347, 377 (1994).

121. See 33 U.S.C. § 1342(d), ELR STAT. FWPCA § 402(d).

122. PADDOCK & KEINER, supra note 7, at 50.

123. Related, critics have faulted EPA's decision to allow states to negotiate a PPG without having a PPA in place, arguing that this allows states "to make decisions about reallocating federal grants without directly connecting them to the strategic planning, priority-setting, and self-evaluations required in a PPA, yet those elements are key to ensuring improvement of environmental results through performance-based management under NEPPS." PADDOCK & KEINER, supra note 7, at 10. To ensure greater accountability of state performance, EPA should likewise consider limiting PPGs to only those states with PPAs.

124. For a discussion of "sunshine" approaches used throughout the world, see EDITH BROWN WEISS & HAROLD K. JACOBSON, ENGAGING COUNTRIES: STRENGTHENING COMPLIANCE WITH INTERNATIONAL ACCORDS 542-46 (1998).

125. JOHN M. MENDELOFF, THE DILEMMA OF TOXIC SUBSTANCES REGULATION 209-10 (1988).

126. Cass R. Sunstein, Informational Regulation and Informational Standing: Akins and Beyond, 147 U. PA. L. REV. 613, 616 (1999) (emphasis in original).

127. OECA, U.S. EPA, SECTOR FACILITY INDEXING PROJECT EVALUATION (1999), available at http:www.epa.gov/sfipmtn1 [hereinafter OECA, SECTOR FACILITY INDEXING PROJECT].

128. U.S. EPA, 2000 Toxic Releases Inventory Data, at http://www.epa.gov/tri/tridata/tri00/index.htm.

129. Bradley C. Karkkainen, Information as Environmental Regulation: TRI and Performance Benchmarking, Precursor to a New Paradigm?, 89 GEO. L.J. 257 (2001).

130. Clifford Rechtschaffen, The Warning Game: Evaluating Warnings Under California's Proposition 65, 23 ECOLOGY L.Q. 303, 341-43 (1996); Clifford Rechtschaffen, How to Reduce Lead Exposures With One Simple Statute: The Experience of Proposition 65, 29 ELR 10581 (Dec. 1999).

131. David Roe, Toxic Chemical Control Policy: Three Unabsorbed Facts, 32 ELR 10232 (Feb. 2002).

132. OECA, SECTOR FACILITY INDEXING PROJECT, supra note 127.

133. Id.

134. See Competing Visions, supra note 65, at 10819-20.

135. Tom Tietenberg & David Wheeler, Empowering the Community: Information Strategies for Pollution Control (Frontiers of Environmental Economics Conference, Oct. 23-25, 1998), at 25 [hereinafter Tietenberg & Wheeler].

136. METZENBAUM, MAKING MEASUREMENT WORK, supra note 70, at 42.

137. U.S. EPA, STATE ENFORCEMENT OF CLEAN WATER ACT, supra note 8, at 55.

138. REINVENTING EPA NEW ENGLAND 26-29 (NAPA 2000), available at http://www.napawash.org.

139. U.S. GAO, COLLABORATIVE EPA-STATE EFFORT, supra note 116, at 44.

140. See David L. Markell, The Citizen Spotlight Process, ENVTL. F., Mar./Apr. 2001, at 32.

141. See David L. Markell, The Commission for Environmental Cooperation's Citizen Submission Process, 12 GEO. INT'L ENVTL. L. REV. 545, 571 n.128 (2000) (quoting FOUR-YEAR REVIEW OF THE NORTH AMERICAN AGREEMENT ON ENVIRONMENTAL COOPERATION: REPORT OF THE INDEPENDENT REVIEW COMMITTEE 5 (1998)).

142. ECOS, RESOLUTION NUMBER 97-5: SECTOR FACILITY INDEXING 1 (1997).

143. See, e.g., OECA, U.S. EPA, ENFORCEMENT AND COMPLIANCE ASSURANCE ACCOMPLISHMENTS REPORT FY 1996, at 3-20 (1997) (EPA-300-R-97-003).

144. See Tietenberg & Wheeler, supra note 135, at 18-19.

145. 40 C.F.R. § 271.22(a). See also 40 C.F.R. § 70.10(c) (withdrawal regulations for CAA Title V permitting program); id. §§ 123.63, 123.64 (withdrawal regulations for CWA NPDES permitting program).

146. For example, in National Wildlife Fed'n v. EPA, 980 F.2d 765, 23 ELR 20440 (D.C. Cir. 1992), the court found that EPA enjoys broad discretion in determining that a state no longer satisfies the primacy criteria of the SDWA. (The court additionally found that once EPA made such a determination, it was required to initiate withdrawal proceedings.) For a decision that took a narrower view of the scope of EPA discretion, see Save the Valley v. EPA, 99 F. Supp. 2d 981 (S.D. Ind. 2000).

147. 42 U.S.C. § 6926(e), ELR STAT. RCRA § 3006(e); 40 C.F.R. § 271.23.

148. See 33 U.S.C. § 1342(c)(3), ELR STAT. FWPCA § 402(c)(3); 40 C.F.R. § 123.64.

149. RODGERS, HORNBOOK SERIES, supra note 12, at 367-68. For further information on the Ohio review, see http://www.epa.gov/region5/ohioreview (last visited June 1, 2003).

150. U.S. GAO, EPA AND THE STATES — ENVIRONMENTAL CHALLENGES REQUIRE A BETTER WORKING RELATIONSHIP 25 (1995) (GAO/RCED 95-64) [hereinafter U.S. GAO, EPA AND THE STATES].

151. Hazardous Waste Treatment Council v. Reilly, 938 F.2d 1390, 21 ELR 21228 (D.C. Cir. 1991).

152. U.S. GAO, EPA AND THE STATES, supra note 150, at 18.

153. See John H. Cushman Jr., States Neglect Pollution Rules, White House Says, N.Y. TIMES, Dec. 15, 1996, at A1.

154. ENVIRONMENT.GOV, supra note 8, at 156. In another recent case, EPA Region 6 threatened to revoke Louisiana's water pollution permitting program because of deficiencies such as permit backlogs, delays in completing enforcement actions, undue reliance on supplemental environmental projects, and failure to recover the economic benefit of violations. See Mark Schleifstein, EPA Puts Pressure on State DEQ, TIMES-PICAYUNE (New Orleans), Feb. 15, 2003, at 1.

155. Rena I. Steinzor & William F. Piermattei, Reinventing Environmental Regulation Via the Government Performance and Results Act: Where's the Money?, 28 ELR 10563, 10573 n.105 (Oct. 1998).

156. Hodas, supra note 2, at 1586 (quoting EPA official).

157. ELI, LAW OF ENVIRONMENTAL PROTECTION, supra note 14, § 6.02, at 6-17. The GAO and others have reached similar conclusions. U.S. GAO, EPA AND THE STATES, supra note 150, at 25-26. Prof. Rena I. Steinzor concludes that "EPA's own funding shortfalls make it difficult for the Agency to threaten to withdraw state delegations with any credibility." Rena I. Steinzor, Reinventing Environmental Regulation Through the Government Performance and Results Act: Are the States Ready for Devolution?, 29 ELR 10074, 10079 (Feb. 1999). (We have seen very little data about the comparative costs of EPA versus the states implementing an enforcement program. One of the few available comparisons resulted when EPA assumed control of Idaho's air quality program for 15 months in 1981, because of funding cutbacks by the Idaho Legislature. One EPA official estimated that it cost the federal government twice as much to run the program as the state. This is discussed in Kuehn, supra note 4, at 2390-91.)

158. NAPA, SETTING PRIORITIES, supra note 115, at 87-88.

159. This decision was upheld in Hazardous Waste Treatment Council v. Reilly, 938 F.2d 1390, 21 ELR 21228 (D.C. Cir. 1991).

160. RODGERS, HORNBOOK SERIES, supra note 12, at 643.

161. SCHEBERLE, supra note 23, at 7.

162. METZENBAUM, MAKING MEASUREMENT WORK, supra note 70, at 75.

163. 42 U.S.C. § 7413(a)(2), ELR STAT. CAA § 113(a)(2); 33 U.S.C. § 1319(a)(2), ELR STAT. FWPCA § 309(a)(2).

164. See Hodas, supra note 2, at 1582-83 (EPA lacks resources to assume enforcement of the CWA in states). Professor Rodgers, in typically colorful fashion, dismisses an enforcement takeover as politically infeasible: "This version of environmental bankruptcy is a nice expression of the strong moralism found in the [CAA], but it has about as much chance of being applied as a provision enabling the Administrator to seek imprisonment of the Governor." WILLIAM H. RODGERS JR., I ENVIRONMENTAL LAW: AIR AND WATER §§ 3.38, 4.40 (1986). He also notes that it would be very hard to take control over solely the enforcement function of an agency without also assuming responsibility for the other interrelated program activities.

165. See, e.g., 40 C.F.R. § 123.64(b).

166. U.S. EPA, FY 2001 ANNUAL REPORT, supra note 60, at II-87 to II-88.

167. See, e.g., Office of Congressional and Intergovernmental Relations, U.S. EPA, Performance Partnership System: Frequently Asked Questions, at http://www.epa.gov/ocir/nepps/faq.htm; U.S. GAO, COLLABORATIVE EPA-STATE EFFORT, supra note 116, at 51.

168. See U.S. GAO, COLLABORATIVE EPA-STATE EFFORT, supra note 116, at 51.

169. See Office of Congressional and Intergovernmental Relations, U.S. EPA, Joint Policy Statement on State/EPA Relations (last modified June 9, 1998), at http://www.epa.gov/ocir/nepps/jps.htm (signed by EPA Administrator Browner and six state officials).

170. See, e.g., Hodas, supra note 2, at 1587. There are signs that EPA is seeking to enhance its capacity for targeted enforcement response. On March 4, 2003, EPA's Assistant Administrator for the OECA announced the creation of a "Special Litigation and Projects Division" in the OECA's Office of Regulatory Enforcement. The purpose of the division is to enhance EPA's ability to provide "targeted and rapid response" to environmental enforcement needs in "any geographical sector or media." The text of the EPA memorandum can be found at Daily Env't Rep. (BNA), Mar. 26, 2003, at E-1.

171. One example of the last type of case is EPA's round of suits filed against older power plants for allegedly rebuilding their facilities under the guise of carrying out routine maintenance and repair activities, thereby evading new source review under the CAA. These suits are described in James A. Lofton, Environmental Enforcement: The Impact of Cultural Values and Attitudes on Social Regulation, 31 ELR 10906, 10912-13 (Aug. 2001) and in Kevin A. Gaynor & Benjamin S. Lippard, Environmental Enforcement: Industry Should Not Be Complacent, 32 ELR 10488 (Apr. 2002).

172. The Relationship Between Federal and State Governments in the Enforcement of Environmental Laws: Hearings Before the Senate Comm. on Env't and Pub. Works, 105th Cong. 190 (1997) (prepared statement of Mark Coleman, Executive Director, Oklahoma Department of Environmental Quality, and Chairman, Compliance Committee, ECOS) [hereinafter Coleman statement].

173. Kuehn, supra note 4, at 2373.

174. The Relationship Between Federal and State Governments in the Enforcement of Environmental Laws: Hearings Before the Senate Comm. on Env't and Pub. Works, 105th Cong. 7 (1997) (statement of Steven A. Herman, Assistant Administrator, OECA, U.S. EPA) [hereinafter Herman Statement]. EPA regulations under the CWA, RCRA, and the CAA contemplate that EPA may overfile under certain circumstances. See 40 C.F.R. § 123.27 (federal authority to seek penalties under state water programs); id. § 271.16 (under state hazardous waste programs); id. § 63.90 (under state air programs). The U.S. Court of Appeals for the Eighth Circuit has concluded that EPA lacks the authority to overfile under RCRA, while the U.S. Court of Appeals for the Tenth Circuit has disagreed. Compare Harmon Indus., Inc. v. Browner, 191 F.3d 894, 29 ELR 21412 (8th Cir. 1999), with United States v. Power Eng'g Co., 303 F.3d 1232, 33 ELR 20027 (10th Cir. 2002). For comprehensive discussions of overfiling, see Elizabeth A. Clysdale, A Look at EPA Overfiling: Can Harmon and Power Engineering Exist in Harmony?, 33 ELR 10456 (June 2003); Jerry Organ, Environmental Federalism Part I: The History of Overfiling Under RCRA, the CWA, and the CAA Prior to Harmon, Smithfield, and CLEAN, 30 ELR 10615 (Aug. 2000); Jerry Organ, Environmental Federalism Part II: The Impact of Harmon, Smithfield, and CLEAN on Overfiling Under RCRA, the CWA, and the CAA, 30 ELR 10732 (Sept. 2000).

175. Ellen R. Zahren, Overfiling Under Federalism: Federal Nipping at State Heels to Protect the Environment, 9 EMORY L.J. 373, 436 (2000).

176. See Competing Visions, supra note 65, at 10807-09.

177. Herman Statement, supra note 174.

178. E-mail message from Gary Jonesi, Office of Regulatory Enforcement, U.S. EPA (Dec. 17, 2002) (listing overfilings initiated by EPA from Jan. 20, 2001 to Dec. 17, 2002) (copy on file with authors).

179. Coleman Statement, supra note 172.

180. Brent Israelsen, Representative Stirs Up a Western Turf War With EPA, SALT LAKE CITY TRIB., Jan. 30, 1999, at A4. Montana, South Dakota, and Wyoming also apparently considered introducing such measures. Id.

181. Specifically, the resolution would narrow the universe of situations in which overfiling could occur to situations where: (1) the state has abused its discretion in carrying out its enforcement authority by failing to take the necessary steps to bring the alleged violator into compliance; (2) the alleged violation involves a significant interstate pollution impact; (3) serious and irreparable harm will occur to the public health and/or environment unless immediate action is taken; or (4) the state requests enforcement assistance from EPA. See E. Blaine Rawson, Overfiling and Audit Privileges Strain EPA-State Relations, NAT. RESOURCES & ENV'T, Winter 1999, at 483-85.

182. Hubert H. Humphrey III & LeRoy C. Paddock, The Federal and State Roles in Environmental Enforcement: A Proposal for a More Effective and More Efficient Relationship, 14 HARV. ENVTL. L. REV. 7, 15 (1990).

183. 33 U.S.C. § 1319(a)(1), ELR STAT. FWPCA § 309(a)(1); 42 U.S.C. § 7413(a)(1), ELR STAT. CAA § 113(a)(1); id. § 6928(a)(2), ELR STAT. RCRA § 3008(a)(2).

184. Stephen C. Robertson, State Permitting: United States v. Smithfield Foods, Inc. and Federal Overfiling Under the Clean Water Act, 23 WM. & MARY ENVTL. L. & POL'Y REV. 593, 594-95, 607 (1999).

185. OIG, CONSOLIDATED REVIEW, supra note 10, at 9.

186. Prof. Janet A. Weiss provides a very helpful summary of different types of information dissemination as policy tools. As she notes, such tools are "marvelously versatile." Among other things, "in communicating what is and what may be, these policies may send strong signals about what should be." Professor Weiss points out that "report cards," an increasingly used form of information intervention that we believe holds great promise, have two purposes, "to discourage undesirable behavior by service providers and to inform, empower, and encourage citizens to exert pressure at the street level for better performance." That is, they "have the potential for two kinds of effects: those on the organizations that create the information and those on the audiences who receive it." Janet A. Weiss, Public Information, in THE TOOLS OF GOVERNMENT: A GUIDE TO THE NEW GOVERNANCE 217, 224, 238 (Lester M. Salamon ed., 2002).

33 ELR 10559 | Environmental Law Reporter | copyright © 2003 | All rights reserved