Good Faith as a Fundamental Principle for Relational Environmental Governance

Rick Reibstein

The author, Assistant Director of the Massachusetts Office of Technical Assistance for Toxics Use Reduction since 1990, is on detail at the U.S. Environmental Protection Agency, Region 1, New England, where he divides his time between assistance and enforcement. He is a graduate of Hampshire College and Brooklyn Law School, and has taught environmental law and policy at Boston University and Northeastern University's Continuing Education program. His opinions are his own and do not constitute official statements of the institutions with which he is associated.

[32 ELR 10219]

In the late 1990s, the Massachusetts Office of Technical Assistance for Toxics Use Reduction (OTA), a nonenforcement agency of the commonwealth of Massachusetts dedicated to the on-site provision of pollution prevention assistance, developed a sector-based program intended to demonstrate a fast path to a common-sense environmental regulatory system. The program, funded by the U.S. Environmental Protection Agency (EPA) as the Massachusetts Auto Body Project, became known as the Collision Repair Auto Shop Help (CRASH) Course, which was the title given to the training materials and documents produced.¹ The CRASH Course demonstrated a new way of simplifying the many complex rules faced by a small business sector that, like many other small business sectors, had never received very much enforcement attention.² The method of presenting requirements also created a "positive" enforcement context for pollution prevention and other best management practices (BMPs). The chief officials of the relevant environmental enforcement agencies of the state signed a joint letter, distributed with the project's materials, to indicate their united agreement on the approach.³

The Massachusetts Auto Body Project used the concept of "good faith" to produce a simplified version of the rules. The good-faith approach allowed for the inclusion of pollution prevention and other BMPs in a way that made them more than just suggestions. No rule promulgation was performed, yet the simplified version of the rules applies the force of law. This Dialogue recommends broadening the application of good faith to more effectively cover activities of regulated entities before they come in contact with agencies, and sending a clear message to all actors, not just high performers, that the Agency will distinguish between those making an effort to comply and/or reduce their impact, and those who do not evidence responsibility. The Dialogue examines the Massachusetts Auto Body Project and how it demonstrates a methodology for building a regulatory system that makes practical or common sense, and why the concept of good faith is important to the future evolution of our system of environmental governance. The first part of the Dialogue explains the specific use of good faith in the project and other contexts. The Dialogue then applies the concept of good faith to the larger issue of the relationship between government and regulated citizenry.

Using the "Good-Faith" Concept for Common-Sense Regulation

"Common sense," as defined by the Common Sense Initiative launched by EPA in 1994, means doing the job of the Agency in a way that is cleaner, cheaper, and smarter. It is cleaner to have a regulatory system that promotes pollution prevention and other BMPs, and it is smarter as well. It is [32 ELR 10220] also cheaper for the regulated party, and if it succeeds in reducing risks and, thus, the need for enforcement or its scale, it is cheaper for the Agency and the public as well. In addition, it is cheaper and smarter to have a system that is well understood, and it may be assumed that a clear system with fewer points of confusion or frustration is one that fosters a higher potential for compliance. If the system is practical and accords with people's sense of logic, they should be more likely to observe it.⁴

The Massachusetts Auto Body Project used the concept of good faith as the principle for organizing a simplified version of the many rules that might apply to an auto body facility. Using the good-faith approach also allowed the incorporation of pollution prevention and other BMPs into the CRASH Course regulatory guidance that was issued.

"Good faith" is a broad term that can mean many different things. Under the Massachusetts Auto Body Project, it refers to the term as it is used in penalty setting policies of the enforcement agencies involved in the project: the Massachusetts Department of Environmental Protection (DEP), the Massachusetts Attorney General's Office, and EPA. All three agencies and the laws that they enforce take good faith into account in the setting of penalties when facilities are found in violation. For example, § 3008(a)(3) of the Resource Conservation and Recovery Act states that in assessing penalties, EPA must take into account any good-faith efforts to comply with the applicable requirements.⁵ The penalty setting process also takes into account a violator's bad faith as well as their degree of willfulness or negligence. Evidence of good faith could conceivably counter a finding of either of these.

Other factors are relevant to good faith: the DEP's penalty policy explicitly includes considering the public interest, and EPA considers the opportunities that exist to establish a useful precedent or send a signal to the regulated community.⁶ Though this may be interpreted as sending a signal about the likelihood of punishment to discourage bad behavior, it can just as well be used to send a signal about leniency to encourage good behavior.

Combining Commands and Suggestions

The idea of the Massachusetts Auto Body Project was simple. The CRASH Course summarized the most important compliance issues, and woven into this guidance was information about pollution prevention and other good management practices, activities not specifically required but that would aid in compliance and reduce health, safety, and environmental risks.

As with most pollution prevention and other BMP activities, these activities were likely to save companies money in the short or long run. Two examples of a pollution prevention activity likely to result in savings in the short run are measuring out spray paint precisely to reduce waste, and using a dedicated gun cleaner to reduce volatile organic compound emissions. These both would reduce purchases of raw materials and pay for themselves in a short time.

Two examples of other BMP activity likely to result in savings in the long run are treating effluents from car washing and controlling air emissions from spray painting. Treating effluents from car washing is an activity that shops consider onerous and of minimal benefit, but there are relatively cheap ways to accomplish it, and potential liabilities from uncontrolled discharge can be quite large, particularly from the washing of undercarriages. Although the typical auto body shop does not expect to be forced to pay for groundwater, surface water, or soil contamination from such discharges, it can happen, and fines can also be levied. Similarly, using adequate filtering to control spray painting emissions is a recurring cost, but if neighbors complain to the local board of health about odors or exposure, the board can use its nuisance powers to close the shop down. (Although it may be necessary in many cases to go beyond regulatory requirements to address nuisance complaints, complying with them as a first step will certainly help.) Avoiding shutdown is a good investment, but of the sort that is not obvious except in retrospect.

Thus, the facility was given a summary of what is required and what is desirable from both the agencies' and the facility's points of view. To ensure that the facility's point of view was taken into account, the OTA developed the document with considerable input and assistance from the Massachusetts Auto Body Association and several individual facility operators.

The Basic Message: Faring Better With Inspectors

The best way to understand the use of the existing good-faith policies and how they apply to the simplified guides to the law, pollution prevention, and BMPs is to paraphrase how the materials were described to the auto body shop personnel attending the many workshops held across the state in 1998 and 1999. What they were told was that if they read the guidance, performed the activities described within it, and documented that they had done so, they would fare better during an inspection than if they had not done so. To assist in the documentation effort, a checklist was provided. It was suggested that they copy the checklist and use it to go through their shop at least four times a year.

Shop owners and staff heard directly from the officials responsible for fining them. For example, the DEP official responsible for Clean Water Act (CWA) compliance told them that if he visited a shop and found that they were discharging oil-laden washwater, and they were making no effort to control it, he might seek penalties of several thousand dollars per day of violation. If, however, he found that the shop was observing the recommended practices in the CRASH Course materials, he would not seek the maximum penalties. In fact, he might seek no penalties at all depending on what he saw, the effect of the discharges, and the response of the facility.

During the planning stage of the project, some had questioned whether the regulated population would understand the complicated idea of good faith. But there was no lack of clarity or need for sophistication in order to receive the message [32 ELR 10221] as intended. The response from the attendees was very consistent. They appreciated the idea that the agencies would distinguish between those trying to do the right thing and those ignoring their responsibilities.

Regulatory Incorporation of BMPs

In order to fully understand the virtue of this use of the good-faith concept, it is necessary to review some of the difficulties that have been encountered in attempts to develop regulatory guidance or systems that incorporate pollution prevention or other BMPs and in attempts to issue simplified regulatory regimes.

Requiring pollution prevention or other BMPs by regulation risks being prescriptive, inflexible, and intrusive. For example, if it is clear that most processes involving degreasing can be made safer by a switch to aqueous cleaners, should we not require it? Unfortunately, it is not that simple. Whether or not a particular pollution prevention option is a good business or practical idea depends on many factors individual to the plant, process, or product. Is there a suitable cleaner for the particular soil that needs to be cleaned off? Can the range of available cleaners do the job effectively and to the level of cleanliness required by the customer? Are the cleaners reliably obtainable at the right price and supply? Is the cleaner compatible with the materials to be cleaned (for example, will it cause rusting)? Does the use of a water-based cleaner require new equipment? Is there space for the new equipment? And does the use of the new cleaner change the production time, add steps, require training, or introduce new health or safety risks? Before an agency requires companies to make specific changes, they need to assess the affect on the company.⁷

Most importantly from an enforcement attorney's perspective, would such an order provide the company with a case that they performed actions at the behest of the agency, and will that then be a "shield" protecting them from agency action? Will it expose the agency itself to litigation if anyone is harmed by the company's actions?

If the agency did not consider the potential adverse impact of orders that specifically command a company to change materials or operations, it could cause undue hardship or the loss of money, time, or customers, which could create unwanted friction and resistance. Agency staff have expertise in environmental matters, and although many in pollution prevention assistance, research, rule development, and other departments have learned a great deal about how businesses operate and what materials and processes they use, they can never have complete assurance that they can prescribe activities without risk. For this reason, pollution prevention assistance offices typically present businesses with a range of options, and many inspectors are careful to couch advice in terms of suggestions, not commands.⁸

A more practical approach is for a regulatory agency to create incentives for the switch and/or disincentives for those who do not address the risks of continuing to use more dangerous chemicals. A standard approach has been to require or encourage the consideration of chemical alternatives rather than to require specific substitutions.

In general, businesses do not appreciate being told what materials, processes, or equipment to use, and unless there is a compelling need justifiable on health, environmental, or safety grounds, environmental policy should balance the needs of protection with the freedoms of economic enterprise to which we have become accustomed.

An example of how pollution prevention and other BMPs can be required is the current approach to regulating stormwater runoff from facilities covered by the stormwater permit requirements. Covered facilities must have a plan that applies BMPs, which are generally described. For example, if you are in the water transportation sector, you are covered by Sector Q, and if pressure washing is used to remove marine growth from vessels, your pollution prevention plan must describe the measures you use to collect or contain the discharges from the pressure washing area.⁹

EPA's 1995 Fact Sheet for Sector Q describes "potential" BMPs for each activity for which pollution prevention planning is required.¹⁰ For pressure washing boats, in addition to [32 ELR 10222] collecting the discharge water and removing all visible solids, BMPS include using no detergents or additives and using solid decking, gutters, and sumps. The idea is that each facility will choose the BMPs that are suitable for them. The Agency did not ban or limit the use of detergents or additives, or require the construction of sumps or decking.¹¹ Presumably, this would have required a higher standard, high enough to withstand the protests that may be expected from detergent manufacturers or from marinas where space for constructing sumps or decking is limited. The rule is flexible, and one who does not implement the particular BMP may be able to argue that it is not appropriate for the location or application.

Similarly, state pollution prevention planning laws have taken a flexible approach. Companies are not made to do specific pollution prevention, but are required to complete plans with specific elements such as options identification, materials tracking, and full-cost assessment. These planning elements are designed to lead the company to make its own appropriate choices.¹²

Another approach is to require pollution prevention planning or other specific or general beneficial environmental activities in settlements. At the federal level this has been formalized as a supplemental environmental project (SEP). EPA's SEP program "encourages the use of environmentally beneficial projects as part of the settlement of an enforcement action," and by this means achieves "environmental and public health protections beyond that specifically required by law."¹³ Many states as well have used the settlement process to promote good practices. Specific activities are generally left to the company to propose.¹⁴ In all of these examples, the enforcement driver is general, and the particular pollution prevention or BMP activity to be employed at individual facilities are not specified in advance.

In the stormwater example, the driving force is the fact that the state or the federal agency can issue specific orders or fines if a water body is found to be contaminated by a facility's activities, and many states have actual bans on nonpoint source pollution on the books to back this up. The idea of an inspector assessing whether a facility actually has a pollution prevention plan (stormwater or other) is also a driving force, although penalties for inadequate planning require a level of expertise in inspector training that demands resources to be developed.¹⁵

The Spill Prevention, Control, and Countermeasure Plan requirement that applies to oil storage is an example of a pollution prevention plan that is easy to evaluate, as it is quite prescriptive and specific, and it requires the sign-off of a professional engineer, considerably reducing the burden on the inspector.¹⁶ Although the Massachusetts Toxics Use Reduction Act requires a "good-faith" effort to plan, the typical driving force in state pollution prevention law is the public scrutiny afforded by reporting requirements. If a company has to report its progress toward being a less environmentally dangerous facility, it has some motivation to take its planning seriously.¹⁷

[32 ELR 10223]

Positive Enforcement Motivation

The good-faith approach provides an enforcement motivation for companies to do the activities that are specified in guidance as opposed to rule. It is a way to accomplish "regulatory incorporation" of pollution prevention and BMPs without going through rulemaking. Because inclusion in a rule brings negative incentives into play, the research necessary to ensure that no one will be unfairly burdened is time-consuming and expensive. In other words, to promulgate the rule one has to show that it would be fair and appropriate to fine people for not performing the activity. One has to meet specific U.S. Office of Management and Budget review criteria and address all the comments submitted by the public before issuing the rule. This entails meeting a rather high standard, and many activities that would be great to do but that do not rise to the level of needing to be required, or are not good for everyone, are not written into rules.¹⁸

Using the positive context of enforcement discretion to afford penalty mitigations for good-faith actions allows their inclusion in regulatory guidance, which is nearly as good as rules, and perhaps better if more widely understood and followed. It does nothing to raise the performance of the recalcitrant, but this is not a shortcoming if the basic deterrence mission of the agency remains strong.¹⁹

This method is a positive tool because although the company is not penalized for failing to implement the pollution prevention/BMP activities described in the CRASH Course guidance, it may receive credit for doing them. If a facility is visited by an inspector who finds that it has left drums unlabeled and proceeds to calculate penalties for these violations, the agency, when calculating the penalty, would take into consideration any good-faith efforts the facility has made in other areas described in the guidance. The facility, therefore, would fare better than if it had not implemented the activities. This could result in a difference of several thousand or tens of thousands of dollars. Thus the positive incentive can be significant.²⁰

Another virtue of the positive tool is that it allows for the inclusion of good-faith activities that would be appropriate for only some facilities. These can be included without harming those for whom they are inappropriate, because there is no penalty for not doing them. One of the basic problems of the conventional command-and-control system—the Procrustean bed of "one size fits all"—is not an issue when using the good-faith approach.

Describing what is good faith by providing specifics in guidance accomplishes a number of objectives. First, complementing the negative approach of penalties with a positive approach of providing for credits changes the relationship of agency and regulated entity. If the agency has tools with which to promote and encourage good behavior, it is in less danger of being seen by those whom it inspects as a perpetual adversary. Its basic relational stance is more sensible, and the chances for willing cooperation with its mission are enhanced. Rules are properly interpreted as existing for the protection of the public, and not as traps for the unwary. Agency staff are more likely to be seen—by those it may penalize—as agents acting on behalf of them and not as enemies of business.

Second, it provides for a flexible method of specifically describing desirable, but not required, activities, which allows the agency to strongly promote "going beyond compliance." A long-standing complaint about the conventional command-and-control regulatory system is that it prescribes a floor, a minimum. Meet the minimum and you've done enough. The minimum is set to achieve a balance between the imposition of burdens and the restriction of freedoms on the one hand, and the necessary protections on the other. This balance is weighed for the entire regulated population. There is a great deal that potential pollution sources can do to improve their environmental performance that is [32 ELR 10224] not included in the rules. For example, the CWA categorical standards, which set limits on discharges from industry sectors, are developed after exhaustive research on what technologies those sectors can economically and technologically employ. If these standards, as proposed, were set so that only a few very wealthy companies could meet them, they might not make it to the promulgation stage because all the companies who would not be able to meet them would protest. What this means is that some companies who could do much more are not required to do so.

More importantly, perhaps, is the fact that technology and our problem-solving abilities constantly improve. We now know how to radically reduce many industrial wastewater discharges by switching input materials upstream. Yet our standards have been set by considering the process as a given and solely by examination of control technology. For example, technology to control discharges of silver from photographic waste is widely available, and we can expect that a limit of two parts per million or so will be achievable by any discharger, and we don't have to trouble ourselves with the concern that small photodevelopers will be put out of business. However, we also know that a larger company could spend a few thousand dollars for a vacuum distillation unit and discharge at the parts per billion level. Or, that they could invest in a digital pre-press system and drastically reduce or eliminate the use of silver-bearing chemicals.

Creating a rule to require this could be a long, drawn-out process. It would entail either forcing the smaller companies to meet the stricter limits or carving out a way to separate them from the wealthier companies, which would involve establishing justifications for doing so. Even after all of this work one would be left with crudely drawn categories, with much variation of capability within them. Such a cumbersome process cannot keep up with rapidly developing technology and improvements in practice.

Publishing guidance that establishes a positive incentive, by defining these technologies as good faith, is comparatively simple. As technologies advance, the guidance can easily be revamped to include new information and new ideas. Each version can be dated with a warning that all representations of good-faith penalty mitigation are "as of" the date of the document or for a certain period.

In this way establishing positive enforcement drivers redirects our attention from the "floor," the minimum, toward the "ceiling," the practical limit of what is achievable. That our attention should be here as well as on the minimum has been widely recognized. One prominent example of this recognition is the establishment of EPA's Performance Track Program, which provides many positive incentives for environmental performance. Many states have established similar programs. These programs have included enforcement-related incentives such as lower inspection priority. But provision of these incentives has been within carefully prescribed boundaries: a company has to apply to the program and meet a series of qualifications.²¹

Spelling out how to obtain good-faith credits provides a flexible enforcement motivator for meeting and going beyond compliance, which can be applied to all companies, not just those that have qualified as performers. Using the positive tool to advance the state of the art can be a more widely applied supplement to our traditional use of the negative tool for ensuring a minimum standard. This way, as with the SEP and performance track programs, one of the other basic features of the conventional command-and-control system—the general inability of the system to promote going beyond the minimum—can be addressed.

Simpler Simplification

Finally, the use of the good-faith approach softens one of the age-old problems inherent in the issuance of regulatory guidance: the fact that simplification leaves things out. Developing plain language, user-friendly guides to regulations that are of any use means deciding what are the most important elements of applicable rules. If, for example, you publish a guide to the Clean Air Act and you neglect to mention thresholds for permits, you are not doing your readers any favors. They may follow your guide and be completely unprotected from inspections that turn up the lack of necessary permits. Including the necessary points is a basic task when developing regulatory guidance. Agencies issue regulatory guidance all of the time, but with a disclaimer that informs the reader that the guidance does not have the force of law, and that thereader still has an obligation to read the actual rules, as they all still apply. Readers may be tempted to ask, does following this guide offer any protection?

The issuance of a good-faith guidance must observe this very same inclusionary principle, but the result is greater than a summary plus disclaimer. Although the reader is still on notice that all the rules in obscure state and federal codes (obscure to many businesses too small to pay professionals to examine them) apply, at least the reader knows that they will be better off if they do what the guidance recommends, even if they neglect to obtain the regulations and study them in the original. Although this is arguably implicit in other guidance, the effect of the disclaimer and the lack of positive incentives tends to interfere with the signal: the end result is a public that is not sure of where they stand.

If the agency seeks to actually simplify the rules it then faces the problem of having to decide which rules it may discard.²² Each rule is there for a reason, which has to be rediscovered and reexamined, and the political risks as well as the environmental, health, and safety risks must be thoroughly understood. For example, EPA has recently extended, for some generators, the rule that requires shipment [32 ELR 10225] of hazardous wastes within 90 days.²³ This took many years, and doing so has created pressure from others who want the extension as well. Every rule change that seems to make our system less protective requires careful study not just of direct effects, but also for precedential effects.

Producing a good-faith-based simplification can be a useful step on the way to a streamlined regulatory system. A version of a guidance established with good faith can be a trial version, a precedent to the actual rewriting of the rules. If it is poorly formulated or there are adverse effects, the guidance can be quickly revised and reissued.

Using the good-faith principle, an agency can orient the public to its priorities without discarding any rules prematurely. If a rule is unimportant and burdensome, it can be left out of the summary of important rules, the observance of which demonstrates good faith. Agencies always use their enforcement planning and discretion for this purpose. They choose to focus on certain activities and not on others. The good-faith guidance is a tool for explicitly communicating these priorities to the public, eliminating the imbalance between most companies and those relative few that have expertise on retainer or staff able to decipher the agency's priorities. A public that knows what is important to the agency can spend less time plowing through complex regulations and more time getting done what needs to get done.

Precontact and Postcontact Good Faith

During the early development of the Massachusetts Auto Body Project, after the idea of the good-faith approach had first been articulated and proposed, it was hard to understand the initial resistance that the OTA encountered from the enforcement agencies. This resistance was overcome after more than a year of meetings and discussions, but one of the difficulties was pinpointing exactly what the problems were. One potential problem that was identified was the idea that promising to mitigate penalties might tie the hands of enforcers and limit their ability to exact appropriately large fines when necessary. This problem was addressed by avoiding specific promises of numerical penalty reductions and maintaining the ability of the enforcement agencies to exercise their discretion. All that was promised was that they would consider implementation of the described activities to be good faith and to make reductions in assessed penalties as appropriate. For the regulated community, this appeared to offer sufficient encouragement.

But throughout the OTA's meetings with enforcement personnel, there was obviously more to the resistance to using the good-faith clauses in this way. Some time after the project was complete, the author of this Dialogue had a conversation with an EPA manager that provided some insight. The manager referred to good faith as the willingness of a violator to cooperate with the agency after the violation had been discovered and the enforcement process had begun. Looking at it from this perspective, the penalty reduction for good faith is not something to give away ahead of time. You hold it in reserve until you see if someone is cooperative or not. This view of good faith as applied in practice has since been confirmed by numerous other enforcement personnel.

For this reason it is important to distinguish the two types of good faith. The good faith recognized and promoted by the Massachusetts Auto Body Project may be termed precontact good faith—a demonstrable commitment to environmental performance before contact with the enforcement process. Precontact good faith creates a different enforcement response for company A, which is making an attempt to do the right thing, than for company B, which does not care. Postcontact good-faith rewards cooperation after the inspection has occurred and the penalties are being assessed, and unless the same companies who try to do the right thing on their own are the same ones who tend to cooperate, it does nothing to distinguish between company A and company B.

Acknowledging that recognition of precontact good faith is called for, attorneys have described the typical negotiation as beginning with virtually every company making a case that they are responsible. It can be difficult to sort out true good-faith efforts from the standard puffery.

Individual inspectors are better able to make these distinctions because they are in the field. Industry also knows that if they have a dirty shop and don't seem to care about the rules, the typical inspector will likely look harder for violations than if they have a clean facility and evidence many indications of making the required and recommended efforts. The problem is ensuring that the inspector as well as the attorney form an objective view of performance. Although we may believe that all inspectors are professional and do not allow personal pique to color their perceptions, if someone at a company is rude or does not show sufficient respect, information about a high level of environmental performance would help create a balanced view. Or, if a company has placed all of its efforts on something the agency considers a low priority, and has missed something the agency considers most important, an accurate representation of the company's intentions may be hard to obtain. The Massachusetts Auto Body Project is an example of a mechanism for documenting specific precontact good-faith activities to enhance the likelihood that the company will know and do what is called for and will be accurately assessed according to their actual performance.²⁴

An objective view of the company's performance must be independent of the vicissitudes of interaction with enforcement. A particular enforcement action focused on one issue, for example, compliance with the Toxic Substances Control Act, may afford no occasion for agency staff to consider a company's stellar hazardous waste reductions (which may or may not be relevant). An argument or appeal on good grounds can make a responsible company look like a recalcitrant. Enforcement attorneys do not conduct site visits, as do assistance staff, and don't generally meet company staff except in an adversarial context. They thus have little opportunity to encounter evidence of precontact good faith on a firsthand basis, uncolored by the skepticism necessarily inherent in the adversarial process. Enforcement staff learn about companies in the context of misdeeds. Conversely, assistance [32 ELR 10226] staff regularly encounter companies going the extra mile to do the right thing. That better recognition of precontact good faith is an issue is a direct result of the fact that providing assistance to the regulated community has led to a more complex set of perceptions of their performance.

Appropriate Use of the Tool

Good faith, as described here, can be a useful tool for the transparent and judicious exercise of enforcement discretion. But allowing too much leeway in penalty setting could be a dangerous tool, allowing corrupted or overly lax agencies to let polluters off the hook. The discretion afforded by the use of a vague term such as good faith, which is highly subject to individual interpretation, can open the doorway to abuse. That is why penalty policies are so valuable. They reduce the chances that the penalty-setting process will be inconsistent, politicized, or corrupted. The development of guides to what might afford penalty mitigation, like the penalty policies themselves, also provides the public with a method of checking abuse of discretion.

Specifying what constitutes good faith should provide the basis for its careful application. Tying its application to descriptions and definitions in widely accessible plain language documents will produce more generally accepted concepts of what it is. It will be transformed from a vague concept to a very specific set of activities: Are hazardous materials stored so that they will never reach the environment? Have you provided information on the risk of your facility to affected parties? Have you calculated how many pounds of chemicals you waste and release? Have you tried to stop using unsafe chemicals?

These specific, concrete questions currently are contained in clear guidance on millions of shelves across the land. People need to know that if they pull those documents off the shelf (or don't throw them in the trash when they arrive in the mail), and instead put them to use, that it could make a difference in their lives.

Relational Aspects of Environmental Governance

If one were to economically describe the relationship between government and the regulated population, and how it has changed over time, one might be tempted to borrow from the language of mathematicians.²⁵ This would oversimplify but clearly describe trends. If, for example, we denote the governmental agency charged with environmental progress as "A" for the authority, and symbolize the regulated population by the letter "P," we can sketch the early stages of the relationship as "A —>t P," with "—>t" standing for "telling P what it has to do." Although it may seem odd to discuss this aspect of governing, concerning the relationship of the parties, in what seems to be mathematical language, these symbols are merely graphic tools that may be used to talk about the relationship created by the command-and-control model.²⁶

If you were to describe the relational nature of environmental governance today you might place "A" at the center of a sun with rays going out and returning in every direction to surrounding stakeholders of every kind. This would include organized large business interests, organized sector-based interests, small business interests, labor, environmental organizations, environmental justice representatives, health professionals, related agency staff, "ordinary" citizens, academics, consultants, etc. Because it is hard to do that on a single line of text, we might describe this interactive relationship in a one-on-one basis, as "A <—> P," where "<—>" stands for "interaction."

That the environmental governance system would have evolved to be interactive is to be expected because of our rights-based democratic system. Our laws are developed and administered through a political process that responds to election results, the Administrative Procedure Act ensures that rules are established with notice and comment, and citizens have the right to judicial process to ensure appropriate agency action. That court orders now take up a large percentage of EPA business demonstrates that agencies can be made to be responsive whether they want to be or not.²⁷

But in addition to this, many environmental agency managers and staff have learned to conduct their business in a far more interactive manner. They conduct outreach and assistance and get to know the regulated population, they form advisory committees to gain input and form consensus, they establish partnerships, and they work closely with affected and other interested parties to develop programs that leverage willing cooperation.

Evolution of the Relationship Between Agency and Regulated Population

From a relational perspective, there is a natural evolution in environmental governance. The first stage, as noted above, results from the passage of laws and the delegation to the agency of responsibility to promulgate rules and ensure their observance. As described above, this first stage may represented as A —> P.

After a while, some portion of the P population, having noticed that others within the population have suffered adverse consequences from being in violation of the rules, will approach A on their own with questions: Do we need a permit? [32 ELR 10227] Are we covered by this rule? We want to be in compliance and don't want to be fined, so what do we need to do? These are essentially requests for compliance assistance. P will also have complaints and will appeal complaints. In the second stages of evolution of environmental governance, agency staff spend more time providing compliance assistance over the phone, through established hotlines, and in public speaking engagements. After receiving many similar requests, the agency recognizes the need for guidance and issues publications intended to clarify P's responsibilities. This second stage, incorporating feedback and response, may be represented as "(A —> P) + (A <— P)," where "<—" refers to questions, concerns, and complaints.

At this point we may characterize two aspects of A: Ae, and Aa. "Ae" is the enforcement aspect of A, and "Aa" is its assistance aspect. In response to queries and cooperative requests for guidance, Aa develops. In response to resistance, Ae develops further. (We may at this time discern also the beginning of "Ai," the portion of the agency that attempts to integrate the enforcement and assistance response; the "i" also standing for innovation or even—daringly or insultingly—intelligence.)

Internal Relationships Affect External Relations

Combining the activities of (A —> P) and (A <— P) can result in a fully interactive picture, as described above, A <—> P. But it may not. It may be that the two sides of the agency, Ae and Aa, split apart and present different, and perhaps even conflicting, faces to P.²⁸ Therefore, a new focus of relational analysis becomes necessary, that between Ae and Aa, to ensure a coordinated message is received by P. Pictorially representing the "transmission" portion of the relationship can look something like this:

[SEE FORMULA IN ORIGINAL]

P's responses to assistance and to enforcement need to be perceived and understood by Aa and Ae so that both have an accurate picture of the response to the agency's messages. The response of P, P —> Ae, and P —> Aa needs to be incorporated, and perhaps differentiated. Take the example of a company that takes pride in what it terms "excellence." Assume that the company has had a small internal spill, and even though it may not have had to report, it errs on the side of obedience to the law and reports the incident. The company is then visited by an inspector who, for one reason or another, fails to recognize that the company is essentially a responsible entity. Or, because of the inspector's focus on one medium, the company's overall performance is undervalued. The inspector then finds a minor paper violation, seeks payment of a fine, and asks for the maximum. Let us say that the company resents the action, regrets having reported the incident, and on principle utilizes the appeals process to resist any tarnishing of the company's reputation. The basic interaction may be represented as "'Ae —> v P' results in 'P —> r Ae,'" where "v" stands for citation for violation, and "r" stands for resistance. The agency should be concerned with keeping unnecessary resistance to a minimum, if for no other reason than to keep administrative costs to a minimum and to have the ability to use resources for addressing a wider range of problems or actors. In this example, the unnecessary resistance stemmed from a failure to recognize good faith and an improper assumption of bad faith.

In contrast, let us assume a different scenario in which the enforcement agency recognizes that this is a responsible company and that the spill is of a minor nature. It commends the company for erring on the side of reporting and suggests the company utilize the services of the agency's assistance program. Assistance staff visit the company and point out the minor paper violation without assessing any fines. The company rectifies the problem immediately. The assistance staff recognizes that the company overall has an excellent program, and (with the company's permission) sings their praises to the enforcement staff. The agency writes up a case study of the company's program and has them speak at public events, serving as a model for other companies to emulate. This may be depicted as "Ae + Aa —> p Per," where "p" stands for promote, and "Per" stands for performance. This is just one sample scenario wherein good faith is recognized. It is not dependent on the coordination of assistance and enforcement, but can occur wholly within the enforcement context. (In other words, the inspector can point out the violation without assessing fines.) But the practice of communicating the model behavior to others is the specialty of assistance offices, and the "safe" assistance office is more likely to be invited to evaluate questionable areas of the plant or to answer questions concerning possible violations.

It may be universally agreed that the situation in which Ae and Aa work together, resulting in improved performance, is the most efficient result. It may be that the "—> v" relationship of citing violations is necessary, and the "—> r" response of resistance is unavoidable, but if the former is not necessary and the latter is not unavoidable, it cannot be considered efficient in a transactional sense.

Benefits of Differentiating Actors

The above examples illustrate the importance of perception and accurate characterization of the entity in the relationship. Aa is predisposed to work with companies that are acting in good faith. They come to Aa asking for help. They are inevitably in contrast with companies that do not take any steps toward doing the right thing. For the Ae-Aa relationship to function efficiently and present an integrated message to P, Ae must also be able to distinguish between companies that are trying to do the right thing and those that are not. We may call those members of the regulated population who are acting in good faith "Pg."

Common sense demands that the agency differentiate between Pg and other members of the regulated population. EPA has been distinguishing between performers with its [32 ELR 10228] Performance Track Program and as part of the Common Sense Initiative's Metal Finishing Strategic Goals Program, which implements a four-level "tiered" approach.²⁹ The first tier represents high performers, whereas the second tier represents companies who would probably do the right thing but need help or a little pressure. The third tier is those who would probably like to go out of business but may be afraid of having to pay for a cleanup if they put their property up for sale (these need assistance to close operations), and the fourth tier perhaps ought to be put out of business. Assistance and recognition programs have been functioning at the state level for many years to reward and foster good behavior, but these programs mostly only recognize good-faith activities by the first tier—the high performers.

Although the new performance recognition and other initiatives have somewhat changed thinking, the essential posture of Ae has been that all companies are supposed to be in compliance, thus, they don't generally deserve credit for being in compliance because that's what they are supposed to do.³⁰ The job of enforcement is to penalize for being out of compliance. As discussed in the first half of this Dialogue, good-faith penalty reductions could be used more effectively to reward preexisting good faith by increasing the ability to distinguish between companies out of compliance.

Performing the job of providing assistance leads to an appreciation of the differences between companies and the fact that rules are so complex that few are in complete compliance. This necessitates a recognition that the difference between a company trying to be in compliance and one not trying is important. Aa recognizes this at a basic level because those not trying do not ask for assistance, and those trying are Aa's customer service population. Whether or not companies ask for assistance, the distinction between those trying in some way, and those not trying at all, can be made at all levels, concerning companies in each tier. By looking at a company's effort, intent, and capability and not just its compliance record (and comparing to the performance of the sector or class to which they belong) we may identify a set of Pg actors larger than those in the first tiers of performance programs, and discern degrees of good faith within that larger set.

The Massachusetts Auto Body Project's checklist, to be filled out frequently and kept on file, was an example of a tool for documenting past efforts that could support a showing of good faith in the context of discovered violations. Having the guidance be distributed with a letter from top enforcement officials, and ensuring that it was received and used by all relevant enforcement staff, was an example of action to create a uniform definition for the enforcement agency to increase the chance that imprecise, vague factors such as "intent" and "effort" would be consistently and objectively interpreted.

Separating Pg from the rest of P, at all levels of performance, would increase the efficiency of enforcement targeting. Generally (and referring primarily to companies at the top of the Pg range), Pg companies are more likely, as in the example given above, to come into compliance with assistance, and non-Pg companies need enforcement attention to do so. Using the term "Pe" to refer to those companies needing enforcement to do the right thing, we can depict the ideal, efficient relationship as "Ae —> Pe" and "Aa —> Pg." At this point it is necessary to remind ourselves that these pseudomathematical expressions are oversimplifications. Aa should also be available to assist Pe members who ask for assistance, and Ae should have enough of a presence with Pg to ensure that we have accurately characterized the Pg companies as responsible and to provide the public with assurance that they remain so.

Strategic Relational Management

There are a number of ways to differentiate the populations. One is to forestall using enforcement tools when a company has established a relationship with assistance staff. It is important, however, to consider that assistance is generally, and most successfully, conducted on a confidential basis. Companies not sure about whether they are in compliance don't want to ask for clarification from someone who might then fine them or refer them for possible fines. They want advice from someone who is safe. The general rule for assistance programs is to report only serious violations or imminent threats.³¹

A better and typical approach is to separate the enforcement and assistance functions so that enforcement proceeds independently. This preserves the Ae presence for all populations, which is necessary, or the Pg population will decline, and can be represented as "Ae = 0, Pg = min." The population acting in good faith is minimized when there is no threat of enforcement.

[32 ELR 10229]

Although some professionals in the assistance community, and likely more in the business community, may disagree, many assistance program staff know that the primary driver for most companies, even those in the Pg category, is the need to avoid enforcement.³² Even for those companies dedicated to excellence and preserving the environment, the pressures of making money can trump all other goals. Each company has a responsibility to meet payroll, maximize profits for owners or shareholders, stay abreast of changing markets, grow, or die. The prospect of enforcement is required or moral imperatives can end up taking a backseat to economic drivers.

But how do we balance preserving the Ae function with avoiding the inefficiency of its unnecessary application to those who are willing to do the right thing on their own or with just a little help? It may be simple and clear enough to state that we ensure its appropriate application. Determining what is appropriate then would be dependent on the formation of accurate perceptions of the good faith of the regulated population and on paying close attention to the relational aspects of environmental governance. This underscores the importance of a functional relationship between Ae and Aa because Aa can provide critical help in both of these areas.

Whether a company is in a relationship with an assistance office is thus one factor that should be considered, and Ae can manage the confidentiality aspect by simply not inquiring about that relationship unless it is necessary. Ae need not predetermine whether Aa has a relationship with an individual P, but should seek to determine—from the company itself—whether there is a positive relationship after engaging in an enforcement process with a particular P.³³ This way, Ae will not infringe on the trust relationship that Aa needs to create and preserve with all Ps, but it can adjust its response after it has begun to engage in an enforcement process.

In terms of general populations, as opposed to specific individual Ps, Ae should make efforts to target its enforcement initiatives according to general information about groups of Ps that are less likely to be in the good-faith category. To do this, Ae can use generic information supplied by Aa. For example, if Aa has been out in the field visiting many kinds of businesses (seeing things that companies hide from inspectors, and hearing things that companies would never ask or tell inspectors), Aa will have developed a sense of what sectors or kinds of businesses are likely to be in violation or to need the enforcement presence. Aa may have formed opinions that, for example, medium sized businesses in a certain region have not paid attention to the need for air permits. Or they may have noted that trucking companies hauling hazardous materials have a lot of issues and are generally hoping they will never be inspected. Aa can report this generic finding to Ae, which can then begin a program of inspecting trucking companies.³⁴

Ae can then visit some companies within the sector, and after doing so, indicate to each company that it is receptive to information about the company's good-faith efforts. In this way Ae would seek to determine if the violations are the result of the company not having completed an effort to come into compliance or if the violations are the result of the company never having tried. At this point the company that sought assistance should understand that it can receive some credit for having made an effort.

It is important to note here that there are reasons not to provide credit to anyone simply because they asked for assistance, or companies may go through the motions without intending to take actions to rectify violations or reduce risks. Companies originally referred to assistance by enforcement are in a different class from companies who initiated the relationship on their own. Companies who have acted on the advice received, or whose request for help was part of a self-initiated program of improvement, are different from those who ignored the advice and have no improvement program.

Having a relationship with Aa should be regarded as a very important indication that the facility may be trying to do better. Ae should look further into the matter and seek information on whether the company has taken action. At this point, the company may provide that information and may wish to waive confidentiality and authorize Aa to provide confirmation of the company's good-faith efforts to Ae (there should be no pressure to waive confidentiality).

Aa's responsibility will be to provide an honest evaluation and not to be a representative or advocate of the company. Ensuring that this is understood should create confidence on the part of Ae and avoid confusion on the part of the company. As each side will have expectations from Aa, this is an important point to clarify up front.

If Ae takes this approach, the effect of targeting and inspecting companies within the sector will be twofold. The message sent to the sector will be that the rules must be taken seriously, and that attempts to come into compliance will put one in a better position.

What this achieves is a realistic approach to environmental governance. Let us contrast it with the current, idealistic approach. The current approach of assuming that everyone is in compliance and being shocked that they are not does not reward attempts that don't completely succeed. It is an all or nothing approach.

[32 ELR 10230]

Although the accurate perceptions of individual inspectors in the field and attorneys able to see past what seems like self-puffery moderate this outcome, this approach can treat a company with a minor, inadvertent violation the same as a company that doesn't give a hoot. Assistance personnel know that companies often expect this outcome if they are unlucky enough to be visible actors and the enforcement agency is attempting to send a message to a sector or region.

Declaration of Policy

A relatively simple way to differentiate and to effect differentiation in the populations is to declare that existing penalty-setting policies, which provide for reducing fines when a company is in good faith, includes precontact good faith. This means that when Ae visits company X and there are indications that company X has been trying to comply (albeit unsuccessfully), the violations will be seen within context, and company X will be fined less than company Y that has not been making an effort.

The P population needs to know what indicators Ae will examine. Clarifying these expectations will promote the desired actions by P. One simple way to do this is to describe the desired actions, such as that the company with an environmental management system or pollution prevention plan in place will fare better than others if the system or plan meets certain standards of adequacy. These standards should be developed by Ae in conjunction with Aa.

For example, a company may claim that it has a pollution prevention plan in place, but it should not get credit if that plan is merely a paper exercise. It should get credit if, for example, it can show that it is truly tracking its materials and performing a mass balance so that losses and trends in chemical use are identified. It should get credit if it can show that it has made a serious effort to look at alternatives. It should get credit if it can show that it has truly assessed the total cost of its current toxic chemical use so that the alternatives will receive a fair comparison. Clarifying and communicating minimum standards such as these will lead to the increased implementation of these activities, thereby greatly increasing the ability of the agency to promote desired behavior and accomplish its mission of environmental protection. Individual enforcement staff in A may be perfectly willing to go easier on good companies, but this information needs to be transmitted. A basic relational activity—clear and understandable communication—needs to take place, as demonstrated by "A —> s P," where "s" means "signal." Environmental agencies must consider the basic signal now sent by referring regulated entities to stacks of complex regulations. A clarified signal will improve the response of message recipients.

Ae should examine all the guidance and assistance publications that have been issued by Aa. These contain recommendations for companies. Some of these are not explicitly required, and some of the recommendations concerning requirements bring a company partly into compliance but not all of the way. The conventional Ae posture could be to completely discount the fact that a company may have implemented the recommendations. Ae should examine these recommendations and provide public confirmation that following Aa regulatory recommendations, though short of complete compliance, will count for something. Similarly, Ae should confirm that following Aa recommendations concerning activities that are not required but are good for the environment also will count for something.³⁵

Response Posture

Ae can adopt the response posture that Aa uses when requests for assistance are made. Many Ps will ask for help from Ae, either because they are already in a relationship with Ae, because they do not know about Aa, or because they wish to have a more definitive answer from Ae, as Ae is the party that makes final determinations. (Aa's regulatory interpretations are not definitive.) Ae should exercise care to distinguish situations that come to its attention by virtue of a P's willingness to come forward. Doing so will encourage Ps coming forward and volunteering information, exposing themselves to the risk of fines in the service of attempts to do the right thing. It would seem that the vast majority of Ae professionals do respond appropriately, but the isolated negative incident can do a great deal of damage because word about it can travel very quickly and perhaps be exaggerated as well. An established policy could be used to prevent these incidents and mitigate the harm if they do occur.

EPA has accomplished some of the development of this relational aspect by issuing the Small Business Compliance Policy for smaller companies and a self-audit policy for larger businesses.³⁶ These policies reward companies for self-reporting violations. But several conditions must be met and violations must be reported. A more general recognition of the principle that voluntarily coming forward always counts for something would cover situations where companies do not actually report their own violations and where they do not meet all the conditions. Although it is reasonable to say that no company should receive a complete waiver for repeat violations, as these policies set forth, it is also reasonable to consider the message sent if good faith is only rewarded in limited situations. If there are two companies that have both had repeat violations, and one is discovered because they asked for clarification on an issue, penalizing that company while the other remains undiscovered and, thus, suffers no penalty sends a clear message that asking questions carries unacceptable risks.

Appropriate for the Smaller Entity

It is universally admitted that our environmental rules are generally complex and difficult to understand. Our expectation has been that companies will hire experts and devote the resources to mastering them. But as our system evolves, it [32 ELR 10231] inevitably focuses on smaller and smaller Ps, as the larger sources of pollution are brought under scrutiny and control and the smaller sources constitute a larger and larger portion of the P that actually threatens public safety, health, and the environment. This population has less and less of an ability to master environmental requirements, and they are less likely to be inspected.³⁷ A simpler system and message is not merely more appropriate for them, it is perhaps the only way they will ever be reached.

Fully Equipped Environmental Governance

At the same time that we are considering applying our environmental requirements to a greater population, our ability and dedication to providing assistance, creating partnerships, recognizing performers, and generally working with the Pg population has increased. We now risk sending two messages to the public: Ae's message is "we will hit you," whereas "we will help you" is the message of Aa. (Ae —> hit, Aa —> help.) The assistance and enforcement messages are two tools for environmental protection and they can have a strong effect if properly combined in a sensible message. We will likely reach our environmental goals most quickly if we use all the tools at our disposal in a combined and synergistic fashion.

Tailoring Messages for Recipients

No message can effectively be sent without adherence to the principles of communication, one of which is that communication is a two-way affair. Thus, the nature and point of view of the recipient is an essential matter. Considering the message recipient is a basic activity necessary to a successful relationship. We need environmental agencies that pay attention to relational aspects and distinguish between Pg and Pe, those acting in good faith and only needing the implicit presence of enforcement and its potential application, and those needing the immediate application of enforcement mechanisms. Relational environmental governance will result in clear messages, simple enough to be communicated by symbols, such as "Ae —> hit Pe, Aa —> help Pg." If Ae and Aa coordinate well, this message won't become confused. Then those Ps who have the capacity or motivation to do so will sit up and say: "I want to be in the Pg category even though I don't understand the rules and can't be sure that I have no violations." Calibrating the enforcement response more precisely to the effort of the regulated entity will seem fair, which should be invaluable in generating support for our system of environmental governance.

Benefits of Good Relational Practice

Acting more appropriately on the basis of more accurate characterizations of the performance of the regulated population is a far better way of improving the relationship of government and citizenry than simply reducing the level of enforcement, which is one politically touted response to criticism of the command-and-control system. Rather, we should ensure that the environmental regulatory system will be more likely to be perceived as just and efficient. Antigovernment attitudes, experienced by many environmental agency staff, should decline and the entire system of environmental governance should enjoy a higher level of respect and consideration. Simply cutting back on enforcement will not achieve this result, and, in fact, it may provide seeming ratification of even the most unsupported criticisms.

Appeals should be reduced, negotiations should be easier to conduct, and compliance will increase. More people will probably go beyond compliance. We won't have to go through the painful process of rewriting our rules every time we need to adjust their effect because we will have developed a clear way of sending signals to the regulated population about desired behavior, and we can then rewrite rules at a more leisurely pace based on what we will have learned from trial runs of guidance.

If the regulated population were properly motivated to come forward when necessary, and did not have to fear punitive repercussions when asking for help, we could describe the relationship as a healthy interaction: A <—> P.

Conclusion

The method of discussion, whether mathematical or otherwise, is relatively unimportant to the issue of environmental regulation. Moreover, it need not be conducted only in the established, professional forums that now exist. It is a concern not just of the mandated agency but of the regulated population and all other stakeholders as well. The quality of environmental governance is an issue that bears directly and heavily on the chances for success of solving environmental problems through democratic means. Further, to address environmental problems democratically, it is imperative that the alternatives be considered. At all costs we must attempt to avoid the choice between solving these vital problems and preserving the self-government aspects of our system. Good faith as a fundamental principle springs from and reinforces the self-governing aspect. It fosters a trust relationship between government and governed, promoting the positive evolution of environmental governance from an adversarial relationship to a nuanced set of relationships, finely adjusted for each member, and inclusive of the potential for alliance.

1. MASSACHUSETTS OTA ET AL., MASSACHUSETTS AUTO BODY ASS'N, MASSACHUSETTS CRASH COURSE PROJECT MANUAL AND TOOLBOX (COLLISION REPAIR AUTO SHOP HELP) (1998), available at http://www.state.ma.us/ota/otapubs.htm. Project Managers: Rick Reibstein and George Frantz. EPA Project Coordinator: Mary Dever. Credit must go to Susan Leite, Project Staff, and George Frantz, Project Coordinator, for the title. Along with Stephen George, these three completed the bulk of the detailed work involved in implementing the Massachusetts Auto Body Project. The author, originator of the good-faith approach and project supervisor, also wishes to credit Frantz with substantial organizational and philosophical contributions to the project. The OTA's director, Barbara Kelley, provided crucial support. Acknowledgment must also be provided to many others, including: Mary Dever, Anne Leiby, and Tom Olivier of EPA Region 1; Kim Kreiton, Paul Reilly, John Reinhardt, Salvador Resurreccion, and Nancy Wrenn of the Massachusetts Department of Environmental Protection (DEP); Anne Berwick, Margaret Van Deusen, and Mary Griffin of the Office of the Massachusetts Attorney General; Nancy Comeau of the Massachusetts Department of Labor and Workforce Development, Division of Occupational Safety; Evangelos "Lucky" Pappageorge of the Massachusetts Auto Body Association; Steve Berard of the Massachusetts Division of Standards; and Steve Vermette of the Boston Police Department. Project funding was provided through grant No. 20000162 from EPA, and CRASH Course manuals were prepared by Tetra Tech EM, Inc., under the OTA's direction.

2. An exception to this generality is that the Boston Environmental Strike Team (BEST), which included the Boston Police Department, had conducted surprise checks on auto body shops not only to observe environmental violations but also to find illegal "chop shops" dismantling stolen autos for the purpose of reselling parts.

3. Letter from Trudy Coxe, Secretary, Massachusetts Executive Office of Environmental Affairs et al., to Auto Body Shop Owners or Operators, CRASH (Collision Repair Auto Shop Help) Course for Compliance and Pollution Prevention Manual (Oct. 15, 1998) (on file with author). Other signatories were: John P. DeVillars, Regional Administrator, U.S. EPA; David Struhs, Commissioner, Massachusetts DEP; Scott Harshbarger, Massachusetts Attorney General; and Robert J. Prezioso, Deputy Director, Massachusetts Division of Occupational Safety. The letter stated that using the CRASH Course self-assessment checklist could "demonstrate to an inspector your good faith effort to comply."

4. A survey of auto body shops, which resulted in nearly 100 responses, found that the distribution of the CRASH Course materials resulted in increases in attention to hazard communication, the use of personal protective equipment, appropriate management of hazardous materials and hazardous waste, pollution prevention, housekeeping, and records management.

5. 42 U.S.C. § 6928(a)(3) (Resource Conservation and Recovery Act (RCRA)), ELR STAT. RCRA § 3008(a)(3). See also U.S. EPA, RCRA CIVIL PENALTY POLICY 30 (1990), ADMIN. MAT. 35281, 35282 (available from the ELR Document Service, ELR Order No. AD-2966) [hereinafter RCRA CIVIL PENALTY POLICY].

6. RCRA CIVIL PENALTY POLICY, supra note 5, at 39, ADMIN. MAT. 35283.

7. For practical and political reasons, at the least. Orders should create desired effects and not waste time and money. In addition, several statutes and executive orders advance the general principle that agencies must be mindful of the affects of their actions. For example, Executive Order No. 12291 requires a regulatory impact analysis of major rules determining "significant adverse effects on competition or employment." Exec. Order No. 12291, 3 C.F.R. 127 (1982), ADMIN. MAT. 45025. In addition, the 1980 Regulatory Flexibility Act (RFA) instructed EPA to carefully consider the economic impact of its rules on small entities. 5 U.S.C. §§ 601 et seq. The 1996 Small Business Regulatory Enforcement Fairness Act (SBREFA) amended the RFA to adopt a range of specific activities to reduce the impact of rules on small entities, including the establishment of programs "which allow for reduction or waiver of civil penalties for small entities under conditions where the violator demonstrates good faith efforts to correct violations and comply with the law." OFFICE OF POLICY, PLANNING & EVALUATION, U.S. EPA, SMALL BUSINESS REGULATORY ENFORCEMENT FAIRNESS ACT FACT SHEET (1997) (EPA No. 100-F-96-038) [hereinafter U.S. EPA SMALL BUSINESS FACT SHEET]. The fact sheet notes that the Agency's Policy on Compliance Incentives for Small Businesses, U.S. EPA, POLICY ON COMPLIANCE INCENTIVES FOR SMALL BUSINESSES (1996), ADMIN. MAT. 35649 (available from the ELR Document Service, ELR Order No. AD-2941), predating the SBREFA, provides that EPA:

Will eliminate penalties for small businesses which discover first-time violations through on-site, government-supported, compliance assistance programs or voluntary self-audits, where the businesses promptly disclose and correct the violation and have not engaged in criminal conduct. If a small entity is found liable for violations, but can demonstrate that the government's penalty request was substantially in excess of the court's judgment and is unreasonable under the circumstances, then it may request the court to award attorney fees if the entity did not violate the law willfully or act in bad faith.

U.S. EPA SMALL BUSINESS FACT SHEET, at 3 (emphasis added).

8. REGION 1, U.S. EPA, POLLUTION PREVENTION WHITE PAPER (1993) (issued by Acting Regional Administrator Paul Keough to New England State Commissioners, State Enforcement Contacts, State Pollution Prevention Contacts, Region 1, and Headquarters personnel). The White Paper dealt with issues arising as a result of the effort to incorporate pollution prevention into all programs, following the Pollution Prevention Act of 1990 and national and regional pollution prevention strategies. It focused on "The Relationship Between Compliance/Enforcement and Voluntary, Non-Regulatory Programs," and specifically noted the difference between information provided in an assistance context and that provided by an inspector. It warned against "misunderstandings," whereby a facility could interpret advice as "an Agency order or a way to avoid enforcement." It stated that "any advice that is given should be issued in an informal manner with the caveat that it is not binding in any way and is not in lieu of possible enforcement action." Id. at 7.

9. U.S. EPA, Final Reissuance of National Pollutant Discharge Elimination System Storm Water Multi-Sector General Permit for Industrial Activities, 65 Fed. Reg. 64795, 64841 (Oct. 30, 2000).

10. U.S. EPA, Final National Pollutant Discharge Elimination System Storm Water Multi-Sector General Permit for Industrial Activities, 60 Fed. Reg. 50803, 50984-92 (Sept. 29, 1995).

11. Id. at 50986.

12. The 1990 Massachusetts Toxics Use Reduction Act (TURA) and its regulations (MASS. REGS. CODE tit. 310, §§ 50.00-.59) require that planning cover certain elements (generally, declared management support; process characterization; materials accounting on a process level and for facilitywide mass balance; options identification, including solicitation of ideas from employees, assessment of the full cost of toxics use, evaluation of option feasibility and option selection; goal setting, and program evaluation). MASS. REGS. CODE tit. 310, §§ 50.40-.49, implementing MASS. GEN. LAWS ch. 21 I, § 11 (1990). The company is not required to select any predefined activity, such as substituting safer materials for identified high hazard materials, even though alternatives may be known to the authorities as practical and implementable. Similarly, EPA's Guidance to Hazardous Waste Generators on the Elements of a Waste Minimization Program states that

EPA believes waste minimization programs should incorporate, in a way that meets individual organizational needs, the following basic elements common to most good waste minimization programs: (1) top management support; (2) characterization of waste generation and waste management costs; (3) periodic waste minimization assessments; (4) appropriate cost allocation; (5) encouragement of technology transfer, and (6) program implementation and evaluation.

U.S. EPA, GUIDANCE TO HAZARDOUS WASTE GENERATORS ON THE ELEMENTS OF A WASTE MINIMIZATION PROGRAM (1993), ADMIN. MAT. 35547 (available from the ELR Documents Service, ELR Order No. AD-2961) (emphasis added). A 1996 compendium of state pollution prevention laws found 18 states with mandatory pollution prevention planning, all with similar elements, varying mostly in the degree to which they defined what chemicals or wastes must be covered and the degree to which they specified the inclusion of generic planning elements. NATIONAL POLLUTION PREVENTION ROUNDTABLE, THE SOURCE: THE ULTIMATE GUIDE TO STATE POLLUTION PREVENTION LEGISLATION (1996). For an evaluation of the effectiveness of such programs and showing substantially greater progress in states with mandatory as opposed to voluntary planning requirements, see Heather M. Tenney, A Comparison of Voluntary and Mandatory State Pollution Prevention Program Achievements (2000) (unpublished thesis, Civil and Environmental Engineering, Tufts University) (on file with author).

13. OFFICE OF ENFORCEMENT & COMPLIANCE ASSURANCE, U.S. EPA, EPA'S SUPPLEMENTAL ENVIRONMENTAL PROJECTS POLICY, QUESTIONS AND ANSWERS FOR THE PRACTITIONER (1999), available at http://es.epa.gov/oeca/sep. No one is forced to do an SEP, but "the final settlement penalty generally will be lower for a violator who agrees to perform an acceptable SEP compared to a violator who does not agree to perform [an] SEP." Id. As a prime example of protection in addition to what is achieved by compliance, the Agency's Q&A policy states that "SEPs involving pollution prevention are preferred over those calling for pollution reduction or control strategies." Id.

14. Proposed projects must meet EPA's criteria, including fitting into a specific approved category (public health, pollution prevention, pollution reduction, environmental restoration and protection, assessments and audits, promotion of environmental compliance, and emergency planning and preparedness), and having a "nexus" (a relationship between the violation and the project).

15. In Massachusetts, inspection of TURA planning has largely consisted of checking on the presence or absence of required elements. But each region has designated TURA specialists who are able to conduct a more in-depth evaluation if necessary. Nevertheless, staff at the OTA, which provides confidential assistance on TURA, have found that plans exhibit a large variety in terms of quality.

16. 40 C.F.R. pt. 112 (2000) (implementing CWA § 1321(j)(1)(C), ELR STAT. FWPCA § 311(j)(1)(C)).

17. TURA's regulations state that "toxics users shall develop information required (by the regulations) in good faith." MASS. REGS. CODE tit. 310, § 50.42(10) (1993) (emphasis added). The regulations also state that "toxics users shall demonstrate a good faith and reasonable effort to identify and evaluate toxics use reduction options." Id. § 50.42(11) (emphases added). What constitutes a good-faith effort is not further defined. Although OTA staff have found that companies' beliefs that they could face penalties for failing to conduct a good-faith effort has motivated many companies to do a thorough job, some companies seem to have reduced efforts as a result of a perception that this aspect of the law is not enforced. In contrast, the public reporting of chemical use has proved a consistently significant motivator for use reduction, at least for those featured in the press as the largest chemical users.

According to Tracy W. Klay, the former director of compliance and enforcement activities for the Massachusetts DEP's TURA program, the DEP "has placed practice restrictions on some [toxic use reduction] planners in violation of the certification requirements." Tracy W. Klay, TURA Liability in Massachusetts: A Trap for the Unwary, Burns & Levinson, LLP, at http://www.b-1.com/publications/articles/Klay. Tura_liability_in_Massachusetts.html (Jan. 2001) (first appeared in the Associated Industries of Massachusetts'e-mail newsletter, EHS NEWS, on Jan. 23, 2001). (All plans must be certified by "toxic use reduction planners" who must themselves undergo certification to be a planner.) Under the regulations, the DEP may "suspend, deny or revoke a planner's certification, or deny recertification for any good cause." Id. Klay details the violations for which the DEP typically assesses fines:

No TURA Plan or done incorrectly; the number of chemicals that should have been reported, in conjunction with a penalty for lack of a TURA Plan; no Form R and/or Form S or the late filing of either or both; no or late Plan Summary for each reported chemical; and no or late or not properly dated, signed or executed Plan Certification Statement.

Id. (emphasis added). The TURA regulations provide that the DEP "may determine that a plan, summary or update is deficient" if it "was developed in bad faith." MASS. REGS. CODE tit. 310, § 50.49(4).

18. The 1946 Administrative Procedure Act (APA), grants a right to judicial review to "[a] person suffering legal wrong because of agency action, or adversely affected or aggrieved by agency action." 5 U.S.C. § 702 (1946). The APA also requires notice of rulemaking and response to public comments. Most importantly for regulatory incorporation of guidance is the issue of whether notice-and-comment procedures are necessary. In Chamber of Commerce v. Department of Labor, 174 F.3d 206 (D.C. Cir. 1999), an Occupational Safety and Health Administration directive establishing a cooperative compliance program (CCP) was vacated. The court held that the CCP, which "effectively obligates employers, under pain of certain inspection," to adopt recommended safety practices, was a standard requiring notice and comment. A voluntary program that is implemented by coercive means may be the equivalent of a rule requiring APA promulgation procedures. One of the aspects of the CCP leading to the decision was the lack of discretion provided to inspectors in the field. The court noted that in American Bus Ass'n v. United States, 627 F.2d 525 (D.C. Cir. 1980), it had found that a "rule" that has only a prospective effect and leaves agency decisions free to exercise their informed discretion in individual cases is a policy statement, excepted from notice and comment. But as noted by Daniel Steinway and J. Barton Seitz, "regulated parties should be wary of new programs adopted by EPA that are said to be voluntary in nature, but impose subtle sanctions on companies that do not participate. Some of these programs may unfairly penalize companies that do not want to participate for legitimate reasons." Daniel Steinway & J. Barton Seitz, EPA Voluntary Programs: Will They Come Under Attack?, Pollution Engineering (Feb. 2000), at http://www.pollution engineering.com/compliance/legal/lra2000/lra0400.htm.

19. Steven A. Herman, recent head of EPA's enforcement office, in an interview with Milo Mason, described the SEP and Self-Audit policies as "positive incentives," and stated that "incentives are an important component of our program." Milo Mason, 12 NAT. RESOURCES & ENV'T 286-87 (1998), cited 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, 15 (2000). To avoid misunderstanding the enforcement chief's recognition of the importance of positive incentives, note his oft-stated opinion that ". . . formal law enforcement is the central and indispensable element of effective governmental efforts to ensure compliance." Markell, at 9.

20. In order for this to become standard practice, however, the connection or "nexus" between the statutory mandate and the actions to be rewarded must be established. Without the nexus or independent authorization it may be difficult for a water inspector to recognize and reward compliance that goes beyond hazardous waste performance.

21. "Participants must satisfy specific performance criteria designed to ensure that they exceed regulatory requirements." Memorandum from Steven A. Herman, Assistant Administrator of the Office of Enforcement and Compliance Assurance, and Richard T. Farrell, Associate Administrator, Office of Policy, Economics, and Innovation, to EPA Administrators and Counsel, on Enforcement and Compliance Principles for the National Environmental Performance Track Program (Jan. 19, 2001) (on file with author). After examining performance recognition programs in 13 states, EPA concluded that "all reviewed require facilities to commit to future performance improvements," most "require some type of public outreach and reporting," and "all state programs reviewed expect that member facilities will be in full compliance with environmental regulations." See U.S. EPA, Linkage to State Programs, at http://www.epa.gov/performancetrack/partners/linkage.htm (last modified July 31, 2001).

22. For example, the Massachusetts Environmental Results Program, which replaced permitting with self-certification for certain sectors, defines "compliance" as meeting what is set forth in streamlined versions of state environmental requirements, as opposed to the much larger body of regulations from which the streamlined versions are culled. To summarize all of the requirements in a small booklet required essentially leaving out some features of the law and rewriting others.

23. Standards Applicable to Generators of Hazardous Waste, 40 C.F.R. § 262.34(g) (2001).

24. Penalty policies provide for the recognition of both precontact and postcontact good faith (recognition of precontact good faith is built into the choice to pursue penalties and the weighting of factors in penalty calculation, and attempts to comply before contact with inspectors are explicitly cited in penalty policies). Good faith can also be encompassed by such language as "other factors as justice may require" or "public interest," which can be used to reward the promptness of a violator's corrective actions. Many attorneys contacted by the author on this issue have referred to using good faith in settlement, essentially referring to postcontact cooperative behavior.

25. An alternative to bare mathematical expressions would be an artist's rendition. An artist would be able to render the interactive arrows so that they depict the quality of the interaction. Red could be angry, wiggly or frayed could be unreliable, and jagged could be alternating mixed messages. This Dialogue leaves to others, or another time, further exploration of the concept of graphically representing the relational quality of environmental governance.

26. The risk of doing so is to seem to treat the topic too lightly and perhaps thus not contribute toward its serious analysis. The virtue of doing so is to draw attention to the subject and to explore methods for elucidating it and developing a serious response.

$=S

27. Often, "compliance with court orders has become the Agency's top priority," with courts deciding what issues will get attention. An example is a 1984 decision ordering EPA to set effluent guidelines (technology-based discharge standards) for toxic pollutants under the [CWA]. This ruling committed EPA to a [10]-year effort . . . .

DANIEL J. FIORINO, MAKING ENVIRONMENTAL POLICY 83 (1995) (quoting Rosemary O'Leary, The Impact of Federal Court Decisions on the Policies and Administration of the U.S. Environmental Protection Agency, 41 ADMIN. L. REV. 549, 554 (1989) and referring to Natural Resources Defense Council v. EPA, 21 Env't Rep. Cas. (BNA) 1823 (N.D. Cal. 1984)). Former Administrator Douglas M. Costle, in an oral history interview in August 1996, said: "Court decisions could undermine our resource allocation process. Court rulings often reordered priorities." U.S. EPA, Douglas M. Costle Oral History Interview, Courts (Jan. 2001), at http://www.epa.gov/history.

$=S

28. EPA has been both a force for innovation and a counter-force: a financier and promoter of numerous pilot projects and an enforcer preventing those pilots from getting off the ground. That internal contradiction should come as no surprise: EPA has never been a monolith. Rather it is a collection of regions, programs, offices, and individuals loosely bound by shared goals, leaving ample room for entrepreneurs in one corner of the organization to promote ideas that good soldiers in other parts of the organization feel compelled to frustrate.

NATIONAL ACADEMY OF PUBLIC ADMIN., ENVIRONMENT.GOV 58 (2000). The report notes that states "are prone to the same contradictions." Id.

29. The National Performance Goals and Action Plan document for this program envisions changes that will reach "all stakeholders," not just the top performers, and states that the measure of success will not be limited to the achievement of the goals but will also consist of the "extent to which they promote fundamental underlying changes," including a shift from "one-size-fits-all" standards and a shift "to greater use of a performance-based approach that rewards environmental excellence as much as it punishes non-compliance." COMMON SENSE INITIATIVE, U.S. EPA, METAL FINISHING STRATEGIC GOALS PROGRAM 9 (1997). In furtherance of these aims, the program identifies four tiers and identifies needed actions for three categories ("those that are at least close to compliance and will make improvements as able; those that would go out of business if they could afford the 'site-transition costs'; and those that are consistently out of compliance") Id. at 18. The National Performance Goals and Action Plan document is contained in the program's Welcome Aboard Kit, which was issued to program partners in May 1998.

30. Succinctly stated by an EPA Region 1 enforcement attorney in response to the author's brief description of the idea set forth in this Dialogue at a meeting of attorneys. Although penalty policies also contain this idea—the polychlorinated biphenyl penalty policy states that the Agency expects "that all reasonable measures will be taken to ensure compliance"—there is recognition that there are different levels of knowledge or control which necessitate distinguishing between levels of noncompliance. U.S. EPA, PCB PENALTY POLICY 15 (1990). Accidents can happen even when "prudent measures to avoid it" have been taken. Id. Although the RCRA penalty policy notes that "no downward adjustment should be made if the good faith efforts to comply primarily consist of coming into compliance," actions to correct violations before detection by the agency, or to prevent their recurrence may be rewarded. RCRA CIVIL PENALTY POLICY, supra note 5, at 33.

31. The General Laws of the Commonwealth of Massachusetts state that the assistance office shall not make available to the DEP:

Information it obtains in the course of providing technical assistance to a toxics user, unless: (i) the toxics user agrees that such information may be available to the department; or (ii) the information is public record information; or (iii) the information pertains to an imminent threat to public health or safety, or to the environment; or (iv) disclosure to the department is required by law.

MASS. GEN. LAWS ch. 21 I, § 7(E).

32. Based on the author's experience developing and managing a pollution prevention technical assistance program that visited over 1,000 facilities, and on over 13 years' worth of conversations with others in the field.

33. The OTA has asked the DEP and other enforcement agencies not to request OTA letters of suggestions to companies, but rather to suggest to companies that if they want to convince the agency that they have been acting in good faith, they may provide such a showing. OTA letters, or waivers of confidentiality, may then be useful for that purpose at the company's election.

34. Many assistance personnel feel that association with enforcement should be limited to avoid the perception on the part of the regulated community that the assistance program is "unsafe." It is unclear how sharing generic information with enforcement staff will create a negative perception on the part of the regulated population, unless they have somehow assumed that the assistance branch never talks to the enforcement branch. The OTA and many other assistance programs only promise that they will not share information about the particular visited facility (unless it pertains to an imminent threat, see supra note 31). The OTA is an independent assistance agency; assistance branches of agencies also responsible for conduct enforcement may have similar agreements but with lower thresholds for sharing information about particular facilities. For example, at EPA Region 1, assistance staff are to refer "significant" violations to enforcement, not just imminent threats. Reporting generic information that can then be used by the enforcement staff to better target inspections is generally in the interest of responsible companies, which neither wish to be undercut by competitors foregoing the expenses of compliance nor to be the ones to point fingers and report colleagues.

35. EPA's National Compliance Assistance Clearinghouse is a website that links to guidance and other tools that have been developed by EPA regions and states across the country. Go to http://cfpub1.epa.gov/clearinghouse/index. Or, consult EPA's Envirosense website at http://www.epa.gov/envirosense to find guidance on pollution prevention and/or compliance (such as the EPA's Sector Notebooks), or links to the many state programs providing this kind of information.

36. U.S. EPA, Small Business Compliance Policy, 65 Fed. Reg. 19630 (Apr. 11, 2000), available at http://es.epa.gov/oeca/smbusi.html (also available from the ELR Document Service, ELR Order No. AD-4481); U.S. EPA, Incentives for Self-Policing, Discovery, Disclosure, Correction, and Prevention of Violations, 65 Fed. Reg. 19618 (Apr. 11, 2000), ADMIN. MAT. 35764, available at http://es.epa.gov/oeca/finalpolstate.pdf (also available from the ELR Document Service, ELR Order No. AD-4485).

37. Richard Reibstein, The Small Pollution Source, POLLUTION PREVENTION REV., Summer 1992, at 287.