An essay on environmental audit privileges: the right problem, the wrong solution.

Industrial representatives long have argued that the Environmental Protection Agency (EPA) should do more to encourage regulated entities to evaluate their own compliance status and correct their own violations. While both EPA and the Department of Justice (DOJ) have policies in place that purport to encourage these practices, regulated entities have argued that some elements of these policies, particularly when taken together with other aspects of EPA's existing rules, actually tend to discourage efforts to develop comprehensive compliance-assurance programs. When EPA declined to take further steps, regulated entities turned to the legislative process hoping to find a more receptive ear.

And that is exactly what they found. In 1993, the Oregon legislature led the charge by enacting Senate Bill 912, a provision of which creates an evidentiary privilege for documents associated with environmental audits.(1) In 1994, Colorado,(2) Indiana,(3) and Kentucky(4) followed suit. This year, Illinois,(5) Arkansas,(6) and Wyoming(7) added their voices to the chorus. Additionally, in Virginia(8) and Utah(9) bills await the Governors' signatures.(10) The Colorado, Wyoming, and Virginia bills go beyond the typical idea of a privilege by providing regulated entities who promptly address discovered violations with immunity from state enforcement of the relevant violations under many circumstances.

At the federal level, Representative Hefley (R-Colo.) recently introduced an audit privilege bill in the House.(11) Perhaps not surprisingly, the Hefley bill also contains an immunity provision similar to that contained in the Colorado bill. On the Senate side, Senators Hatfield (R-Ore.) and Brown (R-Colo.) have also introduced similar legislation.(12)

EPA, having been caught off guard by the recent flurry of legislative activity, is rethinking its posture relating to environmental audits, and has proposed an interim policy on initiatives in this area.(13) It is the thesis of this Essay that privilege and immunity provisions such as those contained in the above-mentioned state and proposed federal bills are the wrong solution to the right problem. I hope that this Essay will play some role in convincing EPA to oppose federal environmental audit privilege/immunity legislation, to resist further efforts by states to develop privilege/immunity legislation in this area, and to urge those states that already have adopted legislation to reconsider their approaches.

At the same time, this Essay urges EPA to develop an alternative framework to provide the regulated community with greater incentives to implement comprehensive compliance-assurance programs (which, as discussed below, should include, but not be limited to, environmental audits). In order to encourage these programs, EPA should develop a framework that includes a firm commitment not to impose civil penalties in response to self-disclosed violations except in limited circumstances, and to reduce the severity of its enforcement response even where some enforcement is appropriate. Additionally, EPA should agree to take steps, upon request from participating regulated entities, to reduce those entities' exposure to citizen suits in appropriate circumstances.

Everyone agrees that compliance-assurance programs are a good thing. EPA and DOJ have recognized as much in their current auditing policies. Surveys indicate that large corporations apparently agree. One recent survey conducted by Investor Responsibility Research Center found that 85% of the 249 companies surveyed had established voluntary audit programs; the programs' average age was eight years.(14) A 1992 Arthur Andersen survey of 257 companies, which included 38 companies with revenues of less than $100 million, found that 59.2% had conducted compliance audits between 1989 and 1991.(15)

These surveys should not be taken as an indication that the incentives currently in place (the EPA and DOJ auditing policies, as well as the draft Corporate Sentencing Guidelines) strike the appropriate balance between the need to promote industry compliance-assurance activities and the need to reserve adequate enforcement discretion for responding to serious violations. Many small and mid-sized companies currently do not have any auditing programs in place. Further, there may be good reason to question the comprehensiveness and aggressiveness of many of the programs currently in existence. Any candid private sector lawyer would acknowledge as much.

The major problem with the existing policies is that they largely ignore two important issues. First, regulated entities (particularly permit holders) frequently are required to report any identified violations to EPA as a matter of law.(16) And, second, even where reporting is not legally mandatory, the current policies frequently demand voluntary" disclosure as a precondition to any favored treatment in the enforcement process.(17) In neither case is there assurance that even relatively minor violations will be subject to reduced enforcement responses. As a result, there is a realistic threat that companies performing audits will wind up facing enforcement actions for violations that EPA otherwise may never have discovered.(18)

EPA should do whatever it can to increase both the quantity and quality of industry's compliance-assurance activities, so long as in the process it does not seriously undermine either the need for deterrence or the public's right to important information regarding environmental matters. The privilege approach, however, is seriously flawed. In addition to increasing the cost of enforcement, as a practical matter a federal audit privilege would shield many "bad actors" from criminal enforcement. It would also shield some not-so-bad actors from civil enforcement in cases where an enforcement response clearly is justified. Additionally, a federal environmental audit privilege would impose a veil of secrecy over environmental compliance issues that would be inconsistent with both the government's and the public's rights to be informed. These points are elaborated below.

Immunity approaches pose even more serious issues. In this context, the shields referred to above would be absolute, not just a matter of practical difficulties. Criminal behavior would go unpunished and other serious violators would escape enforcement regardless of either the threats posed by their violations or the economic benefits they garnered through their noncompliance. Moreover, under the Colorado, Wyoming, and Virginia immunity provisions, all this would be achieved without any quid pro quo beyond the violators' having performed one audit, notified the regulators, and returned to compliance. None of these bills provide a mechanism for ensuring that regulated entities continue to implement reasonable compliance-assurance programs on an ongoing basis.

Fortunately, there is a better way. Both EPA and states can provide regulated entities with powerful incentives to implement compliance-assurance programs short of passing privilege/immunity provisions. Through changes in their enforcement response and penalty policies, EPA and the states can develop frameworks that provide for immunity-type protection in most circumstances, while still reserving the right to take action in circumstances where public policy compels some enforcement response. Significantly, this can be done in the open, with full disclosure rather than secrecy. Additionally, unlike the privilege or immunity approaches, the administrative approach could ensure that regulated entities maintain an ongoing commitment to their compliance-assurance programs.

I. WHY ENFORCEMENT IS IMPORTANT: THREE ENFORCEMENT IMPERATIVES EPA SHOULD NOT JEOPARDIZE IN THE NAME OF PROMOTING COMPLIANCE-ASSURANCE PROGRAMS

I start with the self-evident proposition that enforcement is important. Everyone will agree that enforcement is necessary to deter truly egregious behavior--that is, behavior that either intentionally violates the law or reflects gross negligence regarding legal requirements. This, of course, is the focus of EPA's criminal enforcement program.

But enforcement programs have significance that goes well beyond flagrant violators. Only a very small percentage of environmental violations result from either wanton disregard for the law or criminal negligence. Most violations result from a more simple lack of care--a lack of sufficient attention being paid to environmental matters.(19) Vigorous enforcement programs can and do have a dramatic impact on the amount of attention that regulated entities pay to environmental compliance matters. Consequently, they have significant effects on the total number of violations that occur.

Environmental violations--even those that do not involve egregious behavior--can have serious environmental impacts. In the Resource Conservation and Recovery Act (RCRA) context, for example, the failure of a surface impoundment operator to maintain adequate freeboard' can have serious ramifications irrespective of whether the operator intended to violate the law or exhibited gross negligence.(20) EPA's RCRA penalty policies long have recognized the need to provide extra incentives for compliance where significant harm can ensue. They have done this by making "potential for harm" the most prominent factor in determining the "gravity-based" (deterrence-based) portion of the penalty. EPA should strongly oppose any privilege/immunity formulations that would decrease the incentive for regulated entities to exert the extra level of diligence that is necessary to minimze the potential for violations that could result in significant environmental harm.

Another extremely important function of enforcement is to "level the playing field"--that is, to force violators to forfeit any economic benefits they enjoyed as a result of either delaying or avoiding required compliance costs. Here again, EPA's penalty policies are instructive. The RCRA penalty policy, for example, makes the recovery of economic benefit its highest priority by disallowing any adjustments to this part of the penalty based on a violator's good faith" efforts to comply. If EPA did not insist on full recoupment of economic benefit, violators could enjoy significant competitive advantages as compared with their industrial peers that spent the money necessary to comply with environmental requirements on a timely basis. EPA should ensure that this possibility remains precluded. Based on the above analysis, there are at least three enforcement imperatives EPA should not jeopardize (or allow to be jeopardized) in the name of promoting compliance-assurance programs. First, EPA must retain full authority to investigate and prosecute potential criminal behavior. There are "bad actors" out there, although they constitute only a very small proportion of the regulated universe. One reason they are so few in number is the threat of criminal enforcement. As the California District Attorneys Association has pointed out, most environmental crime is economically motivated.(21) To the extent that legislatures make it more difficult to prosecute environmental crimes through the enactment of privilege/immunity legislation, the net result very well may be a rise in the number of environmental crimes committed. At the very least, any such legislation would result in fewer prosecutions and convictions even where criminal behavior in fact has occurred. This cannot reflect sound enforcement policy.

Second, EPA should retain full authority to impose civil penalties in response to violations that pose a major potential for harm to human health or the environment. EPA must continue to reserve its most serious civil enforcement responses for violations posing the most serious threats of harm. Companies must continue to receive the message that they must exercise their utmost diligence in circumstances that could give rise to violations posing major threats.(22)

And third, EPA should continue to insist on recouping the full economic benefit that violators derive through their noncompliance. Fair is fair. Companies and other regulated entities simply should never be put at a competitive disadvantage by expending the funds necessary to comply with the law on a timely basis.

II. THE CREATION OF PRIVILEGES IS A SERIOUSLY FLAWED APPROACH TO THE PROBLEM AT HAND

There are at least four significant reasons why audit privileges are not the way to go.

1) Audit privileges necessarily increase the costs of environmental enforcement. Every time a privilege is asserted, which one must assume would happen frequently, the government would be required to go to court to establish either that the privilege does not apply or that the document is otherwise subject to disclosure. This would be true for all enforcement investigations, even those arising in what would otherwise be a purely administrative context. Thus, in civil enforcement cases, EPA would lose the efficiency advantages it traditionally has enjoyed when proceeding administratively.

Criminal cases would, of course, be subject to similar delays. Moreover, in the criminal context, EPA frequently needs information from not only the suspected wrongdoer, but also from other companies with which the wrongdoer may have been doing business. If these companies were to routinely assert privileges, many criminal investigations might die early deaths.

If the existing legislation is any indication, the likelihood of litigation and delay in both civil and criminal contexts is further exacerbated by the vagueness inherent in the definitions of what should constitute an 'environmental audit" or a privileged "environmental audit report." While at least six of the existing state bills (Oregon, Kentucky, Indiana, Illinois, Wyoming, and Arkansas) at least require that the audit be "comprehensive" (without indicating what that means) in order to qualify for protection, neither the Colorado nor the Virginia bill contains any similar requirement. The Hefley and Hatfield bills similarly contain no requirement of comprehensiveness in order for a relevant investigation to qualify as a "voluntary environmental self-evaluation." Under these latter approaches, virtually any information generated pursuant to any inquiry by anyone within a company regarding any compliance issue could arguably come within the privilege. Even more significantly, all of the relevant bills are unclear regarding the crucial issue of whether the privilege also extends to the underlying facts discovered during the audits and reflected in audit reports. The Colorado Illinois, and Virginia bills, as well as the Hefley and Hatfield bills, contain testimonial privileges that appear to reach underlying facts, but are unclear regarding whether they apply where there was preexisting awareness of the violations.

While Congress might be able to resolve some of these ambiguities through the legislative process (if it ever comes to that), at least the first issue--relating to what constitutes an adequate audit--would be better resolved through an administrative process. In the administrative context, EPA would have a better chance to fine-tune the definition of a qualifying audit in a way, for example, that might establish different levels of requirements in different factual contexts. Moreover, on the question of whether the privilege should reach underlying facts, the California District Attorneys Association is correct in noting that "[t]he distinction between discoverable facts and nondiscoverable, subjective conclusions is inherently blurred."(23) It may be impossible to effectively distinguish between the two in delineating the reasonable bounds of any privilege provided.

2) The net effect of audit privileges is that enforcement-related investigations become much more difficult in all cases where a privilege is asserted. In cases where the assertion of the privilege is upheld, cases that otherwise might constitute viable and appropriate enforcement actions may founder for want of evidence. Criminals may escape detection, the deterrence value of the law may be diminished, and those not complying may receive competitive advantages vis-a-vis their complying peers.

Even if a federal bill were to provide for some type of "compelling need" exception (as does the Hefley bill), this standard might be very difficult to meet in the preliminary stages of an investigation. Courts might require the government to show a reasonable basis to believe there is a violation before allowing access to audit reports. This might be impossible to do, again, particularly in the early stages of an enforcement case. In the criminal context, a court might very well require a showing of "probable cause," in addition to a compelling need. as a precondition to releasing the document.(24)

Moreover, as the California District Attorneys Association has pointed out, a privilege would cause severe problems with the "fruit of the poisonous tree" doctrine. Environmental enforcers (including civil enforcers) may routinely be required to have two attorneys on each case, one who handles the privilege proceeding and another to handle the case-in-chief, and may have to create a "Chinese wall" between the two to ensure they can show that no independent evidence is in fact poisoned "fruit."

3) The secrecy that would be inherent in the broad creation of an audit privilege is contrary to trends in both environmental law and civil and criminal discovery. As noted by the Atlantic States Legal Foundation and other environmental groups, the public's right to know about environmental threats "has become a benchmark of U.S. environmental policy."(25) To create a broad class of privileged information would have serious effects on the public's faith in EPA's ability and desire to adequately oversee environmental compliance. It also could have negative effects on industry's public image, with every assertion of the privilege being viewed as an indication that the company has something to hide.

The creation of a federal statutory privilege in this area would also run counter to general trends in discovery law. The clear trend is toward fewer, not more, discovery privileges. This is in accordance with the Supreme Court's edict that "the public . . . has a right to every man's evidence."(26) Elsewhere, the Court has noted in the same vein that privileges and other exclusionary rules "are not lightly created nor expansively construed, for they are in derogation of the search for truth."(27)

EPA should strongly oppose any erosion of these trends if there are other viable methods of accommodating the regulated community's concerns.

4) Finally, audit privileges do not even give the regulated community what it wants. An seven of the state bills enacted so far--as well as the Hefley and Hatfield bills--have clear provisions exempting information that is required to be reported under other laws. If we are to assume that any final federal bill will have a similar exception (as it certainly should!), the privilege will be of no avail in many cases. This is because, as mentioned above, EPA's permits typically require permittees to report all instances of noncompliance. Moreover, many other provisions of environmental law require reporting of discovered conditions that may pose environmental threats, which in many cases could lead EPA to investigate potential violations of law.(28)

The only alternative would be to create a broad-based exception to other environmental reporting provisions. Congress should not do this. These reporting provisions serve two very important functions. First, industry self-monitoring and reporting requirements are at the very heart of the compliance-assurance systerns contained in all of the major pollution control statutes. Neither EPA nor the states have the resources to perform the numbers of inspections that would be necessary to displace this role. Nor should they. It is simply much more efficient to enlist the regulated community's help in this regard. And second, the CERCLA(29)--type notification provisions--that is, those that require notice in the face of significant threats even absent a regulatory violation--are essential to ensure that these threats are identified and properly addressed. Both types of reporting provisions are too central to our pollution control statutes to even seriously consider subjecting them to such a broad-based exception as would be necessary to render a privilege fully effective.

III. IMMUNITY PROVISIONS SACRIFICE FAR TOO MUCH

Immunity provisions like those contained in the Colorado and Wyoming statutes, and the Virginia, Hefley, and Hatfield bills are even worse than privilege provisions. Generally speaking, all of these appear to generally preclude any enforcement action with respect to voluntarily disclosed violations, so long as the violation is corrected diligently.(30)

Under a broad immunity approach (as in the Virginia and Hatfield bills, and possibly the Hefley bill), even the most blatant criminal conduct could go unpunished. Even under the more narrow Colorado and Wyoming laws, acts of criminal negligence and numerous other non-intentional violations resulting in significant environmental harm could be immune from penalties. Additionally, these bills are completely oblivious to any notion of "leveling the playing field" between violators and their competitors who spent the money necessary to comply with the law on a timely basis.

Even putting these concerns aside, these immunity provisions simply do not require enough in exchange for waiving all past penalties. Essentially, the message is, "Come on in, comply with the legal requirements from now on, and well pretend away your past violations." The regulated entity provides no benefit to the regulators or the public beyond what is already required as a matter of law. Sound public policy demands that the violator provide some quid pro quo for the significant enforcement concessions that are made when regulators waive significant penalty amounts for past violations. At the very least, the violator should be made to resolve the past violations through an administrative order that establishes a compliance schedule (if compliance has not already been achieved) and requires the respondent to maintain an ongoing compliance-assurance program for a predetermined time period (perhaps five years).

IV. EPA Can Adequately Encourage Compliance-Assurance Programs Through Changes in Its Enforcement Response and Penalty Policies

A. A Proposed Program

EPA can provide the regulated community with sufficient incentives to undertake compliance-assurance programs without further federal or state legislation (and with the repeal of those state laws that already have been enacted). It can do this by making changes to its enforcement response and penalty policies to reflect the basic enforcement principles stated above. In short, such a scheme would include the following components:

(a) Any entity wishing to receive a favored enforcement response

based on its compliance-assurance activities would be required to

have implemented a comprehensive compliance-assurance program.

Such a program would include at least three

characteristics:

(i) a track record of having performed at least one recent, thorough

environmental audit designed to assess the facility's

compliance with all relevant federal, state, and local environmental

requirements;

(ii) timely disclosure to EPA of all discovered violations, regardless

of whether such disclosure otherwise was required as a

matter of law; and

(iii) a commitment to extend the program into the future, providing

for ongoing compliance-assurance activities on a regular

basis.

(b) If a facility has such a program and reports a violation, it should

be entitled to a favored enforcement response regardless of

whether disclosure was required as a matter of law. Such facilities

should be immune from punitive enforcement actions except

in three circumstances:

(i) Where the violation was the result of truly egregious behavior

(for example, knowing disregard of the law or gross negligence),

all bets should be off. The violator should be

subject to full criminal or civil enforcement within EPA's ordinary

enforcement discretion.

(ii) Where the violation was not the result of egregious behavior

but posed a major potential for harm to human health or the

environment, the violator should be subject to full enforcement

under EPA's civil penalty policies, but would be entitled

to a discount if its history of compliance-assurance

activities demonstrates that it has exerted good faith efforts

to comply.

(iii) Where neither (i) nor (ii) are present, but the violator did

receive a significant economic benefit through its noncompliance,

EPA should bring a civil penalty action to recoup 

the economic benefit. In this circumstance, all gravity-based

penalties should be waived.

(c) In cases where EPA forgoes significant gravity-based civil penalties

(perhaps more than $50,000) due to the violator's having implemented

a comprehensive compliance-assurance program, EPA

should be required to enter into a consent agreement with the

violator that legally binds the violator to extend its compliance-assurance

program for a predetermined period (again, perhaps

five years). In some cases, such as where compliance has not yet

been achieved or where EPA also is settling a penalty case pursuant

to either paragraph (b)(ii) or (b)(iii) above, negotiations toward

a consent order already may be ongoing and the penalty

waiver aspects can be folded into the preexisting negotiations. In

other cases, the penalty waiver and the corresponding commitment

to extend the compliance-assurance program may be the

only component of the negotiations.

(d) In cases where EPA is forgoing only minor penalties (however

defined), EPA may choose simply not to take any enforcement

action. However, in such cases the violator may seek to have the

agreement embodied in a consent order as a way to provide some

protection from citizen suits (see below). In such cases, EPA

should be forthcoming, with the quid pro quo again being a commitment

by the violator to extend its compliance-assurance

program.

B. The Proposed Program Should Be Enough To Encourage the Development of More and Better Compliance-Assurance Programs

The proposal outlined in this Essay should both encourage more companies to begin compliance-assurance programs and motivate companies already having such programs to improve and systemize their current programs. This is true because the proposal guarantees that, absent criminal behavior, companies will receive significant credit for having good programs in the penalty-imposition process. Indeed, in many cases (perhaps even most), companies will be able to have all penalties waived if they are willing to commit to implementing an ongoing compliance-assurance program.

One important feature of the proposal is that regulated entities should receive credit for having compliance-assurance programs regardless of whether they legally are bound to report any violations discovered in their audit processes. The lack of any such credit under the existing EPA and DOJ policies is a major impediment to the implementation of more and better audits. The key point is not whether any resulting disclosure is optional, but whether the underlying audit is optional. Because most companies are not required to implement self-auditing programs, we must provide them with incentives to do so. In my view, the most important incentive EPA could provide would be to ensure that regulated entities will receive credit for having implemented such programs even if they are required to report any identified violations.

Some industrial advocates may argue that the proposal is inadequate because it does not provide a shield when either environmental groups or toxic tort plaintiffs try to use the information generated by audits against the regulated entity. This concern is largely illusory for two reasons.

First, in most cases this proposal would result in a consent agreement between EPA (or a state, if a state also adopted this approach) and the relevant violator. The consent agreement specifically would indicate that penalties have been waived or reduced because the violator agreed to undertake actions (that is implementing the compliance-assurance program) that it was not otherwise required to undertake. The proposal centemplates that the violator always would be able to receive such treatment upon request.

In Gwaltney of Smithfield, Ltd. v. Chesapeake Bay Foundation, the Supreme Court determined that citizen suits are intended to supplement, not supplant, government action.(31) In so doing, the Court discussed a hypothetical, very much like the situation that would be present here, where EPA agrees not to seek civil penalties because the violator agrees to take corrective action that it otherwise is not obligated to take.(32) The Court indicated that a citizen suit in that context would be improper:

If citizens could file suit, months or years later, in order to seek the civil penalties

that the Administrator chose to forgo, then the Administrator's discretion

to enforce the Act in the public interest would be curtailed considerably....

[This] would change the nature of the citizens' role from interstitial to potentially

intrusive.(33)

Subsequent appellate decisions make clear that the lower courts have taken the Supreme Court's message to heart.34 Given this case law, it is very unlikely that most environmental groups would choose to file a citizen suit seeking further relief in the face of such an administrative agreement.

Secondly, in most cases the existence of a compliance assurance agreement will tend to reduce rather than increase the potential exposure of companies in any citizen or toxic tort suits. That is why most enlightened companies have such programs in place right now. The main problem under the current scheme has not been citizen suits or toxic tort actions. Indeed, my research has been unable to turn up a single case where an environmental audit was used to the defendant's disadvantage in either type of action.

To the extent that there has been a problem under the current scheme, it is because there has been essentially mandatory disclosure to EPA (or to the states, or both), with no established framework in place to allow sufficient credit to be given to regulated entities to recognize the benefits of compliance-assurance programs.

The remote threat that a regulated entity may someday be disadvantaged in one of these contexts can in no way justify the creation of an unprecedented statutory privilege. This is true particularly in the context of an administrative solution like the one proposed here, because this solution is both more workable and strikes a better balance between the need to promote compliance-assurance activities and the competing need to reserve a range of enforcement discretion to address serious violations.

V. CONCLUSION

In conclusion, EPA should take immediate steps to preempt the audit-privilege movement by providing an alternative scheme that better balances the competing goals of vigorous enforcement and encouraging compliance-assurance activities. If the Agency decides to pursue this route, it should take all steps at both the federal and state levels to promote such a strategy and to discourage the further development of privilege/immunity legislation.

(1) Or. Rev. Stat. [sections] 468.963 (1993). (2) Colo. Rev. Stat. Ann. [subsections] 13-25-126.5, 13-90-107(j)(I)(a), 25-1-114.5 (West Supp. 1994). (3) Ind. Code Ann. [subsections] 13-10-3-1 to -12 (Burns Supp. 1994). (4) Ky. Rev. Stat. Ann. [sections] 224.01-040 (Michie/Bobbs-Merrill Supp. 1994). (5) S.B. 1724, 88th Gen. Assembly, 1993-1994 Reg. Sess. (1993) (to be codified at Ill. Rev. Stat. ch. 5, para. 52.2 (Supp. 1995)) (on file with auchor). (6) Ark. Code Ann. [subsections] 8-1-301 to -312 (1995), available in Westlaw, General State Databases, Statutes - Annotated. (7) Wyo. Stat. [subsections] 35-11-1105, 35-11-1106 (1995), available in Westlaw, General State Databases, Statutes - Annotated. (8) H.B. 1845, 1994-1995 Reg. Sess. (1994) (on file with author). (9) S.J.R. 6, 51st Legislature (1995) (on file with author). (10) In addition, at the time of this article, bills on this topic have been introduced in 22 states: Arizona, California, Florida, Georgia, Hawaii, Idaho, Iowa, Maryland, Massachusetts, Mississippi, Missouri, Montana, Nebraska, New Hampshire, New Jersey, Ohio, Oklahoma, Rhode Island, South Carolina, Tennessee, Texas, West Virginia. (11) H.R. 1047, 104th Cong., 1st Sess. (1995). The bills has six co-sponsors: Rep. Allard (R-Colo.), Rep. Hyde (R-Ill.), Rep. Schaefer (R-Colo.), Rep. Crapo (R-Idaho), Rep. DeLay (R-Tex.) and Rep. Young (R-Alaska). Id. (12) S. 582, 104th Cong., 1st Sess. (1995). (13) Voluntary Environmental Self-Policing and Self-disclosure Interim Policy Statement, 60 Fed. Reg. 16,875 (Apr. 3, 1995). (14) Environmental Auditing Focus Group, Environmental Protection Agency, Measuring The Extent of Auditing Practices (1995) (prepared for the January 19, 1995 meeting held in San Francisco, Calif. (on file with author). (15) Id. (16) See, e.g., Environmental Protection Agency Regulations, 40 C.F.R. [sections] 70.6(a)(3)(iii) (1994) (Clean Air Act: State Operating Pennit Programs); [sections] 122.41(1) National Pollution Discharge Elimination System); [sections] 270.300) (Hazardous Waste Permit Program). (17) Draft Corporate Sentencing Guidelines for Environmental Violations, 24 Env't Rep. (BNA) 1378, 1382-83 (Nov. 16, 1993). (18) The flip side of this, of course, is that they may receive more favorable treatment than they would have if they had not done an audit and EPA did discover the violation. (19) Still others result from conditions that are largely beyond the control of the violator. Under the Clean Water Act (CWA), for example, technology-based limits are set at levels that even a well-operated facility will achieve only 99% of the time. Not much, of course, can be done to deter the rare violations that even the best run of CWA facilities occasionally will suffer. (20) See Interim Status Standards for Owners and Operators of Hazardous Waste Treatment, Storage, and Disposal Facilities, 40 C.F.R. [sections] 265.221(f) (1994) (regulatory requirement for "freeboard"). "Freeboard" refers to the distance between the top of the surface impoundment and the surface of the waste contained therein. Hazardous Waste Management System: General, 40 C.F.R. [sections] 260.10 (1994). (21) Environmental Protection Comm., California District Attorneys Ass'n, The Concept of Immunity for Self-Clinical Analyses and The Federal Statutory Environmental Audit Privilege 1 (Nov. 1994) (on file with author). (22) The line between violations that pose "major" versus "moderate" or "minor" threats to the environment is, of course, a difficult one to draw. But EPA has years of experience doing so on a case-by-case basis under its various penalty policies. In my judgment, the Agency has done a commendable job. Perhaps more importantly, there isn't any other way--EPA cannot predetermine in the abstract exactly which violations give rise to the most serious threats. The bottom line is that concern for the ultimate goals of our environmental programs--protection of human health and the environment--requires that companies be penalized when they put this protection at serious risk, regardless of whether EPA previously has announced an air-tight analytical construct defining the circumstances under which it will deem these threats to have occurred. (23) California District Attorneys Ass'n, supra note 21, at 16. (24) It is Worth noting that at least four of the state bills drafted to date (Oregon, Kentucky, Indiana, and Wyoming) provide for a "compelling need" exception only in the context of criminal enforcement. (25) Letter from Robert W. Shavelson, Director, Atlantic States Legal Foundation, Inc., to Geoff Garver, Director, Office of Enforcement & Compliance, Environmental Protection Agency 4 (Feb. 2, 1995) (on file with author). The letter is cosigned by representatives of Citizens for a Better Environment, the Environmental Defense Fund, the Public Citizen Litigation Group, the Western Environmental Law Center, Environmental Action, The Good Neighbor Project, and Trial Lawyers for Public Justice. Id. at 5. (26) United States v. Bryan, 339 U.S. 323, 331 (1950). (27) United Stales v. Nixon, 418 U.S. 683, 710 (1974). (28) See, e.g., Comprehensive Environmental Response, Compensation, and Liability Act of 1980, 42 U.S.C. [sections] 9603 (1988.) (29) 42 U.S.C. [subsection] 9601-9675 (1988 & Supp. V 1993). (30) The Colorado law does not preclude criminal enforcement for intentional violations. It also allows for all types of enforcement if the entity is found to have committed "serious violations that constitute a pattern of continuous or repeated violations of environmental laws ... and that were due to separate and distinct events giving rise to the violations." Colo. Rev. Stat. Ann. [section] 25-1-114.5(6) (West Supp. 1994). The Hefley bill appears to be similar to the Colorado bill, but is less clear regarding whether intentional crimes come within the immunity provision. (31) Gwaltney of Smithfield, Ltd. v. Chesapeake Bay Found., 484 U.S. 49, 60 (1987). (32) Id. at 60-61. (33) Id. at 61. (34) See, e.g., North & South Rivers Watershed Ass'n v. Town of Scituate, 949 F.2d 552, 555-56 (1st Cir. 1991); Supporters to Oppose Pollution, Inc. v. The Heritage Group, 973 F.2d 1320, 1324 (7th Cir. 1992).

Craig N. Johnston, Associate Professor of Law, Northwestern School of Law of Lewis & Clark College. This Essay was developed from a letter I submitted to EPA regarding environmental audit privileges on February 28, 1995. Since that time, EPA has published an interim policy on voluntary environmental self-policing and self-disclosure. 60 Fed. Reg. 16,875 (Apr. 3, 1995). The EPA policy is similar in approach to the positions advocated in this Essay, There are, however, important differences. Most notably, EPA's draft approach appears to provide for penalty mitigation only when the regulated entity voluntarily discloses any identified violations. As will be seen, I advocate that regulated entities who perform voluntary audits should receive credit in the enforcement process regardless of whether there is a mandatory reporting obligation. Because of this difference and others, Environmental Law has chosen to publish this Essay in its original form. I would like to thank law students David Voluck, Sonia Montalbano, and Mare Fink for their research assistance.