July 16, 2002
Mr. Chairman and Distinguished Members of the Committee:
It is a great pleasure to be testifying again before these two distinguished Committees in a rare joint session on the very important legal and policy issues raised by the administration of the "new source review" (NSR) provisions of the Clean Air Act.
EPA's many changing interpretations of NSR over the years have created a legal mess of baffling complexity that raises a host of separation of powers and administrative law issues that only a law professor could love. The good news is that the NSR controversy makes a great hypothetical for a law school exam (and I have used it as such at least twice in my administrative law courses at Yale and Georgetown). Unfortunately, the bad news, which is much more important, is that major parts of our country's economic infrastructure -- including but not limited to the electric power industry -- are now threatened with great legal uncertainties and huge penalties. As a result, as documented by EPA's recent NSR report, plants are delaying making needed repairs and changes to equipment. In the long run this threatens the reliability of our electricity supply and keeps inefficient equipment on line when it would benefit our economy to replace it with more modern equipment.
The ultimate solution in my view is to replace the antiquated, inefficient NSR program for existing plants with a modern trading system. But in the meantime, I applaud the Administration's recent attempt to do what it can to resolve the huge uncertainties about what is legal and what is illegal under the NSR program by creating safe harbors through the rulemaking process. It took great courage to touch the issue at all, because NSR is rapidly becoming the proverbial "third rail" of US environmental politics. Any action - no matter how modest and reasonable - will immediately be denounced as a rollback of historic proportions in an election year. It is very easy for us no longer in the political arena to criticize. I must admit that I was General Counsel of EPA at the time of the 7th Circuit's WEPCO decision in 1990, which helped to create the current NSR controversies. Urged on by majorities in both houses of Congress during the 1990 Clean Air Act Amendments to "fix the WEPCO problem," the first Bush Administration came out with an NSR interpretative rule in 1992, which I thought had resolved the WEPCO problem, at least for the electric utility industry. However, as a prelude to the current NSR enforcement initiative, the Clinton Administration attempted to renounce our interpretation of NSR -- without any notice and comment - by renouncing it in a proposed rule in 1998.
So I have to admit that I was unsuccessful in getting the problem resolved when I was in the government, so perhaps it is churlish of me to criticize others. But nevertheless, I do feel that the current Bush Administration did not go far enough in two ways. First, in my opinion, the safe harbor portions of the proposed NSR rule should have been made immediately effective as an "interim final" rule under the "good cause" provisions of the Administrative Procedure Act. Massive uncertainty has been created by vague caselaw (which is almost certainly wrongly-decided under more recent Supreme Court precedents ) and by EPA's misguided NSR enforcement initiative. While notice and comment is important, it is simply untenable to wait another 3-5 years or more for a resolution of this controversy in the courts and through the rulemaking process. Administrative law specifically recognizes the power of agencies to put rules into immediate effect for good cause in the meantime while taking comments. EPA has often used this power in the past when court decisions have created undesirable uncertainty, such as following the invalidation of the mixture-and-derived from rule under RCRA in 1991.
Secondly, I believe that the Administration should immediately conform its litigating position in the pending NSR enforcement cases to the policy position that the Administration has taken in the proposed rules. I disagree with my good friend Assistant Attorney General Thom Sansonetti that it is going to be viable for the U.S. government to pursue multi-billion dollar cases based on the premise that the same words in the law meant one thing in 1985, another thing in 1992, still another in 1996, yet another in 1998 and will someday mean something still different in the future.
I also disagree strongly with those who imply that the pending enforcement cases brought in a previous Administration should somehow disable a new Administration from implementing its views of good public policy. As I indicated in my testimony before Government Affairs last March, I see this as a fundamental Constitutional question of who is running the government - the President and the Officers of the United States confirmed by this Senate, or the career enforcement staff at EPA. Just as the Clinton Administration was free to walk away from the first Bush Administration's NSR interpretation in 1998 - provided of course that proper procedural formalities were observed -- so too the second Bush Administration should be free to reinstate its own NSR interpretations and policies.
Of course, the Congress can make the Administration pay a price politically for its actions. But, in my opinion, NSR is the wrong issue to make the touchstone for good environmental policy. The NSR program is the greatest failure in the administration of our environmental laws in my professional lifetime. It has failed to work for 25 years, and now it badly needs to be replaced with something that does work. Case-by-case, plant-by-plant litigation to force individual plants to install best available control technology is at best an antiquated regulatory technology. It is slow, expensive and uncertain. There has to be a better way. The better way is clear. It is a modern, efficient cap and trade system - a concept that has proven remarkably successful in the Acid Rain Trading system under the 1990 Amendments, and which now has tri-partisan support in both the Administration's "Clear Skies Initiative" and Senator Jeffords' S.556, which was recently reported out by this Committee. A modern, efficient trading system will achieve far greater pollution reductions in far less time and at far less expense - not to mention the side-benefit of putting lots of pesky environmental lawyers out of business! In my view, we should move promptly on a tri-partisan basis as quickly as possible to replace the antiquated, dysfunctional NSR system for existing plants by legislating a modern efficient trading system.
What has caused the great NSR debacle? There is plenty of blame to go around - and I probably share some of it. I had been nominated as General Counsel of EPA but not yet confirmed when the WEPCO case was argued, and I failed to properly supervise my staff and did not know the position on NSR issues that EPA was advocating until after the decision came down. The courts are partially to blame, because the cases to date have temporized by promulgating vague, multi-factor tests that fail to give clear guidance to industry as to what is and is not permitted. EPA has issued multiple and inconsistent interpretations over the years. There have been so many of them that I doubt that any of them will ultimately receive much deference from the courts. When invited by the Administration to review the legal situation, the Department of Justice in its recent report ducked the key issues, and said merely that EPA's latest interpretation was not so clearly wrong that it would be unethical to continue to argue it. DOJ then ducked entirely the key issue of whether industry had been given fair notice of EPA's newest interpretation of NSR requirements, punting that central issue entirely to the courts. Unless Congress steps in, I fear that we are now embarked on a decade-long process of litigation that will require several Supreme Court decisions to clarify the law.
Much of the blame for the current NSR mess must also be laid squarely at the doorstep of Congress. In the text of the 1970 Clean Air Act, Congress created a basic distinction between the pollution requirements applicable to "new" as opposed to existing plants. Then, in so-called "technical amendments" in 1977 -- which were never debated or properly vetted in Committee hearings -- Congress extended the concept of "new" plants to included "modifications" of existing plants. But in its wisdom, Congress failed to adequately define the key operative concept of a "modification" in the statute. That statutory ambiguity over how to define the nature of the "modifications" that convert an existing plant into the equivalent of a "new" plant for purposes of installing state-of-the-art pollution controls has been at the root of a great deal of unproductive and unnecessary NSR legal controversy over the years.
It simply cannot be that literally any modification - no matter how trivial - triggers NSR and converts every existing plant into a new plant. Somehow someone has to distinguish between those physical and operational changes that trigger new source review and those that don't. To date, it has proved impossible for the legal system to come up with any clear dividing line that will stand the test of time. EPA has repeatedly tried to resolve this controversy through a variety of changing rules and interpretations over the years. By rule, EPA has exempted certain activities such as "routine repair and replacement of equipment," and required an increase in emissions for a modification to trigger NSR. But how to define what repairs and replacements of equipment are "routine"? At one point, EPA even officially defined "routine" in the Federal Register as "what is routine in the industry" - which brings to mind Churchill's line about a question wrapped in a riddle wrapped in an enigma. EPA's legal staff also developed the fascinating theory of "potential emissions," so that a plant was considered to have "increased" its emissions even though its actual emissions went down! Over the years, EPA has come out with many shifting interpretations of what constitutes a "routine" repair and replacement, and now in its enforcement cases, EPA is arguing for yet a different definition than the ones that it advanced in the past or the ones that it is now proposing to implement through the rulemaking process. I do applaud the Administration's courageous attempt to bring some clarity to the legal chaos that is the NSR program today through its proposed safe harbor rule. There have already been so many varied and shifting interpretations by EPA in the past, however, that I seriously doubt that the courts will ultimately give much deference to whatever construction EPA now attempts to place on the statutory terms.
That unpleasant fact leaves us with only two real options going forward - either slug it out in many more years of unproductive litigation, probably going to the Supreme Court several times, before we finally find out what the term "modification" really means in the NSR provisions of the Clean Air Act. Or alternatively, as I prefer, Congress should act to put a merciful end to the NSR controversy by legislating a modern, more efficient replacement, such as the trading system advocated by the Administration in its Clear Skies Initiative and also endorsed in Senator Jeffords' proposal.
Realistically, I don't think there can be any serious question that slugging it out in continuing litigation is bad environmental policy that will really only benefit the lawyers -- and law professors - and maybe a few politicians who can claim to be taking decisive action to fight polluters, if not actually to benefit the environment. NSR litigation makes those who participate in it feel good, because they can imagine that they are taking tough action to benefit the environment. But in reality, the NSR approach of case-by-case litigation to force each individual plant to install best available control technology is not going to produce anything approaching the environmental benefits that will come from legislating a trading system to replace the antiquated and dysfunctional NSR program for existing sources.