Can Site-Specific Pollution Control Plans Furnish an Alternative to the Current Regulatory System and a Bridge to a New One?

William F. Pedersen Jr.

Mr. Pedersen is a partner in the law firm of Shaw, Pittman, Potts & Trowbridge in Washington, D.C.

The Republican takeover of Congress has triggered—and promises to continue triggering—a proliferation of suggestions from all political sectors for reforming our environmental regulatory system. So far, media attention has focused almost exclusively on generic proposals to require agencies to support new regulations with cost-benefit analysis and more "realistic" risk evaluation. Meanwhile, a completely different and equally broad-based approach has largely escaped notice.

Our current environmental protection system works largely by imposing detailed regulatory commands on "major sources" of air, water, and waste releases. Quite frequently, 5 to 10 different federal and state laws, and scores of separate, uncoordinated regulatory requirements, will apply to a large factory. A new approach has been proposed to replace this system. Under it, a major pollution source could enter into a legally enforceable contract with its local community promising both to cap its environmental releases and to reduce releases gradually over time. That contract would replace most of the regulatory commands in our current federal laws.¹

To date, no one has analyzed the relationship between such proposals and a more general environmental regulatory reform program. Their proponents have defended them simply as ways to strip the surplus procedural complexity out of the current system without any more basic change. Indeed, the Clinton Administration would make acceptance of all the substantive provisions of current law a condition to any such procedural relief.²

This Dialogue argues that the cap proposals will not lead to basic reform unless their designers recognize that our current system not only controls major stationary sources inefficiently, but also devotes too large a share of its energy to controlling them. The current system thus diverts attention away from the smaller "area" sources that currently cause far more environmental problems. A cap approach that simply transposed existing major-source control requirements into a more flexible format would only perpetuate that problem. For that reason, existing releases should provide the baseline for any cap, unless they conflict with achieving specific air or water quality standards. Reductions beyond that level should be addressed by a plant-wide "pollution prevention" plan worked out between the source and the local community unconstrained by federal requirements.

To support its position, this Dialogue describes some of the major defects in our current regulatory system. It then discusses why those defects support a more flexible approach to designing an emissions cap.

The Need for Reform and the Perils of Reform

The Need for Reform

Detailed government regulations telling major pollution sources exactly what to do lie at the center of our current environmental protection system. That approach, despite its undoubted accomplishments over the last quarter-century, now suffers from three overlapping and increasingly serious defects: Complexity, unjustified requirements, and—most important of all—focus on the wrong problem.

[] Complexity. As noted above, scores of regulatory requirements under the major state and federal environmental laws commonly govern major industrial plants. Most typically, the designers of those regulations did not consult much with each [25 ELR 10487] other on how the regulations would work together. Even if they had, it is beyond human capacity to foresee how such a complicated web of commands will work—or fail to work—in the context of any specific facility.

The need to keep track of these requirements and integrate them with each other already imposes a major paperwork and flexibility burden on plant managers and their companies.³ That paperwork must be updated whenever a new regulation comes out and whenever the plant operations themselves change significantly.

The complexity of these regulations often bars plants from achieving emission reductions in the most effective way. To cite one notorious example:

Under the current regulatory framework, Amoco's Yorktown, Va. refinery is required to reduce about 7300 tons of airborne hydrocarbon emissions per year at a cost of $ 2400 per ton. The refinery found that by using different control strategies from those required by the regulations . . . it could eliminate 7500 tons of both hydrocarbons and listed hazardous waste each year at an average cost of $ 500 per ton.

EPA does not have the statutory authority to approve the Yorktown alternative or other multi-media pollution reduction strategies like it.⁴

$=I

The Yorktown example illustrates a basic and increasingly recognized truth. Because major sources have been regulated intensely for over a generation, under the law of diminishing returns the ability of such a blunt instrument as regulation to achieve further reductions in their total releases is decreasing steadily.

Instead, further release reductions from such sources will far more likely be achieved by "pollution prevention" approaches in which the company itself examines its operations to make them less polluting. But to do that, the company's engineers must have freedom to investigate and produce reductions where they are available, not where the government says they should be.

[] Regulations Without Defensible Goals. The inability of our current regulatory system to achieve its goals effectively may be the least of its problems. Very often, those goals themselves are not worth achieving—at least not in the way the law frames them—as measured by any reasonable cost-benefit test. Generally, our current system's regulations are designed without any direct link to any particular environmental goal, such as clean air or clean water. Instead, they simply direct the sources they address to install "control technology," as defined under some legislative test, on the releases that they describe and to monitor the exact composition and amount of those releases in detail.⁵

Ten years ago, two respected law professors expressed the general academic consensus when they wrote that such requirements "waste many billions of dollars annually by ignoring variations among plants and industries in the cost of reducing pollution and by ignoring geographic variations in pollution effects."⁶ In the past 10 years, that problem has only gotten worse. The 1990 Clean Air Act Amendments require a multibillion dollar control program for "hazardous air pollutants" emitted from major stationary sources that is completely unsupported by any cost-benefit analysis. A Harvard School of Public Health study quoted in the legislative history of these amendments states that because of its lack of health benefits, this program could not appropriately be regarded as a public health measure.⁷

In another study, 8 of the 10 most cost-ineffective regulations in a study of Five Hundred Life-Saving Interventions and their Cost-Effectiveness imposed controls on major source emissions of hazardous air pollutants. The costs ranged from half a billion dollars to a hundred billion dollars per year of life prolonged.⁸

[] Mistargeting of Control Efforts. Due in part to a quarter-century of regulation, releases from major stationary sources in most areas now make up a relatively small and diminishing share of total environmental releases. Water discharges now come largely from "nonpoint sources"—roads, farms, and construction sites—and not from factories.⁹ Only 20 percent of emissions of hazardous air pollutants comes from factories.¹⁰ The same low figure holds true for the volatile organic compounds that cause ozone,¹¹ while road salt, fertilizer, and [25 ELR 10488] pesticide use — not factory emissions—are the major sources of groundwater pollution.¹²

Our current system's focus on the increasingly tangled intricacies of major-source regulation robs it of the economic and intellectual resources it would need to address the controversial and politically difficult questions of minor-source and land-use control. A regulatory system consumed with issuing and updating regulations for major sources of air or water pollutants simply has no time or incentive to realize that these releases only make up 20 percent of the total, or to ask itself how big a problem these releases may be in the first place. But without an ability to refocus on the real questions, our regulatory system will be forever unable to bring about true environmental progress.

The Perils of Reform

For many years, the old system seemed impervious to criticism, as each new law added ever more elaborate branches to the regulatory thicket. But the 1994 election revealed that the old system was actually ready to collapse of its own weight. Since then, the old system has found almost no defenders, as politicians of all stripes have rushed to embrace or appear to embrace the need for change.

To date, the most prominent reform suggestions have been generic, focusing on requiring all new regulations to pass a cost-benefit test and make use of more realistic risk assessments. Some of the competing bills also provide for reexamining old regulations under these new standards.¹³

Environmental groups, and to a significant degree the Clinton Administration, have steadfastly opposed any meaningful action to legislatively compel increased reliance on cost-benefit analysis and realistic risk assessments. Industry, on the other hand, almost unanimously supports that approach. But for two reasons many industry groups, at least in a corner of their minds, also view those suggestions as unpredictable, slightly risky, and certainly falling far short of the full change required.

First, such proposals, if enacted, would have only a marginal effect in their first few years. Their impact on any individual decision would be uncertain. And only over decades, as regulations "turned over," would it be even conceptually possible to change the whole regulatory system simply by addressing new regulations or amending old ones.

Second, these approaches by themselves can be, and are, attacked as camouflage for a wholesale "roll-back" of reductions in pollution releases that have already been achieved. In fact, few if any industries desire this roll-back. Most major companies have internalized pollution control as a standard operating procedure.¹⁴ In addition, reductions that have already been achieved represent a "sunk cost" that is generally not worth undoing. Finally, most industries recognize that the public will not accept a system that allows, or appears to allow, actual increases in pollution from current levels.

The Emission-Cap Proposals

Major areas of agreement already seem to exist among parties to the emission-cap discussion, but so do major areas of clear disagreement and unresolved complexity. To the extent disagreement exists, a workable emission-cap agreement that truly responds to the need for change must reflect more the industry than the environmentalist side of this discussion.

The Area of Agreement on Emissions Caps

All the emission-cap proposals would provide an immediate exit from the current regulatory system for selected major sources,¹⁵ replacing it with contracts tailored to the sources' particular circumstances. The contracts would last for a significant period, probably 10 years, and could be renewed thereafter through the same process used to establish them. During that time, the government could override contract terms only in cases of clear environmental necessity. In return for this stability and flexibility, sources would undertake release-reduction obligations.

The emission-cap proposals agree that because this approach rests on a contract between an individual source and its local community, it will be voluntary by its very nature. They further agree that the cap program must be "transparent" and accountable. Maintaining public confidence in industry control efforts during the shift to a new environmental protection system is at the heart of these proposals. Both for that reason, and simply to make sure that this innovative new approach actually works, it is essential:

. that the local community be a party to any cap, a requirement that would provide both political acceptability and a check on abuse of the cap approach;

. that all caps be established through full notice-and-comment proceedings, based on full disclosure of their underlying assumptions and rules;

. that reliable quantification measures be put in place to make sure that releases in fact do not increase and that progress in pollution prevention can be measured; and

. that violation of the cap's terms be accountable both through government enforcement action and through citizen suits.

Baseline and Flexibility: The Areas of Debate

[] Baseline. How to set the "baseline" — that is, the cap's release levels that can never be exceeded — presents the most basic unresolved question in the cap debate. As noted,¹⁶ the Clinton Administration proposals would require compliance with all "existing and reasonably foreseeable" regulatory requirements [25 ELR 10489] as the price of admission. No counterposition has yet been defined.

The chances of success for both the cap proposals and for environmental regulatory reform in general would be increased by a far less aggressive baseline position under which only emission reductions actually achieved were incorporated in the "never to be exceeded" limit.¹⁷ For reasons of basic acceptability, a cap approach should preserve gains in pollution control actually achieved to date. However, only release reductions that have been reached in physical reality represent such actual achievements. As discussed above,¹⁸ the existing regulatory system is so full of irrational and excessive control requirements, particularly where major sources are concerned, that its prospective commands should not receive automatic validity. To grant them such validity would use the cap approach to defeat regulatory reform by forbidding reexamination of unjustified existing control requirements by those who wanted to use the cap. That in turn would greatly reduce both the chances that the cap approach would succeed and the chances that the existing control system would be reformed. It would also perpetuate the current focus of our regulatory system on major sources, which damages environmental protection by diverting resources from more pressing problems.

In all probability, not even the cause of future release reductions would be damaged by relatively lenient baseline conditions. "Capped" sources would have to adopt pollution prevention programs to reduce their emissions over time. The tighter and more detailed the baseline specifications, the fewer caps there will be, and the less room existing caps will allow for pollution prevention. To the extent pollution prevention is more effective than insistence on the letter of existing regulations, a wider use of caps stemming from more lenient entry conditions will lead in the end to greater overall emission reductions. Put another way, it would be a mistake to damage the chances that this new approach will succeed by making the conditions for using it too stringent.

[] Flexibility. Under all the cap proposals, sources would remain subject to regulatory obligations expressly designed to achieve ecologically protective measures of air, water, or soil quality in their local areas.¹⁹ In addition, all capped sources would enter into an ongoing "beyond compliance" commitment to reduce releases beyond the level specified by existing regulations.

It is appropriate to insist in this manner that as companies get more efficient over time, environmental releases reflect those productivity gains. But once ambient air, water, and soil are protected, broad discretion in reducing releases is appropriate on both policy and practical grounds. If no clear link to air, water, or soil quality can be shown, what is the justification for severe or tightly articulated mandatory reduction requirements? And if we are to rely on private knowledge of production technology to achieve reductions, it only makes sense to give that knowledge freedom to operate.

For these reasons, sources and their communities should have the maximum freedom to define the scope and timing of "beyond compliance" reductions. The new cap law itself should require no more than steady progress in reducing environmental releases per unit of product. Sources should be allowed to do this by reducing their own releases, by reducing the pollution that arises from using their product, by on-site recycling, or by "take back" or "product stewardship" programs to promote product recycling.

Pollution per unit of product is the proper measure of industrial progress in "ecological efficiency," because it is not affected by changes in plant activity due to the business cycle or market ups and downs. In very rare cases, where expanding production at a site offsets the per unit decline in releases, releases at an individual site might not go down over time. But even there, releases would not go up, and the environment would benefit overall from the reduction in emissions at other sites where that expanding production would have otherwise occurred.²⁰

Areas of Complexity: Monitoring and Emissions Trading

Any cap proposal would also need to address the proper role of monitoring requirements and emissions trading in its overall design. Although these are not areas where the positions of different interest groups are readily predictable, they are areas whose complexity could imperil the drafting of any cap legislation.

[] Monitoring. As our current approach focuses on smaller and smaller amounts of pollution at lower and lower concentrations, and distinguishes more minutely between different types of pollutants, monitoring costs tend to become an ever greater fraction of total pollution control costs. Given the increasingly small amounts of pollutants involved, and our current inability to distinguish solidly between the different risks posed by different pollutants, it is hard to see how this development can be justified, particularly since it leads to an inflexible control system so full of regulatory commands that any alternative approach is almost bound to violate one or more of them.

At the same time, experience to date with pollution prevention shows that nothing is more important than measuring performance and progress.²¹ That measurement, of course, must be a function of the exact approach to pollution prevention that has been selected.

These facts probably support an "existing releases" approach to designing the cap. The monitoring approaches used to verify compliance with release levels already achieved will generally be better understood and less burdensome than those that accompany future regulations. To the extent future release reductions rest on pollution prevention approaches in which raw materials are substituted and processes are redesigned, they will probably be easier to monitor than reductions that rely on "end of the pipe" technology. Even where this is not the [25 ELR 10490] case, the contractual approach to determining beyond-compliance reductions will allow sources and communities to fit the monitoring methods to the design of the reduction obligation more cost effectively than the current regulatory system generally permits.

[] The Role of Emissions Trading. Many cap proposals go beyond simply requiring a cap on all releases at a source, followed by further reductions, to allow even more flexibility in beyond-compliance controls. Some proposals allow adjustments to the baseline by "trading" control obligations, so that, for example, a reduction in air emissions could be offset by an increase in water discharges, as long as no violation of water quality standards resulted. Other proposals would allow sources to satisfy their beyond-compliance obligations by obtaining release reductions off-site from related or unrelated sources.

Such approaches may well be worth including both in a cap program and in an individual cap. They do not, however, address the defects of our current system nearly as directly as the issues of baseline and flexibility. In addition, the design of such sophisticated approaches must confront substantial new issues. In trading one form of pollution for another, how do we set a "trading ratio" between, for example, releases to air and releases to water?²² In trading releases from different sources, what additional constraints — over and above those applicable to a single source — should be set to make sure that the reductions are equal and comparable? Given the benefits to be expected from an unadorned cap approach, it would be a mistake to make the establishment of the cap framework hostage to such questions.

Conclusion

We cannot know at this stage how many caps would actually be established should Congress approve this new approach. Given the large investment in designing and negotiating a creative and enforceable contract that establishing such caps might require, the number of caps might be small. But even in that case, a cap approach could serve three important larger purposes. First, it would provide a new model for regulating releases from industrial sources that might well have an impact, as such models do, on the government's approach to many facilities that did not formally qualify for cap treatment. Beyond that, the debate over adoption of a cap approach would raise two issues central to the overall design of a new environmental protection system — the current system's excessive focus on major stationary sources and the appropriate way to reduce emissions from major sources.

To realize its true reform potential, a cap approach must make clear that our current system focuses too much on major stationary sources when it should be focusing on smaller activities that are politically less vulnerable, but environmentally more important. The cap approach would respond by allowing major stationary sources to remove themselves from the regulatory system's central focus, thus liberating it to address more generic issues.

Even if major sources have become less important contributors to environmental problems, reducing their overall releases over time remains an important public goal. The cap approach acknowledges that this goal is now better served by far greater reliance on a voluntary approach, without the mandatory short-term deadlines and exact control commands that have marked our efforts in the past. By raising these two new "design specifications" for discussion, the cap approach might both provide specific regulatory relief and help pave the way for a redesign of our entire system.

1. Although no legislation has been introduced, draft proposals along these lines have been put forward by the Clinton Administration, the National Academy of Public Administration, and the Harvard Group on Risk Management Reform. See PRESIDENT WILLIAM J. CLINTON & VICE PRESIDENT ALBERT GORE JR., REINVENTING ENVIRONMENTAL REGULATION (1995) (especially suggestion 19 on the "XL" program); 60 Fed. Reg. 27282 (May 23, 1995); NATIONAL ACADEMY OF PUB. ADMIN., SETTING PRIORITIES, GETTING RESULTS—A NEW DIRECTION FOR THE ENVIRONMENTAL PROTECTION AGENCY 31, 37 (1995); HARVARD GROUP ON RISK MANAGEMENT REFORM, REFORM OF RISK REGULATION: ACHIEVING MORE PROTECTION AT LESS COST 37-42 (1995). On the industry side, draft bills incorporating similar suggestions have been circulated by General Electric, 3M, and several other companies.

2. Specifically, any project eligible for the XL program must be "able to achieve environmental performance that is superior to what would be achieved through compliance with current and reasonably anticipated future regulation." 60 Fed. Reg. at 27287.

3. The environmental vice president of International Paper Company gives a standard briefing with a slide that shows separate boxes labeled "Toxic Release," "MACT," "SPPC," "Dredge and Fill," and many other acronyms descending into a funnel hanging over a man standing by a factory. He tells his audience that no one in our society other than a plant manager is expected to make sense of all these different requirements by makingthem work together.

4. NATIONAL ACADEMY OF PUB. ADMIN., supra note 1, at 31.

5. Such tests come in a great variety of legislative descriptions, although they all have the common feature of forbidding to a greater or lesser degree the basing of regulations on a rigorous analysis of costs and benefits. They include provisions for treating water discharges to levels defined by use of the "best practicable control technology," "best available technology," and "best conventional technology" under the Clean Water Act, for treating hazardous wastes to "best demonstrated available technology" levels before land disposal under the Resource Conservation and Recovery Act, and for installing controls under the Clean Air Act to achieve control levels defined by use of "maximum available control technology" or "best available control technology" or to attain the "lowest achievable emissions rate." See 33 U.S.C. § 1311, ELR STAT. FWPCA § 301; 42 U.S.C. § 6924, ELR STAT. RCRA § 3004; 42 U.S.C. §§ 7412, 7475, 7503, ELR STAT. CAA §§ 112, 165, 173.

6. Bruce A. Ackerman & Richard B. Stewart, Reforming Environmental Law, 37 STAN. L. REV. 1333, 1335 (1985); see also BRUCE A. ACKERMAN & WILLIAM T. HASSLER, CLEAN COAL/DIRTY AIR (1981). Professor Ackerman is a liberal Democrat; Professor Stewart is a moderate Republican.

7. H.R. REP. NO. 490, 101st Cong., 2d Sess. 318 (1990).

8. TAMMY O. TENGS ET AL., FIVE HUNDRED LIFE-SAVING INTERVENTIONS AND THEIR COST-EFFECTIVENESS, tbl. 8, at 23 (1994).

9. OFFICE OF WATER PROGRAM OPERATIONS, U.S. EPA, REPORT TO CONGRESS: NONPOINT SOURCE POLLUTION IN THE UNITED STATES 1-14 (1984) (70 percent of chemical oxygen demand, 66 percent of phosphorous, 90 percent of kjeldahl nitrogen, 70 percent of oil, and 57 percent of lead in water attributable to nonpoint sources). By contrast, the only pollutant listed as more than 60 percent attributable to point sources were mercury, arsenic, and cadmium. Although some toxic heavy metals are largely attributable to point sources, no statistics are available for toxic organic chemicals. For sediments, the contribution of nonpoint sources is even greater. LEONARD GIANESSI ET AL., NONPOINT POLLUTION: ARE CROPLAND CONTROLS THE ANSWER?, tbl. 1 (Draft, Oct 1, 1985) (available from Resources for the Future, Washington, D.C.) (4 million tons per year of sediment runoff from point sources, 2.719 billion tons from nonpoint sources).

10. See H.R. REB. NO. 490, supra note 7, at 317.

11. In 12 northeastern states, major sources contributed 16 percent of these emissions in 1990, and are projected to contribute 14 percent in 2005. In Los Angeles, the figures are 19 percent for 1987 and a projected 21 percent for the year 2010. Richard E. Ayres, Developing a Market in Emission Credits Incrementally: An 'Open Market' Paradigm for Market-Based Pollution Control, 25 Env't Rep. (BNA) 1522, 1528-29 (Dec. 2, 1994).

12. A.D. TARLOCK, LAW OF WATER RIGHTS AND RESOURCES § 2.03[2] (1988).

13. The major bills are H.R. 9, which has passed the House of Representatives as part of the "Contract With America," and its Senate counterparts, S. 291, S. 333, and S. 334. See H.R. 9, 104th Cong., 1st Sess. (1995); S. 291, 104th Cong., 1st Sess. (1995); S. 333, 104th Cong., 1st Sess. (1995); S. 334, 104th Cong., 1st Sess. (1995).

14. Many major American companies have their own internal pollution control codes, largely reflecting developed country standards, for use by their plants in countries without strong environmental protection programs.

15. There is no reason that smaller sources, or groups of smaller sources, could not also be eligible for a cap. But the complexity of the issues a cap would need to address would probably effectively restrict its use to major sources, at least initially.

16. See supra note 2 and accompanying text.

17. If a source was out of compliance with its existing emissions limits, it should be ineligible for a cap until it had achieved compliance.

18. See supra note 6 and accompanying text.

19. Even for such obligations, the cap approach should allow more flexibility in methods and timing than our current system.

20. Since production increases in the economy as a whole are never greater than 5 percent per year, any decrease in pollution per unit of product that exceeded 5 percent would be bound to reduce overall national releases, even if releases at the particular site did not decline.

21. See Marcia E. Williams, Pollution Prevention: Industrial Leadership and the Lawyer's Role, A.B.A. SECTION ON NATURAL RESOURCES, ENERGY, AND ENVIRONMENTAL LAW, 24TH ANN. CONF. ON ENVTL. L., tab 28 (1995).

22. This problem might be addressed by allowing trades only as long as they reduced the releases to the more environmentally sensitive medium. For example, air decreases could be traded for water or waste increases, water decreases could be traded for waste increases, and waste increases could not be traded for anything.