Economic Efficiency in Pollution Control: EPA Issues "Bubble" Policy for Existing Sources Under Clean Air Act

The Environmental Protection Agency (EPA) has been the target in recent years of growing criticism that its pollution control requirements are in many instances unnecessarily rigid and excessively expensive. These pressures, in combination with EPA's developing interest in economically based alternatives to traditional regulatory approaches to pollution abatement,¹ have led the Agency to take several steps toward a regime that will, in its view, reconcile improved environmental quality with economic growth at the lowest possible cost. In this vein, the Agency is now making a major effort to encourage the development and use of alternative emission reduction strategies which are less costly and more flexible than current regulatory approaches. The basic aim of all these measures is to allow sources in certain circumstances the freedom to adopt more economically efficient pollution controls.

As a central component of this effort, EPA recently promulgated² its final policy on the use of the "bubble" concept by existing sources to abate air pollution emissions. The concept is so named because it treats a source with multiple emission points as if an imaginary bubble with a single vent had been placed over it. The focal point for regulation under this approach is the aggregate of all emissions from the source. Provided that increases and decreases in emissions from plant components offset one another, theoperator would be free to adopt a different set of controls on stacks and other emission points as long as the alternative control strategy does not delay compliance with statutory deadlines or have an adverse impact on air quality. Emission reduction thus may be lessened on those plant components that are most expensive to control and correspondingly maximized on those emission points that are least costly to clean up.

The final policy statement recommends that states consider revisions in their implementation plans to allow existing sources to use bubble-based alternative pollution control strategies. The Agency expects the new policy to result in significant cost savings by allowing individual operators to develop and implement what they feel would be the most cost-effective method of compliance. As an example of the dramatic cost savings which it believes use [10 ELR 10015] of the bubble approach can yield, EPA points to an estimate by DuPont that by using alternative control strategies it can reduce its air pollution control costs from $135 million to $55 million.³ EPA also believe adoption of the bubble will provide an incentive for sources to develop innovative control technologies.

The Agency is plainly excited about the possibilities of its new policy, which it sees as one in a series of regulatory reforms aimed at using economic incentives to enlist industry's cooperation in controlling pollution.⁴ Other EPA initiatives which similarly use market mechanisms to permit the "controlled trading" of emissions are the Agency's emissions offset and banking policies⁵ regarding new or modified source construction in non-attainment areas. In announcing issuance of the final bubble policy, however, EPA Administrator Douglas Costle labeled it "the most important single reform of the regulatory process" accomplished in the last several years.⁶ The new policy also provides EPA with a concrete defense against the recurrent charges of over-regulation and calls for drastic cutbacks in the Agency's statutory authority to control pollution.

Although in theory the new policy seems environmentally unobjectionable, environmental advocates fear that in practice use of the bubble will often result in compliance date extensions, serious enforcement difficulties for EPA and state agencies, and actual deterioration in environmental quality. Industry, while generally supportive of the new policy, shares the worries of many states that the substantial limitations on the use of the bubble are overly restrictive and will discourage its widespread adoption. Although there may be some lingering doubt as to the legality of adopting the bubble approach in regulating existing sources,⁷ more important questions now are (1) whether the conditions attached to the policy are sufficient to assure that compliance with national ambient air quality standards will not be delayed, (2) that the revised state implementation plan (SIP) provisions remain enforceable, and (3) whether, given these conditions, the bubble approach will in practice be both workable and widely available.

Background

The bubble concept was first put into effect in EPA's 1975 regulations⁸ regarding the application of technology-based new source performance standards (NSPS) to modifications at existing plants.⁹ The regulations provided that a source planning a physical or operational change that would entail an emissions increase could escape application of the strict NSPS by more than offsetting this increase with reductions in emissions of the same pollutant from other components of the plant. This was accomplished by broadly defining the term "stationary source" as "any one or a combination of … facilities"¹⁰ and stating that a modification subject to the NSPS would be deemed to have occurred only if the total emissions of any pollutant from all these facilities (e.g., an entire source) increased.¹¹ This aspect of the regulations was judicially invalidated¹² as being facially inconsistent with § 111 of the Clean Air Act, which defines a source in terms of a single facility.¹³

Undaunted, the Agency included the bubble approach in its 1978 regulations¹⁴ governing the prevention of significant deterioration (PSD) of air quality under §§ 160-169 of the Act, as amended in 1977.¹⁵ These statutory provisions require preconstruction air quality impact reviews and use of the best available control technology when major stationary sources are modified. The PSD regulations defined a source as "any structure, building, facility … (or combination thereof) which is located on one or more contiguous or adjacent properties …."¹⁶ Under the rules, a modification of one or more facilities within a source would not trigger the PSD requirements as long as "no net increase in emissions … would occur … taking into account all emissions increases and decreases at the source which would accompany the modification,"¹⁷ and there would be no adverse overall air quality impact.

These provisions prompted the second judicial "battle of the bubble," and challenges to their legality were recently resolved in Alabama Power Co. v. Costle.¹⁸ In the course of its lengthy final opinion, the United States Court of Appeals for the District of Columbia Circuit ruled that the Agency had defined the term "source" more broadly than the Act allowed but concluded that Congress did intend an entire plant to be considered a single source for PSD purposes. More importantly, it upheld the provision that physical or operational changes within a major source that do not result in a net increase in emissions are not subject to the PSD requirements. The central components of EPA's incorporation of the bubble concept into its PSD regulations thus received firm judicial approval.

The Agency decided not to adopt the bubble approach [10 ELR 10016] as part of its emissions offset policy¹⁹ for allowing modifications of major sources in non-attainment areas. This decision was based not on doubts as to the legality of applying the bubble in this context; rather, the Agency decided on policy grounds that it would be inappropriate to provide an exemption from strict technology-based control requirements in areas where ambient air quality standards are currently being violated. An industry challenge to EPA's omission of the bubble concept from its emissions offset rules is currently pending before the D.C. Circuit.²⁰

EPA is also considering the possibility of adopting the bubble concept in regulating water pollutants through the national pollutant discharge elimination system (NPDES) permit program under § 402 of the Clean Water Act.²¹ The legal issues in this context are thorny, however, and there are deep disagreements within the Agency as to whether and exactly how the bubble approach should be incorporated in the NPDES program.²² Moreover, strong opposition from key members of Congress, particularly Senator Muskie (D-Me.) may effectively block its inclusion in the NPDES regulations.

Early last year EPA proposed²³ using the bubble concept in the sphere where it would have the widest potential application under the Clean Air Act — the regulation of existing sources under state implementation plans. The proposed policy was generally aimed at encouraging state air pollution control agencies to approve proposals from source operators for alternative mixes of emission controls on plant components so long as net emissions of specific pollutants did not increase and a number of other environmentally protective conditions were met.

Final Bubble Policy

On November 29, 1979, EPA issued its final policy statement²⁴ on the use of the bubble concept by existing sources. The basic premise of the final policy is to allow a source operator to develop and adopt, subject to state and federal approval, an alternative and presumably less costly emission abatement strategy provided there is no adverse environmental impact and the source remains subject to enforceable controls. Theoretically, this will result in greater economic efficiency by allowing sources to put more stringent controls on plant components where the marginal reduction cost is lowest instead of further reducing emissions where the cleanup cost is high.

The policy statement first enunciates EPA's conviction that SIP revisions based upon the bubble approach can provide important economic benefits. It encourages states to publicize the availability of the bubble approach to regulated sources and to adopt proposals from sources for bubble-based SIP revisions. The operator of a source must take the initiative in developing detailed alternative compliance proposals that will meet the SIP's total emission control requirements. This proposed alternative strategy will then be subject to state review and adoption as an SIP revision. EPA approval would also be necessary before the alternative mix of emission controls could be put into effect.

To ensure that adoption of the bubble concept does not undercut the enforceability of emission controls or adversely affect environmental quality, EPA set forth a number of protective conditions. The policy statement emphasizes that compliance with these requirements is a prerequisite for EPA approval of a bubble-based alternative emission reduction strategy.

Eligibility Requirements

There are two prerequisites for eligibility. First, sources may propose to adopt the bubble approach only if they are located in areas that are expected to attain ambient air quality standards for the relevant pollutants by the statutory deadline or can show reasonable further progress toward that goal.²⁵ A question the policy statement leaves unanswered is whether the requisite demonstration can be made for areas subject to an SIP that has received only conditional approval on the ground that attainment is less than assured. EPA did threaten, however, that if attainment is not achieved, sources may be required to install more stringent controls on points where emission reduction was relaxed under the bubble.

The second requisite for eligibility is that a source must have emission limitations and compliance schedules in effect for all emission points involved in the bubble transaction. Such requirements may be set forth in the current SIP, an EPA-approved compliance schedule, or a court decree which EPA finds satisfactory. A history of non-compliance will not make the source ineligible to use the bubble approach if at the time it submits an alternative proposal it also agrees to adhere to a compliance schedule.

Environmental Conditions

The source operator bears the burden of demonstrating that the alternative controls he suggests are equivalent to the existing or proposed SIP requirements in terms of environmental impact. Specifically, the source must first show that its proposal will not jeopardize timely attainment of ambient air quality standards for the pollutant(s) involved. It must also demonstrate that all emissions to be exchanged under the alternative control strategy are quantifiable and that the proposed trades among them are even. Direct measurement is the preferred method of quantification, but indirect calculation based on work practices and equipment standards is also acceptable. The test for equivalence is generally to be consistent in its level of detail with the method used in the SIP's demonstration of attainment. Air quality modeling will always be required, however, where the alternative strategy involves more than one plant or a net increase in emissions.

EPA acknowledged that in certain circumstances the Clean Air Act permits SIP revisions that allow increases [10 ELR 10017] in overall emissions. The Agency nonetheless stated that it does not favor the adoption of such emission-increasing arrangements under the bubble concept and will in fact disapprove such SIP revisions to the extent it has the legal authority to do so.

The pollutants exchanged under the bubble arrangement must be comparable; trades across pollutant categories are thus impermissible. Similarly, increases in pollutants posing severe health hazards cannot to exchanged for corresponding decreases in less harmful pollutants, even if within the same category. SO₂ therefore could not be traded against hydrocarbons, nor could increased carcinogenic coke oven emissions be traded for reductions in other particulates. Increased emissions of a pollutant currently listed as hazardous pursuant to § 112 of the Act²⁶ must be accompanied by a corresponding decrease in emission of the same pollutant at another emission point. Decreases in hazardous pollutant emissions may be traded against increases of nonhazardous pollutants within the same category, however.

Enforcement-Related Requirements

An alternative control strategy must contain specific emission limits for each emission point involved. It must also be treated as an addition to the current SIP rather than as a replacement for the presently applicable emission limitations to assure that no lapse in enforceability results from legal challenges to the bubble proposal.

Third, an alternative control strategy should generally have the same compliance deadline as the existing SIP requirements. Acknowledging that it may be impossible in some instances for a source to implement an alternative approach within the existing schedule, EPA stated that it lacks authority to approve a compliance extension solely for the purpose of allowing the source to adopt the bubble concept. The Agency did note, however, that such extensions are possible for sources that would otherwise qualify under § 172 of the Act.²⁷

EPA has specified that existing enforcement actions shall not be delayed while the state reviews bubble proposals. Moreover, until the proposal is approved by EPA, noncompliance penalties assessed under §§ 113 and 120of the Act are to be based on preproposal compliance cost figures. Finally, the Agency warned against the modification of existing judicial decrees regarding compliance to incorporate bubble provisions but stated that the bubble approach can be made part of the negotiated settlement incorporated into future consent decrees.

Significant Changes From Proposed Policy

EPA's final bubble policy statement includes several noteworthy changes from the provisions originally proposed early last year.²⁸ Interestingly, most if not all of these changes tend to increase the policy's flexibility by cutting back on the accompanying protective conditions and limitations.²⁹ One important departure from the policy as proposed is the deletion of the prohibition against adoption of alternative compliance strategies involving more than one plant. The final policy allows an alternative abatement strategy to cover other plants or sources within the same area. A modeling demonstration that no adverse effect on air quality will result is required as a safeguard in such cases.

Perhaps the most significant change is the allowance of open-dust trades. While the proposed policy expressly forbade reductions in fugitive dust emissions from roads or storage piles to be traded against increases in particulates emitted from industrial processes, the final policy statement takes the position that trades will be permissible in some circumstances.³⁰ Open-dust trades against stack emissions will be approved only if equivalence is demonstrated through actual monitoring readings rather than through air quality modeling.The Agency believes the reliability of models in predicting air quality impacts is higher for stack emissions than for open dust. Source may use modeling demonstrations for open-dust trades which do not involve the lessening of controls on industrial process emissions, however.

The third noteworthy change reflected in the final policy concerns the impact of adoption of the bubble approach on compliance schedules. The proposed policy would have required an alternative control strategy to have the same compliance date as the original SIP provisions. Under the final policy, this requirement has been diluted to a general admonition and compliance date extensions will be available to sources using the bubble that can qualify for them under other statutory provisions.

Fourth, whereas the proposed policy placed primary emphasis on the principle that alternative compliance strategies adopted under the bubble concept could not result in a net increase in emissions of any specific pollutant, the focus of the final policy has shifted noticeably toward the looser requirement of no net adverse effect on air quality. EPA has now explicitly acknowledged the possibility of state-approved SIP revisions based on the bubble approach which envision net emissions increases. In an attempt to discourage the states from approving emission-increasing SIP revisions, EPA voiced its intent to disapprove such revisions to the extent of its authority under § 110(a).³¹ But its acknowledgement that it lacks authority to completely bar such SIP revisions serves as a clear indication that the requirement of a net decrease or even trade in emissions is no longer an absolute rule.

Issues, Problems

There may be some lingering doubt as to the legality of EPA's policy for allowing use of the bubble approach in the regulation of existing sources through state implementation plans under the Clean Air Act.³² The debate is [10 ELR 10018] now focused, however, on the advisability of specific provisions as well as the policy's overall feasibility and acceptability in practice. A central issue is whether the protective conditions are adequate to prevent further compliance delays, gaps in SIP enforceability, and backsliding in environmental quality. The limitations attached to use of the bubble approach were weakened in a number of respects from the provisions in the Agency's original policy proposal, increasing the policy's flexibility and broadening its applicability. At this point, there is no way of estimating the environmental price of enhancing the bubble's advantages to industry in this manner, but these changes understandably reinforce the preexisting fears of some observers that in the long run the environmental harm associated with use of the bubble concept will outweigh its putative economic benefits.

A related and perhaps equally important question concerns the burden that adoption of the bubble approach will impose on sources and state air pollution control agencies. A source wishing to propose an SIP revision must incur significant front-end expenses in evaluating alternative compliance strategies and meeting EPA's exacting demonstration requirements. Similarly, state agencies will have to commit a significant portion of staff time to reviewing proposed SIP revisions and then negotiating with EPA for final approval.

The concern here is that these costs may be so great that neither the sources nor the states will find bubble-based compliance alternatives to be worth the trouble. EPA takes the view that the bubble concept will be of interest primarily to medium or large industrial facilities for whom the initial development and demonstration costs will be dwarfed by the prospective savings from using a less expensive control strategy. The Agency also attempted to counter the specter of the states turning a cold shoulder to the bubble policy by downplaying the associated burdens and promising to work with the states in expediting the SIP revision process. EPA is stating the obvious in its assessment that in most instances where use of the bubble approach could make a real difference in a source's compliance costs, the savings will more than outweigh the initial expense of developing and demonstrating the alternative strategy. But something more than reassurances may be needed to overcome well-founded fears that implementation of the policy will overwhelm the small and already overburdened staffs of state air pollution control agencies.

In addition to these general issues, the final policy statement raises several more specific problems. First, SIP revisions incorporating alternative control strategies that involve net increases in emissions, which EPA discourages but acknowledges may nevertheless occur, may serve to exacerbate acid rain and other difficulties associated with the long-range transport of air pollutants. Second, where an existing source is emitting less than the amount permitted under the relevant SIP provisions, allowing it to use the bubble may result in air quality deterioration if a phantom reduction in its "paper" emission rate is traded for real increases in emissions at another point or plant. The policy prohibits such trades only if the actual emission rate rather than the permitted rate was used in the SIP's demonstration of timely attainment.

A further question lurking behind a number of the policy's provisions regards the consequences for sources using the bubble approach in areas in which, despite the earlier demonstrations, national air quality standards are ultimately not attained by the 1982 or 1987 statutory deadlines. The final policy statement warns that should this happen such sources could be required to place more stringent controls on emission points where abatement was lessened under the approved alternative control strategy. This threat may be greeted with skepticism in some quarters, however, as the substantial increase in compliance costs that such a move would entail could serve as a strong impetus for additional legislative delays of the statutory timetable.

Conclusion

In announcing its final policy calling for widespread adoption of the bubble concept by existing air pollution sources, EPA has taken a major step towards harmonizing pollution controls with economic principles, a goal that has been advocated by a small but growing cadre of commentators.³³ The policy also represents a noteworthy attempt by the Agency to introduce flexibility and individual initiative into its regulatory program at minimal or no environmental cost.

Industry has generally welcomed the final bubble policy, although there is some concern that the extensive conditions and limitations will significantly and unnecessarily restrict its availability. Environmentalists predictably take the opposite view and worry that the conditions, particularly insofar as they have been diluted since the original policy proposal, are insufficient to prevent weakened SIP enforcement and a decline in environmental quality. They also fear that the costs of rigorous implementation of the policy will be more than state and local air pollution control boards can bear.

Issuance of the final policy statement is a major advance in EPA's long campaign for widespread adoption of the bubble concept. However, its successful implementation is far from assured given its dependence on the initiative of sources and the ability and inclination of states to review and approve such proposed SIP revisions. Despite these and other uncertainties, EPA retains high hopes for the success of the policy. Unless it languishes unused or brings a flood of inadequate applications that swamps state and EPA officials, neither of which is likely, the bubble approach seems here to stay. Moreover, if the policy proves successful in practice, it will almost certainly be followed by further EPA actions to incorporate market mechanisms into pollution control.

1. EPA OFFICE OF PLANNING AND MANAGEMENT, REGULATORY REFORM INITIATIVES: PROGRESS REPORT 1 (Oct. 1979).

2. 44 Fed. Reg. 71780, 10 ELR 30001 (Dec. 11, 1979).

3. Remarks of Douglas Costle, EPA Administrator, Press Conference on New "Bubble" Policy, Washington DC (Dec. 3, 1979).

4. See generally, EPA OFFICE OF PLANNING AND MANAGEMENT, REGULATORY REFORM INITIATIVES: PROGRESS REPORT (Oct. 1979).

5. 40 C.F.R. pt. 51, app. S; 44 Fed. Reg. 3274 (Jan. 16, 1979); 44 Fed. Reg. 51925 (Sept. 5, 1979).

6. Remarks of Douglas Costle, supra note 3.

7. See Comment, EPA's Widening Embrace of the "Bubble" Concept: The Legality and Availability of Intra-Source Trade-Offs, 9 ELR 10027 (1979).

8. 40 Fed. Reg. 58416 (Dec. 16, 1975).

9. See Clean Air Act § 111, 42 U.S.C. § 7411, ELR STAT. & REG. 42216.

10. 40 C.F.R. § 60.2(d) (1977), 40 Fed. Reg. at 58418.

11. 40 C.F.R. § 60.14(d) (1977), 40 Fed. Reg. at 58419.

12. ASARCO, Inc. v. EPA, 578 F.2d 319, 8 ELR 20164 (D.C. Cir. 1978). See Comment, D.C. Circuit Rejects EPA's Use of the "Bubble" Concept in Applying New Source Performance Standards, 8 ELR 10052 (1978).

13. A stationary source, "for purposes of" § 111, is defined in § 111(a)(3) of the Act as "any building, structure, facility or installation" that emits or may emit an air pollutant. 42 U.S.C. § 7411(a)(3), ELR STAT. & REG. 42216.

14. 40 C.F.R. § 52.21 (1978), 43 Fed. Reg. 26388 (June 19, 1978), ELR STAT. & REG. 46611.

15. 42 U.S.C. §§ 7470-7479, ELR STAT. & REG. 42233-36.

16. 40 C.F.R. § 52.21(b)(4) (1978), 43 Fed. Reg. at 26404, ELR STAT. & REG. at 46627.

17. 40 C.F.R. § 52.21(j)(4) (1978), 43 Fed. Reg. at 26407, ELRSTAT. & REG. at 46630.

18. __ F.2d __, 10 ELR 20001 (D.C. Cir. Dec. 14, 1979), superseding 606 F.2d 1068, 9 ELR 20400 (D.C. Cir. June 18, 1979) (per curiam).

19. 40 C.F.R. pt. 51, app. S; 44 Fed. Reg. 3274 (Jan. 16, 1979); 44 Fed. Reg. 51925 (Sept. 5, 1979).

20. Manufacturing Chemists Ass'n v. EPA, No. 79-1112 (D.C. Cir., filed Jan. 19, 1979).

21. 33 U.S.C. § 1342, ELR STAT. & REG. 42141.

22. See Comment, EPA's Widening Embrace of the "Bubble" Concept: The Legality and Availability of Intra-Source Trade-Offs, 9 ELR 10027 (1979).

23. 44 Fed. Reg. 3470 (Jan. 18, 1979).

24. 44 Fed. Reg. 71780, 10 ELR 30001 (Dec. 11, 1979).

25. The Agency stated that it will make an exception to this condition for sources of volatile organic compounds included in control techniques guideline categories located in areas that cannot demonstrate attainment of the ozone standards by 1982 provided they use Reasonably Available Control Technology. See Clean Air Act § 172, 42 U.S.C. § 7502, ELR STAT. & REG. 42238.

26. 42 U.S.C. § 7412, ELR STAT. & REG. 42219.

27. 42 U.S.C. § 7502, ELR STAT. & REG. 42238.

28. 44 Fed. Reg. 3470 (Jan. 18, 1979); see generally, Comment, EPA's Widening Embrace of the "Bubble" Concept: The Legality and Availability of Intra-Source Trade-Offs, 9 ELR 10027, 10029-30 (1979).

29. Senator Muskie (D-Me.) has expressed "strong reservations" concerning the final policy because the original proposal has been "altered extensively." Letter from Sen. Edmund Muskie, Chairman, Subcommittee on Environmental Pollution of the Senate Committee on Public Works and the Environment, to Douglas Costle, Administrator, Environmental Protection Agency (Jan. 4, 1980).

30. Senator Muskie has already directed harsh criticism at this provision and urged that open-dust trades for industrial process emissions be barred completely. Id.

31. 42 U.S.C. § 7410(a), ELR STAT. & REG. 42212.

32. See generally, Comment, EPA's Widening Embrace of the "Bubble" Concept: The Legality and Availability of Intra-Source Trade-Offs, 9 ELR 10027 (1979).

33. See, e.g., F. ANDERSON, et al., ENVIRONMENTAL IMPROVEMENT THROUGH ECONOMIC INCENTIVES (1977); W. IRWIN & R. LIROFF, ECONOMIC DISINCENTIVES FOR POLLUTION CONTROL: LEGAL, POLITICAL AND ADMINISTRATIVE DIMENSIONS (GPO 1974); A. KNEESE & C. SCHULTZ, POLLUTION, PRICES, AND PUBLIC POLICY (1975).