A Conceptual Framework for an Acid Rain Control Program

Larry Banks Blackwood

Mr. Blackwood, a graduate of Stanford Law School and the North Carolina State University School of Engineering, is presently an attorney in the Office of Enforcement and Compliance Monitoring of the United States Environmental Protection Agency. From 1984 to 1988, he worked on acid rain policy and implementation issues in EPA's Office of Air and Radiation. Previously, he was an Assistant Attorney General practicing environmental law for the state of Illinois, and a consultant on environmental and energy policy. The author would like to thank Brian McLean, David Bassett, Eleanor Leonard, and Ernestine Hall of the Environmental Protection Agency for their invaluable help in formulating the ideas in this article. The views expressed are those of the author, and do not represent the position of the Environmental Protection Agency.

[19 ELR 10166]

The long national debate on the acid rain problem may be approaching a turning point. President Bush and his EPA Administrator, William Reilly, have stated their belief that the time has come to take action. The new Senate majority leader, George Mitchell of Maine, is one of the leading advocates in Congress for acid rain control. The ten-year National Acid Precipitation Assessment Program (NAPAP) is nearing its conclusion. These developments may shift the focus of the national debate from whether there should be a new acid rain control program to how that program should be designed.

Numerous acid rain program designs have been proposed in the attempt to find a politically acceptable compromise. The design selected will have major consequences long after a program is enacted, however. It will be a key factor in the ultimate success or failure of implementation of the program by the Environmental Protection Agency (EPA), the states, the courts, and the private sector.

One major element of program design is the number and difficulty of the issues that are resolved by Congress in the enacting legislation, and the corresponding extent of the issues that are left to the implementation process. This Dialogue explains the importance of this aspect of program design, describes four alternative types of acid rain control programs differentiated by it, and outlines the likely course of implementation of each type of program. It concludes that a program based upon congressionally mandated annual emission ceilings for individual sources, calculated as the product of an average emission rate deemed to be acceptable and the actual utilization of each source over some representative period, would be the simplest and surest for EPA and the states to implement. At the same time, such a program could allow the greatest flexibility of compliance options to the regulated sources.

This Dialogue does not address the initial questions of whether or how much U.S. emissions of the sulfur dioxide and nitrogen oxide precursors of acid rain should be reduced, or whether significant reductions could be achieved without a new regulatory program. Its premise is that if a new regulatory program is enacted, it will be in the long-term best interest of all parties to make that program as workable as possible.

The Resolution of Program Issues Through Statutory Objectives

Any federal pollution control program can be conceptualized as a long series of issues and decisions. These range from the scientific questions regarding the environmental effects of various pollutants, to the values to be ascribed to those effects, to the engineering, economic, and equity questions involved in deciding which sources are to do what to control each pollutant. The statutes, regulations, and ongoing management of each program constitute an elaborate system that allocates authority over various issues among the many interested parties and institutions, and guides their decisions.

The most basic decisions in this process are the legal objectives established by Congress for the program. These translate a broad societal concern about an environmental situation (e.g., in the present Clean Air Act, ambient air pollution) into obligations for the implementation process (e.g., attainment of health and welfare based national standards for criteria pollutants).¹ Their establishment acts to resolve (for program purposes) all of the issues subsumed within them — e.g., whether the situation in question constitutes a problem requiring a federal response, and what response is appropriate. Congress is declaring that the situation is a national problem, and moreover that satisfaction of the statutory objectives represents a necessary and sufficient response to it, whether some parties continue to disagree or not. Ongoing debate as to the seriousness of the situation, or whether the statutory objectives aretoo severe or not severe enough, is thereafter legally irrelevant to the remaining decisions to be made by the implementing parties.

The statutory objectives can thus resolve issues because Congress possesses great discretion in establishing them. It can base its decisions on subjective factors or political compromise, or bypass arguments and issues bearing on the objectives altogether. Its decisions are nevertheless to be carried out, provided only that they pass constitutional muster and unless they are changed by further legislation.

Decisions by EPA and the other implementing parties have much less power to resolve issues, because they possess much less discretion than Congress. EPA decisions can be reversed by the courts if they are "arbitrary, capricious, an abuse of discretion, or otherwise not in accordance with law"; "contrary to constitutional right, power, privilege, or immunity"; "in excess of statutory jurisdiction, authority or limitations"; or "without observance of procedure required by law."² The obligatory rulemaking process further requires, in effect,³ that virtually every argument on every possible issue be specifically considered and objectively decided. The discretion of the other implementing parties is still more limited, in that they must follow EPA's decisions. Parties who disagree with any of the implementation decisions, or who simply want to avoid or delay action, thus have abundant opportunities [19 ELR 10167] to raise issues again and again through the agencies and the courts and thereby prevent them from being finally resolved.⁴

The number and difficulty of program issues that are resolved by Congress is therefore a key factor in the speed, the administrative cost, and sometimes even the possibility of implementation. Congress selects which issues are to be resolved in the enacting legislation, and which are to be deferred to the implementation process, through the design of the program and particularly through the statutory objectives established. Thus, in the present Clean Air Act Congress decided that ambient air quality standards sufficient to protect public health and welfare should be achieved throughout the country. This left the issues of which types and concentrations of pollutants should be subject to national standards to satisfy that objective, and how to attain those standards, to EPA, the states, the pollution sources, and the courts. Prior to the 1970 amendments to the Clean Air Act, Congress stopped short of requiring national standards, and thus also left the issues of whether, and how much, to regulate ambient air quality to the states and localities. It could alternatively have gone further, and specified in the statute the types or concentrations of pollutants to be regulated or even the emission controls to be required. This would have correspondingly removed those issues from the implementation process.

The extent of the issues in any new program to be decided by Congress is thus itself a complex public policy question. Favoring congressional decisions are the desirability of speed, certainty, and minimum administrative cost in the implementation process. Favoring deferral are the possibility of a better resolution of some issues through scientific or engineering expertise, fine tuning, or case-by-case consideration. Congressional resolution of these issues may hinder implementation by making the entire program less effective or feasible.

The issues that ideally should be deferred to the implementation process are those for which the participants in that process are, for some reason, better able to decide than Congress. On the other hand, to the extent that Congress defers to the implementers issues that the latter are no better (or, indeed, less) able to decide than Congress, success of the program will be delayed or made less likely and will require greater administrative cost.

Programs, however, are not created at one point in time by any one person or even any one institution. They evolve. It is seldom possible to make complete decisions as to which issues ought to be resolved at which stage of the process or by which parties; and it would probably not even be desirable to try to decide all issues prior to actual experience with the program. Often, the possibilities of political compromise dictate which issues can be resolved by congressional decisions and which must be deferred. Such decisions can further the overall debate even if they do not go as far as would be desirable from a pure program design perspective. An understanding of design concepts can help to bring about a successful program, however, even though design concepts cannot be the sole determinants of program design.

Options for Acid Rain Program Design

The Range of Acid Rain Issues

The acid rain problem has the same range of scientific, value, and equity issues as every other pollution problem. They differ from most cases in that they are on a larger scale, are characterized by even greater uncertainty, and affect more varied interests. These differences are sufficient, however, to warrant a reexamination of basic program design concepts.

Briefly, the issues are:

(1) The effects of acid deposition on receptor lakes, forests, materials, and human health, and the value to be placed on those effects.

(2) The relationship between levels of acid deposition and extent of those effects at different receptors.

(3) The relationship between levels of emissions at some locations and levels of deposition at other locations.

(4) The measures to be taken by source areas and/or individual sources to control emissions.

A program to control acid rain could theoretically be based upon statutory objectives encompassing any extent of these issues.

A Program Based Upon Environmental Effects

Legal authority already exists for a limited acid rain program based upon environmental effects in other countries (i.e., Canada). Section 115 of the Clean Air Act⁵ provides that if the EPA Administrator has reason to believe that U.S. emissions "cause or contribute to air pollution which may reasonably be anticipated to endanger public health or welfare in a foreign country," and if the foreign country gives reciprocal rights to the United States, he shall give formal notification thereof to the governor of the state in which the emissions originate. The state is then to revise its Clean Air Act implementation plan insofar as the plan is "inadequate to prevent or eliminate the endangerment." A finding by the Administrator to trigger § 115 must be made through the formal rulemaking process.⁶

Section 115 embodies an implicit congressional objective that U.S. emissions should not endanger public health or welfare in a foreign country. However, all of the scientific, socioeconomic, and equity issues that must be decided to carry out that policy are left to the implementation process. Thus, EPA and/or the states, courts, and sources would have to decide whether Canadian health or welfare is threatened by acid deposition, the nature of that threat, the levels of acid deposition responsible for that threat, the portion of those levels attributable to U.S. sources, the allocation of the U.S. portion among states and sources, and the control measures to be taken by each source. But because none of these implementing parties has the ability of Congress to resolve issues, those decisions would be extended indefinitely. Moreover, the procedure for making [19 ELR 10168] those decisions would be expanded to include all of the political, administrative, and legal forums available to all of the interested parties to advocate their individual perspectives and self-interest in the outcome of the decisions.

Successful implementation of such a program would be very doubtful. Indeed, due to the intermingled issues of scientific uncertainty and subjective values involved in the concept of a "threat" to Canadian health and welfare, the very definition of "success" would remain open to debate.⁷ At best, implementation would require much more time and administrative cost than a program based upon a more specific statutory objective.

A Program Based Upon Deposition Loadings

The statutory objective enacted by Congress could go further, and specify quantitatively the levels of acid deposition deemed to threaten public health or welfare (in Canada and/or the United States). Alternatively, it could define the kind or extent of threat to be prevented; and set up a two-stage implementation process whereby EPA would set quantitative deposition standards corresponding to that degree of environmental protection, and EPA and the states would develop plans to attain the standards. Such an approach would be analogous to the national ambient air quality standard/state implementation plan program which is the focus of much of the present Clean Air Act.

A deposition loading objective would resolve or simplify the issues of the degree of environmental protection to be achieved in the implementation process. Once the objective was set, further debate as to its merits would be irrelevant to the program until and unless it was changed. This would leave to the implementing parties, however, the complex issues of long-range source/receptor relationships and the allocation of control requirements among states and sources to achieve the objective.

A Program Based Upon State Emission Limitations

Most of the acid rain bills before Congress have specified limitations on emissions of sulfur dioxide and nitrogen oxides from each state, to be achieved through EPA-approved state implementation plans. They vary as to whether the emission limitations are in the form of total annual emission ceilings or average emission rates, the amount and timing of required emission reductions and associated costs, adherence to the "polluter pays" principle versus spreading of control costs through various types of fees and subsidies, measures to encourage control through conventional and/or innovative technology and energy conservation, allowance of emissions trading, applicability to new emission sources, provisions for "default" by a state in its program obligations, and a host of other details.⁸ All of them, however, would subsume all of the debate about the causes and effects of acid rain, the value of those effects, and source-receptor relationships in the congressional decision that the mandated state emission limitations represent a necessary and sufficient answer to the acid rain problem.

This resolution of these complex issues would certainly simplify the program compared to environmental effects or deposition-based programs. It is usually argued that the remaining issues regarding the plans to meet the specified state emission limitations are, like those involved in the plans to meet the ambient air quality standards, best left to the states. Thus, each state would have flexibility to achieve its acid rain control obligation in a manner that represents least cost to the state as a whole (usually through fuel switching), least disruption to high sulfur coal markets and employment (usually through technological controls such as scrubbers), maximum protection of the environment (including related concerns such as ambient air quality and scrubber sludge disposal), or any combination of these and other state objectives.

The converse of this flexibility, however, is that all of the issues involved in identifying options and choosing among possibly conflicting state objectives would be open. Unlike ambient air quality planning, in which short-range air pollution modeling can usually objectively identify the emission control requirements for each source, many different plans could satisfy most of the state acid rain control targets that have been proposed. The corresponding obligations on individual sources could range all the way from no control to shutdown. Therefore, the political, administrative, and legal processes of each state would have to deal with the technical complexities of a wide variety of possible plans, the weighing of possibly competing economic and social policies in choosing among possible plans (i.e., whether to minimize unemployment effects or costs), and the equities among sources of different possible control allocations.

Throughout the planning process, conflicting interest groups in each state can be expected to advocate their positions on these issues just as forcefully as their counterparts on the national level have for the past ten years. In addition, some individual sources or groups of sources might seek (through the state political, administrative, and legal forums) to put as much as possible of the burden of meeting the mandated state control targets on other sources.

Most discussion of state emission limitation programs has assumed that most states would adopt control plans that represent least cost to the state as a whole. But a least cost state plan would probably impose higher costs on some sources, or even whole regions of the state, than alternative plans. Thus, a least cost plan for a hypothetical state might involve the installation of scrubbers at one or more large coal fired power plants belonging to one utility, which happen to be the least expensive plants in the state to control, and little or no action at relatively difficult-to-control plants of other utilities in the state.⁹ But this would impose most or all of the costs of the state's control plan on the utility owning the plants to be scrubbed, and its ratepayers. Those [19 ELR 10169] ratepayers would thus, in effect, be subsidizing lower electric rates in the rest of the state than would be necessitated under some alternative control plan. While some individuals and institutions might give primary consideration to least cost for the state as a whole, most people are likely to be more concerned with least cost to themselves.

It might even be in the interest of some sources to prevent or delay adoption of any state plan until after the federal deadline, and thus force the state into default. This could be the case, for example, for an industrial source that could feasibly install controls, if (as under a number of proposals) the statutory default provision imposed control requirements on each utility source. A state plan would probably require controls on the industrial source, whereas under default it would have to do nothing.

These problems would be exacerbated by the fact that the electric utility industry, which accounts for about two thirds of U.S. sulfur dioxide (SO₂) emissions, would be the most important factor in most state acid rain control plans. The electric power system is inherently much more difficult for individual states to control than industrial sources, because its operation is essentially interstate in character. The power plants or service areas of many utilities are located in more than one state, and electric power is routinely bought and sold between utilities and across state lines. Emission control plans are intimately related to system operations. The emission control plans of each state would thus directly affect, and would be affected by, the plans of every other state with which it is interconnected on the utility grid.¹⁰ The only mechanism provided in most of the proposed state emission limitation programs to coordinate the control plans of interconnected states is the required EPA approval. EPA, however, would have neither the time (under the proposed statutory deadlines) nor the expertise to do so effectively.

A Program Based Upon Individual Source Emission Limitations

Some of the basic acid rain control proposals discussed in Congress,¹¹ and many of the default provisions, would place federal emission control requirements directly on individual sources. These requirements could be stated as performance or equipment standards. They could be applied to all sources in a specified category (e.g., all coal fired steam generators above a certain size, or reaching a certain age, or both); or to the largest or most feasible to control sources according to some criteria; or simply to specifically named sources. They could be limited to existing emission sources, or they could take into account future sources either by requiring their emissions to be offset or by requiring sufficient initial reductions to accommodate new sources. Whatever the form and scope of the required controls, however, the congressional decision to impose them would resolve or subsume (for program purposes) most of the issues from environmental effects and values down to and including the initial allocation of control obligations among sources.

Implementation of such a program would be correspondingly straightforward. For EPA and the state environmental agencies, it could involve simply expanding the present programs that implement the new source performance standards to include the specified existing sources. Indeed, one of the arguments for this approach is that the new source standards were originally premised on the belief that existing uncontrolled power plants would be retired and replaced by controlled new plants at about 30 years of age. According to this view, the continuing acid rain problem is largely due to the extension of the lives of these uncontrolled plants to 40 to 60 years or more for economic reasons (including the cost of the emission controls that would be required on replacement plants); and the solution is to bring them under controls comparable to new sources. The most commonly proposed performance standard for such programs is therefore 1.2 pounds of sulfur dioxide emissions per million BTU's of heat input to the plant, corresponding to the original new source performance standard for coal fired power plants.

A primary objection to such approaches is that individual existing sources vary greatly in the feasibility and cost of emission controls, and therefore should not be subjected to a common requirement. It would be possible, however, to design an individual source emission based program with as much or more flexibility in the means by which control obligations could be satisfied as a state emission based program. Moreover, a source based program [19 ELR 10170] could give the decisions about the exercise of that flexibility to the owners and operators of the sources who understand them best, rather than to state planners.

This flexibility could be achieved by specifying the source limitations as annual emission ceilings, rather than the average emission rates that have usually been discussed. The annual emission ceiling for each subject source would be calculated by multiplying the average emission rate deemed acceptable (e.g., 1.2 pounds of sulfur dioxide per million BTU) times the average utilization of the source over some representative period (e.g., average BTU input to the source over a five-year period). Each source could then be brought into compliance with its emission ceiling through whatever means, or combination of means, was determined to be best by its owners or operators. Technological controls or fuel switching could be applied to meet the base emission rate, leaving the owners and operators free to utilize the source up to its average utilization over the specified period. Alternatively, they could "overcontrol" the source and be free to operate it more than the previous average; or apply less or no emission controls and meet the ceiling by operating it correspondingly less than the previous average. The latter course might be preferred, for example, for an old power plant which was to be phased out or used as a "peaker" to meet short-term load demands. Reduced operation could also be made possible through energy conservation, or by dispatching plants on the basis of emissions instead of (or in addition to) operating costs. The annual ceiling for each source could also be satisfied through an emissions trading program by "overcontrol" at other sources. Finally, and significantly, this flexibility of compliance options could continue for the life of the source. An old power plant could comply initially through reduced operation or a temporary emissions trade (an "emissions lease"). Then, if at any time in the future system needs or economics call for increased utilization of the plant, or technological controls improve so as to become feasible for it, the mix of emission controls, utilization, and emission trades for the plant could be altered.

The role of EPA and/or the state environmental agencies in implementing this program would be to calculate the emission ceilings for each source, and review and keep track of the compliance options applied to ensure that they are met. This would undoubtedly involve new issues, some of them complicated and controversial, such as the design of an emissions trading program and the means of enforcement of the source emission ceilings. States desiring to incorporate their own goals in the program (e.g., protection of local high sulfur coal markets and employment) could do so by modifying or restricting the compliance options allowable to sources within their borders (e.g., limit or ban fuel switching). These issues would involve, however, nothing like the scale or complexity of allocating basic control obligations among all the emission sources in each state and the country.

Conclusion

The design of an acid rain control program is independent from the level of environmental protection desired. Any level of protection, and associated costs, the Congress determines to be appropriate could be sought through any of the types of programs described in this Dialogue. Likewise, provisions to distribute the costs of control among the various possible groups, to encourage or discourage particular control options, to account for future emission sources, and to address the other issues involved in the acid rain problem could be incorporated in any of the program types. The probability that the country will actually achieve the level of protection from acid rain desired, however, and the time and administrative expense that will be required, will be dramatically affected by the type of program selected.

It will be years or decades before there is a scientific concensus on the various possible effects of acid deposition, let alone on the quantitative relationships between levels of deposition and levels of effects. If those relationships are ever established, they will be different for each effect (e.g., lakes, forests, and materials) and even for the same effect at different locations (e.g., lakes in ecosystems with different natural chemistry). Even if the extent of damage to the ecosystem can be determined, its valuation will always be at best very difficult and highly debatable. Source-receptor relationships over thousands of miles, affected by complex atmospheric chemistry and constantly changing meteorology, will always add additional levels of uncertainty between desired environmental effects and emission control measures to be taken. Therefore, if Congress decides to take action to control acid rain, it should simultaneously subsume or resolve these issues for purposes of the program through the objectives established for it.

Economic and equity issues will always make the allocation of control obligations among emission sources by federal or state governments controversial. Allocation decisions that are perceived by individual sources (and their stakeholders) as unreasonable or inequitable, or that allow them to avoid or defer control or shift obligations to other sources, will expand or prolong the controversy. Conversely, decisions that are perceived as reasonable and final will restrict or resolve it. Due to the inherently interstate operation of the electric utility system and continuing internal opposition to any action, many states will not be able to make such allocations possessed by EPA and other agencies are also available to Congress. No agency, however, has the ability of Congress to resolve the equity issues involved. Therefore, a congressional decision to control acid rain should also allocate control obligations among individual emission sources.

An important aspect of "reasonableness" in imposing control obligations is that they should allow sources as much flexibility as possible to meet their obligations in ways that best suit their particular situations, and involve government (federal or state) as little as possible in the exercise of that flexibility. This will minimize the cost of meeting the program objectives both for the sources and for goverment. It can be achieved by framing the control obligations for individual sources as annual emission ceilings, to be met (subject to additional restrictions that may be imposed by the states) by any combination of emission control measures, source utilization, and emission trading. This, therefore, should be the format for a federal acid rain control program.

1. Clean Air Act §§ 108-110, 42 U.S.C. §§ 7408-7410, ELR STAT. CAA 006-011.

2. Clean Air Act § 307(d)(9), 42 U.S.C. § 7607(d)(9), ELR STAT. CAA 046.

3. Although the Administrative Procedure Act does not require every comment in a rulemaking process to be addressed (St. James Hospital v. Heckler, 760 F.2d 1460 (7th Cir. 1985)), EPA and other agencies are acutely aware that they cannot be challenged for answering a comment, but they may be challenged for a failure to answer. The course of prudence, therefore, is to answer virtually every comment.

4. See Small Refiners Lead Phase-Down Task Force v. United States Environmental Protection Agency, 705 F.2d 506, 13 ELR 20490 (D.C. Cir. 1983), for a discussion of the procedural and substantive standards of review of EPA decisions. For an example of how lengthy and complex the rulemaking process and associated litigation can be in an area related to acid rain, see the 17-year history of the "stack height" regulations related in Sierra Club v. Environmental Protection Agency, 719 F.2d 436, 13 ELR 21001 (D.C. Cir. 1983), and Natural Resources Defense Council, Inc. v. Thomas, 838 F.2d 1224, 18 ELR 20519 (D.C. Cir. 1988).

5. 42 U.S.C. § 7415, ELR STAT. CAA 017.

6. Thomas v. New York, 802 F.2d 1443, 16 ELR 20925 (D.C. Cir. 1986).

7. The issue of whether EPA should, or must, take action under § 115 to control acid deposition in Canada has already occupied two presidential administrations, a U.S. District Court, and a U.S. Court of Appeals. Id. It is now once again before the Court of Appeals for the D.C. Circuit in three separate actions. Her Majesty the Queen in Right of Ontario v. Environmental Protection Agency, No. 88-1778; New York v. Thomas, No. 88-1812; and Sierra Club v. Thomas, No. 88-1780.

8. See, e.g., S. 52 (Stafford) and H.R. 4567 (Waxman-Sikorski) from the 99th Congress, and S. 1894 (Mitchell) and H.R. 4331 (Cooper) from the 100th Congress.

9. Acid Rain Legislation and Utility Commission Regulation, Carol L. Etter and Dr. Daniel M. Violette, RCG/Hagler, Bailly Co. for U.S. EPA, (Sept. 1988), Section 2.2, 2.5.

10. The service areas of many utilities span all or parts of more than one state. Many power plants are owned by or serve, in whole or in part, utilities that are centered in states other than the state where the plants are located. Most utilities constantly buy and sell power across state lines, so that their combined load demand can be met in the most cost-efficient possible manner by their combined generating facilities. Thus, the system is for operational purposes virtually a single grid encompassing many states.

Least cost operation of the utility grid requires that individual power plants be "dispatched" (i.e., put on line) in priority of their cost of operation. Thus, the more cost-efficient a plant is relative to the other plants available to the system, the more it will be used. The imposition of emission controls on a power plant, however, usually decreases its cost-efficiency of operation and its resulting priority of dispatch relative to uncontrolled or less controlled plants. This decreased utilization of more controlled plants relative to less controlled plants has two major consequences for control plans. First, the total emission reduction from the system will not be as great as would be predicted from the amount of control imposed if there were no resulting changes in plant utilization. Second, the imposition of power plant emission controls in one state may increase the utilization, and the resulting emissions, of plants in other states.

In an extreme case, if the control plan of one state called in part for the closing of an old coal fired plant whose power production was primarily exported to interconnected states, those states would somehow have to replace that power supply. To the extent that replacement power came from increased utilization of coal-fired plants within their borders, their state emissions would increase. Their control plans could then fail to meet the federally mandated targets, unless they had been formulated in light of the plans of all the interconnected states.

More insidiously, some states might deliberately attempt to shift part of the costs of their control strategy to ratepayers in other states. This could be done by imposing disproportionate control requirements on power plants located in the state that are owned by or serve utilities in other states. The mere possibility of such a strategy could delay the implementation process through lengthy debate and litigation over whether such a strategy was in fact incorporated in a state plan and if so, whether the plan was to that extent unconstitutional as a discrimination against interstate commerce.

The only apparent mechanism to coordinate these interrelated aspects of the control plans of different states is their review and approval by EPA. That agency is not, however, expert in the intricacies of the operation of the electric utility system. Even if it were, most of the proposed state emission based programs would allow all of the state plans to be submitted to EPA on the same day, and would give EPA as little as six months to review them all.

11. See, e.g., H.R. 3400 (Waxman) from the 98th Congress and the "Mitchell Compromise" discussed in the 100th Congress.