Below is the uncorrected machine-read text of this chapter, intended to provide our own search engines and external engines with highly rich, chapter-representative searchable text of each book. Because it is UNCORRECTED material, please consider the following text as a useful but insufficient proxy for the authoritative book pages.
Automotive Emissions Research DONALD KENNEDY Stanford University THOMAS P. CRUMBLY Health Effects Institute We must notforget the policy objective behind auto emission regulations to protect the public health. Short-term tactical battles over standards cloud much more fundamental issues- namely, will reduction of auto pollution actually improve ambient air quality significantly, and will better air actually improve public health? These ques- tions have not yet been answered satisfactorily. Rep. David Stockman Harvard University October 19, 1978 Emissions Control: The History of Conflict The United States has had laws to control air pollution for the past 30 years. Since the passage of amendments to the Clean Air Act in 1970, our society has made an heroic effort to reduce automotive emissions, as one cornerstone of an overall air pollution control policy. And as a result of these laws, we have, in fact, seen major reduc- tions in several pollutants. Some of the Air Pollution, the Automobile, and Public Health. ~ 1988 by the Health Effects Institute. National Academy Press, Washington, D.C. 3 benefits of control are obvious: in parts of the country, the air is noticeably cleaner; and with recent improvements in many . . . inspection anc maintenance programs, fewer cars will present gross problems that lead to emissions beyond current standards. The costs of this control have also been great and not only in terms of dollars expended by taxpayers, stockholders, and car owners. Throughout the 1970s, our efforts to make meaningful emission reduc- tions and to ensure that these reductions were carried out provided difficult tests for some in our society. For a time, perceived attempts by some car companies to resist regulation brought the industry into disre- pute with much of the public. In addition, the Clean Air Act amendments raised ex- pectations that have not yet wholly been achieved. With the advantage of nearly two dec- ades of hindsight, it seems indisputable that the 1970 law was an important exercise in symbolic politics. The car companies had permitted themselves, one could argue, to be cast as forces of evil. The Congress, responding to the burgeoning environmen- tal movement so well symbolized in the Earth Day of that year, seized the high moral and political ground as a way to transform a policy situation that had be
4 The Social Context of Automotive Emissions Research come increasingly unsatisfactory. The Congress overwhelmingly enacted stan- dards and mandated pollutant reductions that would protect the public and provide, in the words of the statute, an "adequate margin of safety," for all people-includ- ing particularly sensitive populations. The Congress further empowered the presi- dent, now acting through the new Envi- ronmental Protection Agency, to take strong enforcement action against compa- nies and states that did not comply with tough timetables for action. The struggle to meet these new standards in a timely way led, at various times during the 1970s, to threats by major corporations to shut down, and to governmental plans for state compliance that seemed to many as exer- cises in futility. In the eyes of many citi- zens, the reputations of government as well as industry suffered during this period. Benefits of Regulation Have the dollars, antagonisms, and lost credibility been worth the benefits our so- ciety has derived from mobile source reg- ulation? Should we continue to control a major sector of the economy in the same way? Should we continue to explore alter- native methods of regulation, such as emis- sion fees, that have always proven politi- cally unacceptable in the past? The answers will only be found by assessing the benefits we have already achieved, and in looking at what might be achieved in the years ahead. We need to see whether we are achieving the fundamental purposes of the Clean Air Act. Are we adequately protecting the pub- lic health? In particular, are we meeting the special needs of the old, the young, and those already compromised by disease pro- cesses or by other biological processes that we are only just beginning to under- stand? In the early days of program implemen- tation, the answers to these questions seemed easy. Anyone familiar with the famous London fog of 1952 that ultimately killed 4,000 people, or with the "incident" at Donora, Pennsylvania, in which another 48 people died, knew that air pollution at high levels causes disastrous health prob- lems. We also knew that automobiles were significant contributors to the air pollution problem because of seminal work done in the Los Angeles basin. On any hot day in Washington in the late 1960s, the Senate Environment and Public Works Commit- tee knew without asking that automotive emissions just could not be good. The initial results of control were visible. Air pollution alerts decreased (independent of any changes in how these alerts were defined), and auto emissions came down. Hydrocarbons, carbon monoxide, and ox- ides of nitrogen have all been significantly reduced. After much struggle, we have also made considerable progress in reducing lead emissions from autos. In noteworthy contrast, we have not sought to control heavy-duty truck and bus emissions until more recently because of energy concerns. Accordingly, many of us still complain bitterly about these emissions, especially if we are behind a vehicle that is not well malntalnec ,. Current Problems for Motor Vehicles The air is cleaner. Not surprisingly, as progress was made, public pressure about motor vehicle emissions declined. Because of the determined efforts of many scientists and environmentalists, the leaded gasoline issue has retained a high profile, and the government took strong action during 1985. Ozone control is still a vexing prob- lem, from both a health and an environ- mental perspective, and no strategy for control is without substantial practical problems for industry or state and local governments. Inspection and maintenance, and associated tampering, also appear on the public agenda from time to time, and we have not yet fully explored the health ramifications of the diesel engine. For the most part, however, other air pollution and environmental issues have displaced mobile sources at the top of our environmental policy agenda. Problems of hazardous waste, radon, acid rain, and pes- ticides now seem more pressing. This is no
Kennedy and Grumbly 5 sign of failure, but of success, despite the presence of a revisionist movement that castigates the efforts at emissions control as a case of regulatory overkill. But, to paraphrase the famous question from another context, "What do we know, and when did we know it?" A variety of commentators concerned about control policy, regardless of their political persua- sion, quickly dispose of the science of the health effects of auto emissions, citing the inherent problems of dealing with uncer- tainty in assessing benefits. Science in this, as in some other policy arenas, often seems like the poor cousin- one to be quickly introduced and just as quickly sent away, lest the rest of the family be embarrassed. This treatment is often meted out by law- yers, economists, and various other profes- sionals who somehow assume that the rest of the risk assessment/risk management . . . equation IS more certain. The hard fact still is, of course, that we don't know how successful our air pol- lution and motor vehicle emission con- trol policies are from a public health per- spective. Stockman's question at the open- ing of this preface is still the relevant one. We have reduced acute health problems through our regulation, but we cannot be sure how much. We know that the policies have had effect, but we cannot assume that biological damage reduction is proportion- ate to pollution reduction, particularly as we learn more about biological systems. And all of our modes of analysis, whether they in- volve experimentation with people or ani- mals, epidemiology, or in vitro laboratory work, are still imperfect. Given the compar- ative expenditures in control technology and relevant environmental health research, how- ever, it is surprising that we know so much about the health benefits of air pollution control. There is obviously nothing we can do about past policy, but a great deal that we can do about policy for the future. Importance of Continuing Health Research The question inevitably arises about the value of additional expenditures to find out whether auto emissions and other air pol- lutants are, in fact, risks worth worrying about. Even if we take it for granted that current emission control regulations have reduced the most serious acute health ef- fects problems, why is research still impor- tant? The answers flow directly from the social context as we just described it. Regulations: Form or Substance? If it is true that the structure and some of the imperatives of the current Clean Air Act sprang significantly from an exercise in "symbolic politics," that is, a political sit- uation in which positive societal action required the construction of a stage on which good and evil played clearly defined roles, then it is now important to see whether rational inquiry conducted in a less-heated time supports the symbols. This is important not only as an exercise in political science and evaluation, but as a step in rebuilding a base of public trust. We Americans are just now emerging from two decades of internal conflict. Perhaps we now have the luxury of reexamining the choices we made in more turbulent times. And, as a matter of the continuing credi- bility of the government, the need for intellectual integrity alone justifies a con- tinuing examination of the reasons we have undertaken a cleanup costing billions of dollars. We need to see if our past choices were right or just expedient. 7 Safety Margins and Susceptible Groups The Clean Air Act is quite specific in setting out the need for adequate margins of safety in establishing standards. The debate in the Congress, and subsequent reports from the relevant congressional committees, also makes it clear that these safety margins need to take into account so-called susceptible populations. Neither in the statute nor in the legislative history is the concept of susceptible populations ade- quately described. The courts and policy makers are left to their own devices in interpreting the term. From a standard- setting process, then, identifying and deal- ing with groups of people who may be either more susceptible to disease from any
6 The Social Context of Automotive Emissions Research given dose of pollutant, or who begin to exhibit "health effects" at levels below those seen in the general population, are critical. Indeed, it can be said to drive the regulatory system with respect to criteria pollutants, and it can only be dealt with through continuing research. Complicating the problem, of course, is that the current "revolution in biology" was not envisioned by the framers of the amendment. Accordingly, we are develop- ing a range of techniques that can show, even at the subcellular level, the effect of a particular dose of toxic pollutant. We do not, however, yet know how to translate much of this knowledge into terms that can help us quantify risk. But to leave the science at its current stage of development, or simply to leave to chance its capacity to generate information that can help regula- tion, would be to hinder the regulator in an already difficult task. Accordingly, this commandment to seek out and deal with susceptible populations forms a fundamen- tal scientific driving force in the implemen- tation of the Clean Air Act. Costs and Benefits From a practical perspective, it is important to understand whether additional invest- ments by taxpayers and consumers are nec- essary to further reduce emissions of the so-called criteria pollutants. Many econo- mists (including those with biases toward environmental regulation) now believe that we are in a condition of declining marginal benefits for each dollar invested, and that we have been in this condition for some time. If that is right, then our society should be thinking hard about reallocating motor vehicle emission control dollars to other problems in risk reduction where the return will be greater. The only practical way to approach this issue is through re- search investments designed explicitly to quantify the risks to human health from pollutants at levels that approximate cur- rent use, or that are likely to exist without additional control in the next decade. That this research investment has not already been made is undoubtedly a function of several things. These include the absence of the scientific tools with which to make major progress, and to put the matter plainly the reluctance of government and industry to move the issue away from symbolism and into reality. Fortunately, the tools are being developed, and as we describe below, the attitudes of many for- mer combatants seem to be changing for the better. We would not argue, of course, that uncertainty about risk can ever be elimi- nated through science. To say this, how- ever, is different from saying that we can- not reduce uncertainty into the range in which rational policy makers could have more confidence in their control and re . . . source Recisions. Emergence of a New Regulatory Structure in Mobile Source Emission Regulation The 1970 Clean Air Act amendments di- rected special attention to a particular group of pollutants, the so-called "criteria pollutants." Given the technology of 1970, this was arguably appropriate. Atmo- spheric chemists and biologists have known for some time, however, that our approach to understanding and controlling air pollutants, including motor vehicle emissions, is too simple. Not only are many more products actually emitted than are currently controlled, but atmospheric and biological reactions multiply signifi- cantly the numbers and the complexity of the ultimate pollutants that hit the "target" organ. We say this not to downplay the need to control "criteria" pollutants, but to emphasize the importance of using those parts of the law that focus on noncriteria or "unregulated" pollutants. We live in a world of complex chemical mixtures, and new technology or fuels further complicate that mixture. For the Environmental Protection Agen- cy, these new facts may mean that we are about to enter an era in which the EPA begins to look more like its older regula- tory cousin, the Food and Drug Adminis- tration. Because of what seemed at the time in 1977 to be rather minor amendments to the Clean Air Act, the EPA and the man
Kennedy and Grumbly 7 ufacturers now have the responsibility to determine whether new emissions or the technologies that produce them will result in an "unreasonable risk" to the public health. The EPA will find itself more and more in the classic regulatory mode of premarket approval in an era in which quantitative risk assessment provides the standard of evidence. The EPA will only be able to quantify the risk of new mixtures or new technology in the presence of consid- erable health research information. It will be able to sustain its decisions only if such research is credible within the wider scien- tific and public policy community. Research and Comprehensive Risk Management The assessment of risk through the appli- cation of science will never be, nor should it be, the sole criterion for decision making. However, it seems clear that future corpo- rate leaders and regulatory managers will increasingly look at issues of comparative risk in deciding how to deploy limited resources. Just as important, it seems likely that we are also entering an era in which we are beginning to look at the risk of a substance from each of its sources before making control decisions, so that we max- imize control in a cost-effective way and do not transfer risks from one media to an- other. In the motor vehicle arena, this means looking closely at all the sources of nitrogen oxides, carbon monoxide, partic- ulates, formaldehyde, benzene, and other chemicals before implementing additional controls. The techniques to do these anal- yses are still lacking, and research is re- quired particularly in the critical risk assess- ment arena. Regulatory Credibility and the Maturation of Toxicology In the past several years, we have witnessed a determined attack on what some have called the "new social regulation," includ- ing many areas of environmental protec- tion. Indeed. it can be argued that a presi- dent of the United States was elected in part because of the success of these arguments. , ~ This attack goes well beyond arguments concerning the costs and benefits of any particular regulation, and to the core of whether much of the government's regula- tory behavior was inherently just. Science per se cannot deal with this latter question, but science can narrow the argument. Over time and with effort, science can substan- tially reduce the uncertainties in extrapolat- ing from animal and laboratory data to human experience. Ironically, then, science can change the debate from a technocratic one to one in which we really do grapple with the fundamental question of"how much control and regulation do we want as a society?" Our ability to understand and manipu- late basic genetic material offers the possi- bility of radically improving toxicology. It does this by providing ways in which we can compare the results of traditional ex- periments with results by some of the new methods. That does not mean that tradi- tional toxicology will be less important. Indeed, we will need more chronic testing in order to maximize the utility of some of the newer methods such as "computer modeling." To take advantage of new op- portunities will require, however, the long- term effort to bridge historic gaps between the basic and the applied biological sci- ences. It requires ways to bring together scientists of a variety of disciplines under mission-oriented banners. In short, it will require money as well as the right institu- tional structures. The Health Effects Institute and the Social Context This preface has briefly concentrated on some of the antagonisms, concerns, and possibilities that surround the science and regulation of auto emissions. The concern about credibility and the search for an institutional structure that could marry the basic and applied sciences formed the fun- damental bases for establishing the Health Effects Institute, the institutional frame- work from which this book has developed. In 1979, it had become apparent that
The Social Context of Automotive Emissions Research health research related to mobile source emissions needed to be organized in a dif- ferent fashion. There was simply too strong an adversarial tone; government- sponsored research, especially if it was done within the regulatory agency, was mistrusted on the grounds that it was aimed at regulatory outcomes, whereas in- dustry-sponsored research was seen as di- rected toward the exoneration of its own products. Fortunately, realization arrived almost simultaneously at the Environmen- tal Protection Agency and at the automo- tive industry. A group of motor vehicle and engine manufacturers and Douglas Costle, then Administrator of the EPA, asked Archibald Cox, Donald Kennedy, and William Baker to be the founding Directors of a new organization called the Health Effects Institute (HEI). It came into being in 1981 under the Executive Direc- torship of Charles Powers, and its research program was launched during 1983. The basic idea behind the HEI was sim- ple enough. It was to solve two problems: first, public mistrust of the sponsorship of research on the health effects of mobile source emissions; and second, the lack of involvement of the best academic scientists in that kind of research. Clearly the two problems are related. Most scientists want to work on interesting questions, but- all other things being equal would prefer that the answers yield useful outcomes. But, where controversy and suspicion sur- round the sponsorship and evaluation of research, the best scientists will exercise their option to busy themselves elsewhere. The HEI sought to change this situation by clearing the sponsorship of suspicion and finding ways to reengage the interest of first-rate people in the hope that others would follow. The basic concept, then, was that the HEI would draw research resources in equal measure from the regulatory agency and from regulated industry. It would then assemble groups of scientific leaders who would elicit proposals from good research organizations and make the awards. In that way, public confidence in the objectivity of the work could be established and main- tained. At the same time, the continuing availability of resources and credible spon- sorship would build interest and confidence within the research community, so that first-rate scientific groups would be pre- pared to make long-term commitments to work in the area. That would further rein- force public belief that the science could be trusted; and so on. The history of the HEI has, so far, borne out these expectations. The first step was to recruit a group of outstanding scientists for the two committees that would do the main work of the Institute: the Research Committee, responsible for the develop- ment of research objectives and the evalu- ation of proposals; and the Review Com- mittee, responsible for scientific peer review of the work done. Under the re- spective chairmanships of Walter Rosen- blith of the Massachusetts Institute of Technology and Robert Levy of Columbia University, these committees have been able to develop rosters of outstanding sci- entists, retain them, and above all en- gage their attention in a sustained way. Everyone who has served on policy com- mittees knows how difficult this last task is. The committee flies in to the meeting site, having (in many cases) done its homework on the airplane; over 36 hours many prob- lems are discussed and tentative conclu- sions reached; and then everyone goes home before a permanent memory trace has been established. This mode of scien- tific committee work, once termed BOG- SAT (the acronym for Bunch of Guys Sitting Around a Table), is fraught with hazards. But the HEI committees have largely been free of these, in large part because of the Institute's insistence on pur- chasing a greater commitment of time and energy from the committee members. In its equilibrium state, then, HEI seems to be working. It has an annual budget of approximately $6-7 million and over 40 research projects under contract. Only six- teen have been completed, but that is enough to give us confidence in the pro- cesses of review and evaluation. The Insti- tute is constantly being asked to do more, or different, things; and as is usual for any research field, priorities are constantly shifting. It is better not to let the setting of
Kennedy and Crumbly priorities fall entirely to external demand, because often the urgent preempts the im- portant and long range, and potentially comprehensive solutions may be missed. By 1985 then, the Institute had recognized that its own intellectual capital needed replen- ishment. Accordingly, we brought together a group of people from government, indus- try, and academia to write and critique pa- pers that would not only review the relevant fields, but that would also identify ap- proaches and a strategy that would most likely lead to improving our risk assessment capability. Only by improving this capability will we ultimately be able to answer Stock- man's opening question: What are the health effects of emissions? 9 The noted political scientist Aaron Wil- davsky has written, "There is no point in having good ideas, if they cannot be carried out." We believe that this book, far from being some "academic" exercise in research planning, represents another step in the social policy context of trying to get a good idea, that is, preventing health problems from air pollution, to work. This book is an important link in helping us determine whether further control is necessary, whether symbolic politics can be made real in our society, and whether the public should believe, as we do, that government intervention, so visibly represented by emissions regulation, can ultimately be jus- tified by the facts.
