EPA 560/3-76-001
ANALYSIS OF THE NATIONAL
ACADEMY OF SCIENCES' REPORT
DECISION-MAKING FOR REGULATING
CHEMICALS IN THE ENVIRONMENT
JANUARY 1976
Prepared by
OFFICE OF TOXIC SUBSTANCES
ENVIRONMENTAL PROTECTION AGENCY
WASHINGTON,D.C.
-------
EPA 560/3-76-001
ANALYSIS OF THE NATIONAL ACADEMY OF SCIENCES'
REPORT DECISION-MAKING FOR REGULATING CHEMICALS IN
THE ENVIRONMENT
January 1976
Prepared by
Office of Toxic Substances
Environmental Protection Agency
Washington, D.C.
-------
PREFACE
In July, 1975, the National Academy of Sciences, pur-
suant to a contract with the Environmental Protection Agency,
published a Report entitled Decision-Making for Regulating
Chemicals in the Environment. This Report included 34
specific recommendations for improving the decision-making
process.
The Office of Toxic Substances, with input from all
interested offices in the Agency, prepared the following
analysis of the 34 recommendations. This analysis was
undertaken to assist in further consideration of these
recommendations but is not intended to represent the views
and policies of the Environmental Protection Agency.
-------
ANALYSIS OF THE NATIONAL ACADEMY OF SCIENCES'
REPORT DECISION-MAKING FOR REGULATING CHEMICALS IN
THE ENVIRONMENT
General Comments and Recommendations
The 34 recommendations vary widely, from quite specific
recommended practices to broad general precepts. Those at
the latter end of the spectrum are useful in focusing atten-
tion on problem areas, but are somewhat difficult to evaluate.
Several specific recommendations, though sound, would involve
significant resource expenditures. Additional guidance as to
how they are to be implemented, and clarification of the resource
and policy implications, would be helpful. Recommendation 29,
urging use of monitoring systems to detect changes in concen-
trations of toxic substances in biological tissues and changes in
death patterns, is a case in point. These objectives are clearly
worthwhile, and absent competing claims on scarce resources should
be pursued. Without guidance as to what priority EPA should
assign to this recommendation within the framework of its
overall program and objectives, however, its utility is
limited.
Many of the recommendations are general in nature and
appropriate for almost any non-routine government decision
(e.g., deregulation (of crude oil prices). For example, recom-
mendations 9-16, dealing with openness and access to the decision-
making process, fall in this category. By and large the
recommendations are more useful to the extent that they
specifically address issues peculiar to regulating chemicals.
A primary problem is selecting those decisions to which
these recommendations should apply, since the resources
involved clearly preclude across-the-board application. The
Report suggests limiting application to non-routine decisions,
defined as those requiring the direct policy involvement of
the agency-head. But the practical question remains: viz.,
how to determine which decisions require the Administrator's
personal involvement. It would be most helpful if the
Academy, in its current study, could recommend specific
criteria or methods to (1) differentiate minor from major
decisions, (2) establish gradations between these extremes,
and (3) define the appropriate decision-making approaches to apply
at each gradation. In defining these approaches, it would be
useful to identify who within the Agency should be making what
decisions.
An additional problem is that generalized precepts may
conflict with one another. In the present report, for example,
-------
- 2 -
recommendation 3 calling for revision of. existing statutes to
permit "consideration of any relevant factors in the decision-
making process", seems somewhat inconsistent with recommenda-
tion 4 urging "increased and consistent statutory guidance as
to the relative importance that should be given to health,
environmental and economic factors". Another example is the
apparent contradiction between the likely effect of
recommendation 16, requiring that all ex parte communications
be made public, and the objective, sought by recommendation
12, of encouraging inputs from all affected parties.
Recommendations 1, 9 and 30-32 are among the most important
elements in the Report. Recommendation 1 is based on the sound
principle that the costs of showing a net value to society
resulting from a new chemical use should, when such a demonstra-
tion is required, be borne by those who will benefit from
production or use of the chemical. Federal regulatory action
in this area is justified in that the operation of the free
market may fail to equitably distribute the costs and benefits.
For example, persons who incur a cost, or risk, pursuant to the
production of a chemical may not be adequately recompensed by
the benefits accruing from the commercial production and use
of the chemical. On the other hand, such Federal regulatory
activity may unduly discourage research and development, product
innovation and other essential economic ventures, and therefore
any legislation placing the burden of proof upon the producer-
to-be of a new chemical should require that the regulatory
agency set forth by regulations and guidelines the type of data
which must be submitted to the agency to determine if a chemical
should be withheld from the market. Such guidance to potential
producers will allow better estimates of the time necessary for
market approval and also will help prevent the imposition of
unnecessary data requirements for individual chemicals.
The Agency is taking steps to incorporate the valuable
principles set forth in recommendations 9 and 30-32 into
the decision-making process. Certainly all aspects and
impacts including health, social, international and
economic should be explicitly incorporated and considered
in- any decision to regulate chemicals.
In this regard recommendation 30 is perhaps misleading
in its statement depreciating the utility of "highly
formalized methods of benefit-cost analysis". This seems
to conflict with the sense of recommendations 31 and 32, the
excellent discussion and the definition of benefit-cost
-------
- 3 -
analysis provided on page 39 of the Report:
"Benefit-cost analysis ... refers to the systematic
analysis and the 'evaluation of alternative courses
of action drawing upon the analytical tools and
insights provided by economics and decision theory.
It is a framework and a set of procedures to help
organize the available information, display
tradeoffs, and point out uncertainties. In this
way benefit-cost analysis can be a valuable aid;
but it does not dictate choices, nor does it replace
the ultimate authority and responsibility of the
decision maker."
The above suggests that a rewording of recommendation 30 would
more clearly state the intent i.e., not that benefit-cost
analyses seldom can be used for making decisions, but that
such methods cannot, in and of themselves, mechanistically
make the decision.
The discussion on pp. 46-50, advocates making explicit
the tradeoffs that any regulatory decision implicitly makes,
including those between dollars and human life. Consideration
will be given to the possibilities of achieving such explicit
valuation of noncommensurables in the standard-setting process.
Over and above any cost-benefit analytical advantages, adherence
to this procedure should enhance both the quality and consis-
tency of EPA's diverse regulations.
A mild demurrer is offered to recommendations 9-14 regarding
compiling, organizing, summarizing and arraying the background
information and cost-benefit data for regulatory actions and
presenting the various options for the decision-maker. Numerous
formal and informal mechanisms exist to achieve openness in
decision-making: Advance Notice of Proposed Rulemaking, Working
Groups with State and local government participation, solicitation
of comment from affected and knowledgable parties, OMB Circular
No. A-85 (providing for consultation with heads of State and local
governments in development of Federal regulations), Interagency
Review prior to proposal, public hearings and notification, Notice
of Proposed Rulemaking with subsequent public comment periods of
30 to 60 days, Environmental Impact Statements, Inflation Impact
Statements, Statements of Basis and Purpose, publication or
notice of availability of Criteria or Development Documents and,
finally, virtual daily contact with public interest and environ-
mental groups. Given these extensive procedures the recommenda-
tions cited raise the following questions:
1. How many additional resources should EPA divert
to this effort, and from where?
-------
- 4 -
2. How much additional delay in the standard-setting
process should be countenanced to further
accommodate these objectives?
3. At what point does aggressively educating and
obtaining support from the affected public become
propagandizing?
4. In taking extraordinary steps to solicit comment
from some affected parties how can the Agency avoid
discriminating against, and being charged with
discrimination by, others not so notified?
Other philosophical questions regarding this group of
recommendations derive from the underlying principle set forth
on page 23: >
"... the decision process is equitable when the
consideration given to the interests of potentially
affected individuals is proportional to the antici-
pated effects of the decision on those individuals.
In other words, the groups represented in the
decision process should reflect the groups in
society impacted by the decision and the degree
of impact of the decision on those groups."
Aside from the practical problem of identifying parties to
a decision and the degree to which each is impacted, this
principle is subject to various, perhaps unintended,
interpretations. Does it mean, for example, that more per
capita consideration should be given to a few shareholders of a
company that will be heavily impacted economically, than to the
many residents of the plant's community whose health may be
somewhat adversely affected? Or, conversely, that the health
interests of an extremely small and vulnerable segment of the
population be stressed over a very large economic cost spread
over the population as a whole? Perhaps the Academy's
current study could reexamine this issue to (1) determine the
feasibility of EPA attempting to assess the degree of potential
impact of proposed regulations in order to differentially weight
the consideration given diverse opinions, and (2) provide
specific guidance on how to go about it.
The Report might have given more consideration to the
programmatic or procedural framework necessary to provide the
informational wherewithal for sound decisions. The Agency's
efforts would profit from an outside investigation in this
area. How can relationships be improved within EPA between
(1) the decision-maker, (2) the regulation writers, (3) the
-------
- 5 -
data gatherers, (4) the litigators, and (5) the enforcers?
How can research programs be planned and instituted so as to
provide the requisite information when it is needed to develop
sound, defensible regulations? What offices should be involved
in defining these research programs and what institutional
arrangements should govern their interaction? How may it be
assured that substantive and procedural enforcement issues as
well as resources for enforcement, are adequately considered?
These and similar questions go to the very heart of institutional
decision-making and could be usefully addressed by the
current Academy effort,.
Another area which might be given more attention is how to
select the most reasonable approach for controlling a given
chemical problem from among the many options available. In
addition to the various Federal statutes FIFRA, FWPCA, CAA,
OSHA, FD6CA, CPSA, et al there are many State and local
approaches and non-regulatory options including jawboning,
publicizing the hazard, and reasoning with the source. Within
each statute, moreover, there are several authorities that can
be exercised. In practice a chemical control problem is rarely
approached from the perspective of the various available means
for controlo More typically the available statutory authority
determines the chemical hazard discerned i.e., problems are
selected to fit authorities rather than the other way around.
This too could be usefully explored in the current NAS study,
with an eye to recommending one or more specific strategies for
systematically evaluating alternative means of control.
Finally, a major problem in rationalizing the decision-
making process not addressed by the Report is how to ensure
that decision-makers actually want and use the prescribed
process. The Report ably discusses the problem of scientists
intruding their sociopolitical biases into their ostensibly
objective analyses, thereby usurping the function of the
politically accountable decision-maker to make the necessary
value judgments. There is also a common counterpart frailty
among some decision-makers namely, the willingness to
have the controversial value-laden decisions remain obscured
and buried in the form of extrapolation methods, "safety
margins", "application factors", "no threshold" assumptions,
reluctance to use uncertain data to assign dollar values to
human life, and so forth. It would be extremely useful, in
this regard, if the Academy would explore and present specific
ways and means of (1) ensuring that the analyses received by
the decision-maker are not only free of buried value judgments
but, in addition, quite explicitly highlight those political and
socioeconomic decisions that should be made, together with an
array of options, and (2) maximizing the likelihood that the
agency-head will use these analyses to make and make
explicit these decisions. These ways and means should include
-------
- 6 -
not only the essentially coercive methods suggested in the
present Report simultaneous release of the background
documents to the agency-head and the general public, man-
datory external review by outside experts, autonomy for the
Science Advisory Board, et al but also means of rendering
the consequences of making these politically difficult
decisions, if not pleasant, at least palatable. It would
seem preferable not to employ methods based upon diluting
the responsibility, such as panels or committees of
advisors, since political (as opposed to technical or
scientific) decisions emanating from such groups often tend
more toward the less controversial options than decisions
from a single, politically accountable, agency-head.
That concludes the general comments. Following are
specific comments on each of the 34 recommendations.
-------
Recommendation 1; As a general principle the burden of proof
that society will obtain a benefit should rest with those
proposing a new use of a chemical. It may be desirable to
make statutory changes to reflect this principle.
This important recommendation is based on the sound
principle that the benefits of using a chemical should outweigh
the risks, and the reasonable precept that, where it is
necessary to demonstrate that a net benefit will accrue from a
new chemical, or new use, the costs and other burdens so imposed
will be borne by those who stand to profit from the commercial
use proposed.
In distributing the burden, however, it should be kept
in mind that the "beneficiaries" of a new chemical use include
not only the producers but also the employees, wholesalers,
retailers and consumers involved, and these groups are
difficult to identify prior to commercial production and
impossible if production is disapproved. The Agency is con-
sidering ways to determine the cost-benefit information neces-
sary to support various decisions which will include evaluating
these costs and the roles and interests of the parties involved.
The open market may fail to equitably distribute the true
costs of commercial use of a chemical, clearly justifying
Federal regulatory action, however research and development,
product innovation or other essential economic ventures must
not be unduly discouraged in the process. To this end any
legislation imposing upon the producer-to-be the burden of
demonstrating the desirability of a new chemical (which burden,
in the nature of things, can evolve into a series of require-
ments to prove negative hypotheses) should obligate the
regulatory agency to establish, by regulations and formal
guidelines, the scope and cost of the evidence it will require.
This will allow the producer to better estimate the time the
agency will take to reach a decision and will preclude
imposition of arbitrary or onerous requirements in specific
cases.
Finally, the recommendation might be reworded to
explicitly include the initial use of new chemicals as well
as the new use of existing chemicals.
-------
- 8 -
Recommendation 2; Once the government has made a reasonable
case that the challenged use of an existing chemical creates
an excessive hazard to human health or to the environment,
the burden of producing evidence should shift to the propo-
nent of use, who must then make an appropriate showing that
the continued use is desirable.
This is a sound recommendation. It would be helpful if
NAS would define the key term "reasonable case" and indicate
what changes this recommendation implies for the current
procedures of FDA, CPSC, OSHA and EPA.
The word "excessive" might also be clarified, or perhaps
deleted. It is difficult to envision a situation in which
the proponent of use would be able to demonstrate the
"desirability" of any substance that creates an "excessive"
hazard. Generally, in such instances, the Agency would
invoke the "imminent hazard" provisions of one or more of
its authorities, in which case the burden of proof does tend
to shift to the proponent of use. Presumably a less clear-
cut situation is envisioned here, and deleting the word
"excessive", or substituting "unreasonable", would seem more
appropriate.
Recommendation 3; Statutory provisions should not preclude
consideration of any relevant factors in the decision-making
process. Those provisions that prevent such consideration
should be considered for possible amendment.
This important admonition is sound but appears somewhat
inconsistent with recommendation 4, calling for increased
statutory guidance as to the "relative importance" to be given
these "relevant factors". Inflexible legislative proscription,
on either the factors that may be considered or the relative
importance that may be assigned to those factors, is undesirable.
-------
- 9 -
Recommendation 4; Whenever possible, and without precluding
administrative consideration of relevant factors, Congress
should provide increased and consistent statutory guidance
as to the relative importance that should be given to health,
environmental/ and economic factors in regulating chemicals.
Clear statutory guidance is certainly desirable. However,
inflexible legislative priorities that impair the Agency's
ability to respond, in a changing environment, to the
exigencies of particular rulemaking situations are undesirable.
On the one hand the legislative intent should be clear; on
the other, Congressional legislation is often a remote and
inflexible vehicle for making judicious regulatory tradeoffs
among various competing and changing interests.
Congressional guidance should not necessarily be "consistent"
in any strict sense of the word. Different statutes have
different purposes, deal with different substances, and have
different frameworks. The relative weight assigned to
health, economic, and social considerations in regulating food
additives, for instance, could quite reasonably differ from
the relative emphasis assigned in a law to regulate industrial
effluents.
Finally, and as noted above, it may be difficult for
Congress, in practice, to give "increased ... statutory
guidance" without "precluding administrative consideration of
relevant factors" in the decision-making process.
-------
- 10 -
Recommendation 5; EPA should undertake a study to (1) identify
neglected areas in hazard or pollution control and (2) determine
whether existing legal authorities are available and unused or
whether new laws should be sought from Congress. Examples of
such .needs include the problem of choosing optimal waste dis-
posal methods considering the total environment and the question
of indoor air pollution.
This recommendation defines a major aspect of EPA's mission.
The Agency continuously reviews environmental problems and the
existing legislative framework to identify programmatic gaps
that can be filled either by more aggressive utilization of
existing authorities or by seeking new legislation. In
addition, EPA conducts a comprehensive annual review of its
regulatory framework for this purpose. Although the Agency, of
course, could improve its utilization of these reviews to better
anticipate problems, it seems unlikely that additional studies
would produce significant new information at this time.
Insofar as the two cited examples of legislative needs
are concerned, the proposal to seek Federal legislation to
regulate waste disposal has been deferred in accordance with
Administration policy concerning new spending authorities. It
is not clear that additional authority to regulate indoor
air pollution is necessary, in large part because the Consumer
Product Safety Commission, under the definition of "Consumer
Product" in Section 3(a)(l) of the Consumer Product Safety
Act and under Sections 7, 8, 9 and 12 of that Act, may already
have some authority in this area.
-------
- 11 -
Recommendation 6; Congress should review the adversary
procedures that have led to unduly protracted hearings
(such as pesticide cancellation) and determine how best
such marathon decision making can be shortened.
The amount of time consumed in such proceedings is a
major source of concern and all possible means to expedite
the process should be explored.
Congress is, in fact/ conducting a detailed review of
all aspects of the Agency's activities, including hearing
procedures. EPA has responded to a detailed questionnaire
from the House Special Subcommittee on Oversight and
Investigations and the results of this study might provide a
basis for the Academy's current analysis of EPA's operations,
Recommendation 7; Exposure to many chemicals comes simultane-
ously from two or more media. An interagency committee should
be established to ensure that each agency's standards reflect
appreciation of such multiple exposures.
The desirability of ensuring that standards are set
taking total exposure into account is clear. Recently EPA's
Office of Research and Development initiated a program to
produce STAR Documents, which emphasize all routes of expo-
sure, to assist in this process.
However, standing interagency committees are unlikely
to be particularly useful or vigilant in this regard. Actually
the necessary and sufficient institutional arrangements already
exist in the Interagency Review administered by the OMB; the
purpose would be served if steps were taken to ensure that all
agencies adhered to this established review procedure.
-------
- 12 -
Recommendation 8; Within the Committee there were divergent
opinions on tvhe optimum organizational structure for making
chemical regulatory decisions. It is recommended that Congress
give consideration to alternative decision-making structures.
In particular, it should consider whether responsibility
should be vested in a board or commission with fixed terms in
office or in a single administrator.~~
There are advantages and disadvantages associated with
each type of decision-making organization. Congressional
consideration of the issue may well be useful in further
clarifying the considerations involved.
Recommendation 9; The essential elements of decision making
should be part of the public record. The agency should publish
a "white paper" for each important regulatory action undertaken.
The paper should include the key details of the economic, legal/
scientific/ and other considerations taken into account in
reaching the decisions. It should be issued when the agency
decides to take an action but sufficiently in advance of a final
decision to permit considered response. An important decision
to take no regulatory action, or to defer such action/ should
be accompanied by a "white paper".
Since most if not all the essential elements of decision-
making are already part of the public record, in one form or
another, this will largely be a coordinative task. Consolida-
tion of these diverse documents into a single, comprehensible,
standard format, eliminating redundancy, is required. Indeed
this recommendation is so critical that steps are being taken
to ensure that its principles, along with those set forth in
recommendations 30-32, are incorporated into the formal rule-
making process.
-------
- 13 -
Recommendation 10; Any information available to an agency on
the hazards of a chemical that is regulated by that agency
should not be considered proprietary and thus should be avail-
able for public inspection in a timely fashion during and
after the decision-making process.
Such information is available to the public unless it is
deemed protected by Title 18 USC 1905, in which case it is
inviolable. This, in turn, is an ad hoc determination made in
accordance with the requirements set forth in the law in question.
Some laws prohibit disclosure of any information which, in
the Administrator's judgment, contains or relates to trade,
commercial or financial information. In contrast, Section 308
of FWPCA and Section 114 of CAA require that any information
obtained under those authorities be made available to the
public except upon a showing satisfactory to the Administrator
that such information would divulge methods or processes
entitled to protection as trade secrets. These latter sections
seem to strike an equitable balance between competing but
legitimate concerns, and are preferable.
Two other legislative changes would be helpful:
1. Vest authority in EPA to obtain information needed
to develop standards regardless of whether or not
the information source is to be subject to the
proposed regulation. While EPA has ample authority
under both FWPCA and CAA to obtain information
from sources that are the focus of the proposed
regulatory action, difficulties arise when they are
not.
2. Exclude, in statutes administered by EPA, all
narrowly defined (e.g., toxicological) health data
on chemicals currently in commerce from the pro-
tective umbrella of the Trade Secrets Act.
In any event EPA will still receive some information
legitimately protected under the Trade Secrets Act which is
relevant to regulatory decisions. Therefore, the Agency will
continue to set standards on the basis of information not com-
pletely shared with the public. There is no obvious way out of
this dilemma. It is unreasonable to deprive a company of its
legitimate competitive advantage, and it would be irresponsible
not to utilize all of the available, pertinent information.
-------
- 14 -
Recommendation 11; The early and open exchange of information
and opinions on a proposed decision should be encouraged to
reduce the current dependence on subsequent judicial challenge,
The EPA Administrator should hold public hearings at the
earliest feasible stages of the decision process. He also
should facilitate pre-hearing exchange of information among
parties (for example, through depositions/ interrogatories,
and other discovery procedures).
Such "early and open exchange of information" is certainly
desirable, although there is no assurance that this will "reduce
the current dependence on judicial challenge". The Agency
already goes to some length to ensure such information exchange
occurs and has adequate authority to obtain information.
Specific comments from NAS would be helpful regarding
the adequacy or focus of these efforts. If additional steps
are considered necessary, it should be noted that such pro- .
cedures could be used by those with an interest in impeding
rather than facilitating rule-making.
-------
- 15 -
Recommendation 12; At appropriate points in the decision-making
process the agency should actively seek to identify the affected
parties and solicit suggestions and comments from them. Ways
should be explored to better represent the interests of future
generations.
EPA already pursues this goal. To go beyond current
practice would require arrogation of the responsibility for
defining "affected" parties and the degree of impact. Also,
active identification and solicitation of comments from
affected parties would probably draw charges of favoritism
or discrimination from those whose suggestions and comments
were not sought. It would also attract disruptive political
pressures early in the decision-making process.
The Agency will consider including with the NPRM a list
of known affected parties, such as trade associations, which
might have an interest in commenting, but to actively, and
exclusively, seek such comments and suggestions would be
risky from a regulatory standpoint. Anyone whose particular
comments were not expressly solicited could make an arguable
case that implementation of the ensuing regulation should be
further delayed until he was given an equal opportunity to
comment. The whole point of the Federal Register notice is
to even-handedly provide notice.to all interested parties.
The system may not be perfect but the recommended solution
seems less desirable.
Future impacts of decisions may certainly be significant
and need to be addressed. However, in this rapidly changing
environment, it is extremely difficult to anticipate
technological or scientific breakthroughs, natural or man-
made catastrophes, or other such radical changes. Thus any
decision made could easily be rendered meaningless, no
matter what framework had been employed to accommodate future
generations. The Agency should consider chronic effects of
chemicals in any decision to regulate, however, methods
utilized to deal with the issue of the future social, economic
and health impacts of such decisions are severely limited; any
insights or guidance from the Academy would be helpful.
-------
- 16 -
Recommendation 13: The press as well as other interested parties
should be told when a standard-setting process begins and when
the proposed standard is ready for publication; further dis-
cussions during this process should occur as often as warranted.
This reasonable suggestion largely reflects current EPA
practice. To ensure it, the Office of Public Affairs should
routinely notify the press whenever a standard-setting process
is initiated. This event is rarely newsworthy, however, and
that fact prompts the following comment on the text supporting
this recommendation.
The extent and nature of press coverage of any toxic
"story" is often unrelated to an analysis of scientific,
toxicological or economic data but is, rather, a function
of the story's "news value" i.e., its dramatic appeal.
For this reason the press must be viewed as an uncertain
ally in the effort to mount a deliberate and considered
regulatory program.
In this same vein, the following passage from page 27 .
apparently misapprehends EPA's role:
"Regulation of toxic substances is, by its very nature,
a subject that will evoke sympathetic reactions from
the press. It is a story with a built-in advantage
for the government which is, after all, engaged in the
business of protecting the same public the press sees
itself as representing."
This suggests that EPA's position will and should coincide with
popular opinion and against the regulated industry. In point of
fact, however, EPA's responsibility and intention is, to the
maximum extent possible under the applicable statutes, to
strike the optimum balance among the relevant costs, risks
and benefits associated with the use of a chemical. In so
doing, the Agency will not necessarily reach the "popular"
position.
-------
- 17 -
Recommendation 14 r EPA and other agencies should initiate
programs aimed at training and encouraging citizens to
participate in the decision-making process.
The Agency does have an obligation to lay before the
public, as clearly and concisely as possible, the hazards,
costs and benefits associated with a range of alternatives,
and to listen with an attentive ear to the ensuing response
before reaching any major regulatory decisions.
However, there are three reservations concerning this
recommendation. First, the propriety of EPA's training
citizens to participate in the decision-making process is
questionable. Such programs tend to turn, in time, into
publicly-funded but agency-directed efforts to sway public
opinion.
Secondly, there is an inherent tension, if not conflict,
between the idea that "training and encouraging citizens to
participate in the decision-making process" is worthwhile, and
the suggestion in recommendation 17 that the quality of
chemical regulatory decisions is largely dependent upon the
adequacy of the scientific data base.
Finally, the statement on pp. 2'9-30 in the supporting
text that
II
many significant decisions have taken far too
long to make, but this has more often been due to
parties with a specific interest in delaying a
decision than to public participation ..."
highlights the fact that either party to the dispute,
against which the imminent decision seems to be going, will
generally resort to delaying tactics. This is the price of
open decision-making.
-------
- 18 -
Recommendation 15; The Department of Commerce should develop
an educational resource to help small businesses acquire the
information on chemical regulatory matters that is at present
routinely available to large corporations and major trade
associations.
The Department of Commerce might well be useful in this
area. Trade associations (which usually participate in the
review of standards proposed by EPA) also generally perform
this function and the DOC would need to take care not to
duplicate the efforts of these groups.
Recommendation 16; All ex parte communications, including those
from Congress, members of the Executive Branch, private corpor-
ations, and citizen groups, on any adjudicatory decision pending
before a regulatory agency should be made public with sufficient
time for comment before a decision is made.
Strictly speaking this recommendation is already mandated
by law -- i.e., ejc parte communications in adjudicatory pro-
ceedings must be made public. Presumably the intent here is
to apply this requirement to less formal proceedings and to
expand the definition of ex parte communications to include
communications from sources other than direct parties to the
dispute (e.g., Congress), and to persons other than the
"decision-maker" (e.g., to staff members of EPA). The crux
of the proposal is spelling out the precise coverage and
definitions intended so as to create a workable method that
does not unduly inhibit productive information exchange.
The objective of exposing to public scrutiny any pressures,
entreaties, or persuasion received from vested interests,
while laudable, works at cross-purposes with recommendation
12, designed to encourage comment and criticism from affected
parties. Many of the most affected and knowledgable sources
will be inhibited by the exposure sought here.
In any event, the recommendation addresses a serious and
legitimate concern. It would be useful if the Academy's current
study could sort out the definitions intended by the term
"ex parte communication", and provide specific workable methods
for attaining the objectives.
-------
- 19 -
Recommendation 17; The quality of chemical regulatory decisions
is dependent largely upon the adequacy of the available infor-
mation. To develop an adequate data base research efforts in
basic clinical and environmental toxicology and epidemiology and
in economic analysis must be strengthened" and professional
training in these areas must be supported. An interagency com-
mittee consisting of the relevant federal research and regulatory
agencies should be established to gain maximum use of existing
information.
It would be useful to document the conclusion, on page
52 in the supporting text, that:
"... society suffers large and unnecessary expense
because of inadequate investment in determining
the hazards of chemicals."
And similarly, for the statement on page 53, that:
"It usually will cost more to institute the more
stringent controls resulting from the larger
safety (ignorance) margin than it would to
improve the precision of the data base."
The validity of the recommendation hinges largely upon the
soundness of those arguments and, since implementation will
require substantial resources, they should be buttressed
with the supporting cost-benefit data.
The recommendation and supporting text apparently urge
increased Federal outlays to achieve these objectives. If so,
the recommendation conflicts with the thrust of recommendations
1 and 2, which would shift this burden to the private sector.
The Academy's present study might address this issue. There
are arguments for "internalizing" the costs of controlling
chemical hazards by creating a regulatory framework that
compels the private sector to assume the burden. There are
also counter-arguments for imposing these costs and
responsibilities on the general taxpayer. It wouldhelp to
obtain clarification from the Academy regarding which of
these policies should be pursued. If, indeed, a mix expand-
ing both approaches is desirable, advice on the relative
emphasis to be placed on each would be useful.
The proposal for another standing interagency committee,
in this case to gain maximum use of existing scientific and
technical information, is unappealing. Informal ad hoc
arrangements among the technically involved and qualified
personnel would be more effective.
-------
- 20 -
Recommendation 18; Existing toxicological and other information
systems related to the regulation of chemicals should be
examined with a view to improving their coordination and
utilization.
The Agency is supporting the current efforts of the
Department of Health, Education and Welfare directed to improving
systems concerning health effects information. In addition,
EPA is evaluating ways to better use its own data, and is
involved in international efforts to develop mechanisms for the
exchange and use of information on toxic chemicals.
Recommendation 19; For optimal regulatory decision making, a
procedure to conduct retrospective analyses of the impact of
given decisions should be adopted. This should emphasize
evaluation of the accuracy of the predictive models for health,
economics, and environmental sciences in the original decision-
making process. EPA should also develop a formal system that
reviews and identifies information necessary for optimal future
decision making and implements the appropriate programs to
generate this information in a timely fashion.
While most EPA offices informally do this to some extent,
the Agency does not have a formal procedure for retrospectively
assessing the consequences of major decisions. Clearly, in the
absence of such feedback, many opportunities to learn from
experience will be lost.
Guidance is needed on how to mount such follow-up
analyses and the priority they deserve. Would it be better,
for example, to assign the task to the individual units that
were responsible for the decision and its implementation in
the first place, on the assumption that any possible bias
stemming from self-evaluation would be more than compensated
by the high level of interest and competence in the subject
matter; or to a centralized and impartial, but relatively
uninformed and less interested "auditing" unit? What criteria
should be used to select those decisions requiring retrospective
assessment? How can it be assured that the results of such
analyses will actually be used to guide future decisions or
consider revision of past ones if necessary? How soon can
tangible results be expected from this process? How can the
probable output and limitations be defined to create realistic
expectations of these efforts? This is fertile ground for the
Academy's current study.
-------
- 21 -
Recommendation 20; All agencies that regulate chemicals should
establish mechanisms for external scientific review of the
technical data base presented to the decision maker. The re-
sults of the review should be available directly to the decision
maker and to the public as early as feasible. The major
research and development programs undertaken by EPA specifically
should be reviewed routinely by panels of qualified experts.
External review by outside experts is one important approach
to ensuring the validity of the scientific data base used to
reach regulatory decisions. Views of affected parties
during the public comment period following Notices of Proposed
Rule-Making are also often very effective.
Increased emphasis on review by external experts raises
a number of questions the Academy might address. Of paramount
importance is the question of how to select these experts in
such a way as to minimize, or at least equitably distribute,
the sociopolitical biases represented an issue of no less
concern here than with in-house EPA scientists.
Should the outside experts operate independently or
should they sit as a committee? If the reviews are prepared
independently, who should arbitrate in the case of conflicting
assessments, and within what framework? If they deliberate
and report as a committee, how should discrepancies between
their findings and those of the in-house scientists be handled?
On some occasions the outside review should not be made
available to the decision-maker and the public simultane-
ously. Often responsible rule-making will be better served
if the in-house staff has an opportunity to digest and react
to such reviews before they are made public.
When these reviews are necessary, and to the extent that
committees rather than individual assessments are used, ad hoc
groups or the existing standing committees are preferable to
new committees.
-------
- 22 -
Recommendation 21; EPA should make greater use of expert
committees that represent the spectrum of potential viewpoints,
particularly in helping to anticipate future chemical control
problems.
Many EPA Offices are concerned with anticipating future
chemical control problems. With regard to advisory committees,
better use of existing committees is preferable to additional
committees.
It is not clear exactly how advisory committees would
go about anticipating "future chemical control problems", and
advice in this area would be useful.
Recommendation 22; EPA's Science Advisory Board should be
permitted to formally request the Administrator to subpoena
information needed for the Board's consideration and to allow
the Board to investigate new areas. The request and the
Administrator's response should be a matter of public record.
The existence and responsibilities of the Science Advisory
Board should be mandated by legislation.
The Science Advisory Board was established to increase the
Administrator's options by providing, at his request, informal
advice from a broad spectrum of relevant interests and expertise.
This recommendation would thrust the Science Advisory Board into
a quasi-adjudicatory role, thereby nullifying the informal advi-
sory function, and would constrict rather than expand the
Administrator's discretion. Thus, the Science Advisory
Board should not be permitted to formally request the Administrator
to subpoena information, and its existence and its responsi-
bilities should not be mandated by legislation.
-------
- 23 -
Recommendation 23; The power to subpoena expert witnesses,
now available to EPA under the Federal Environmental Pesticide
Control Act, should be extended to proceedings under other
statutes to develop the fullest and fairest public record.
The lack of such authority has not been overly troublesome,
but the suggestion is sound and should be supported.
Recommendation 24; To improve the quality of scientific
evidence, the right of counsel to cross-examine expert
witnesses should be guaranteed.
As formal proceedings do, in fact, include provisions
for cross-examination, it is assumed this proposal refers to
informal rule-making procedures. In these cases it could
diminish rather than enhance the quality of scientific evidence
generated. The inquisitorial nature of cross-examination
particularly the limitations placed on the mode and scope of
response while appropriate for retrospectively ascertain-
ing facts, is counter-productive for assaying the goodness of
prospective conjecture, which is generally the nature of the
testimony of expert witnesses at these proceedings. Adoption
of this recommendation moreover, would deprive these proce-
dures of the counsel of many able experts unwilling to
subject themselves to this form of inquiry.
-------
- 24 -
Recommendation 25: Other federal agencies should have the
opportunity to participate in internal EPA deliberations early
in the consideration of regulatory options. The same
opportunity should be given to EPA by other agencies that
regulate chemicals.
As noted earlier in another context, if all agencies
participated in the existing interagency review process
administered by the Office of Management and Budget there
would be ample coordination in the Federal standard-setting
process. Interagency coordination at an earlier stage should
(and generally does) occur on an informal basis among the
technical personnel.
Recommendation 26; Federal agencies regulating chemicals should
establish a formal and regular method for obtaining information
from state agencies.
The statement on page 27 in the supporting text that
"the general trend of environmental legislation and
- regulatory activities in the toxic substances area
is clearly away from the direct involvement of state
and local units of government and toward a
pre-eminent federal role"
is not entirely accurate. Large and increasing State and local
responsibility is envisioned under the Federal Clean Air, Water
Pollution Control, and Safe Drinking Water Acts. Insofar as
the specific recommendation is concerned, for example, Sections
110 and 111 of the CAA and Sections 208, 303, 305 and 516 of
the FWPCA establish formal mechanisms for the regular submission
of information from States and localities.
Generally speaking there is a reasonably free and
adequate flow of information to and from EPA and States and
local communities under existing formal and informal arrangements,
In those areas where improvement is necessary fewer, rather than
more, formal procedures may be in order. Adding more formal
procedures to the existing paper-work burden might dry up the
existing, extremely useful, informal flow.
-------
- 25 -
Recommendation 27; The Department of Health, Education, and
Welfare in conjunction with EPA, should attempt to develop a
hazard rating system, placing particular emphasis on evalua-
tion of use patterns.
More attention needs to be devoted to the task of develop-
ing criteria and procedures for assessing the relative risks
posed by various toxic substances. The notion of developing a
comprehensive "hazard rating system" to serve the diverse needs
of various programs within DHEW and EPA will be useful only to
the extent that these several programs are concerned with cor-
relative uses of, or exposures to, the same chemicals. This
is infrequently the case. A hazard rating system appropriate
for controlling pesticides, for example, would not serve well
the purposes of the Clean Air Act. Similarly, a hazard rating
system designed to facilitate implementation of the Food, Drug
and Cosmetic Act would have limited utility with the Federal
Water Pollution Control Act. Any rating system designed to
meet these multiple needs would serve none well.
While important, "use patterns" are only one of a number
of factors that should be incorporated in a hazard rating
system. Depending upon the regulatory purpose to be served,
one would also stress such variables as persistence, bio-
degradability, bio-concentration, bio-accumulation, intermedia
transport, levels of production, and synergism. It is the
differential weight that must be assigned to these various
factors to meet specific regulatory purposes, incidentally,
that makes a general scheme unworkable.
The Agency should improve its several hazard assessment
efforts, however, for the same reasons that are so ably set
forth in the Report's discussion of benefit-cost analysis.
Indeed, it is hard to see how such analyses can usefully
proceed without substantial improvement of EPA's ability to
assess and rank the relative risks presented by the universe
of toxic substances. Experience suggests that such hazard
evaluation systems are much easier to discuss than develop.
Any insights that the present NAS effort may be able to
provide in this area would be very useful.
-------
- 26 -
Recommendation 28; The Department of Health, Education, and
Welfare and EPA should establish a task force including, among
others, representatives of the chemical industry and the
scientific community to develop a system for making unpublished
or proprietary data about chemicals available to governmental
agencies.
A task force for this purpose is neither necessary nor
helpful. If Section 114 of the Clean Air Act or Section
308 of the Federal Water Pollution Control Act cannot obtain
the data sought, there is little a task force can accomplish.
What is really needed in this area is legislation such as the
proposed Toxic Substances Control Act, and the reporting
provisions it contains which would permit EPA to obtain
pertinent data from a source that is not the target of the
proposed regulations. Legislative amendments to the CAA and
FWPCA would also be helpful.
In the absence of any incentive to induce industry to
release proprietary data, there is little point to convening
representatives of the chemical industry, or anyone else, to
develop a system for making such data available. Unless and
until the Toxic Substances Control Act or similar legislation
is enacted it is unrealistic to assume that industry will
voluntarily release data that (1) were expensive to generate,
(2) may aid their competitors, and (3) will frequently be
used by the Government to impose unwanted restrictions.
-------
- 27 -
Recommendation 29: EPA in cooperation with the Department of
Healthy Education, and Welfare should develop and use monitoring
systems that can detect changing patterns in concentrations of
specific toxic substances in biological tissues. They should
also develop and use population surveillance systems that re-^
fleet changes in illness and death patterns due to environmental
pollutant exposure. Data from monitoring systems and from
other sources should be used to adjust past decisions/ if
necessary.
This is an excellent concept. Unless such efforts are
initiated now, and sustained, future efforts will be plagued
with the same information gaps that hamper current ones.
However, the recommendation may be difficult to implement. A
major obstacle is that such programs are so long-range that
the payoff falls beyond the "practical utility" timeframe of
any given set of decision-makers. This and the relatively
heavy expense involved give these proposals relatively low
priority in spite of their obvious importance. The Academy
would perform an extremely valuable service if it initiated
activities to generate the resources and commitment necessary
to develop these monitoring and surveillance systems.
It is not clear from the Report exactly how the results
of the population surveillance system or the epidemiologic
studies are to be related to "environmental pollutant expo-
sure". Given the problems associated with multi-pollutant
exposures, long latency periods between exposure and mani-
festation, and migration in and out of the target areas such
relationships are virtually impossible to establish.
Monitoring for concentrations of specific substances
in tissue should be an invaluable aid to both "early warning"
and standard-setting activities. Presumably this would
include tissues from both indicator organisms and humans.
The recommendation might also have included provision for a
tissue "data bank" to permit retrospective analyses, in
future years, of questions unanticipated at the time the
tissue was excised.
-------
- 28 -
Recommendation 30; Highly formalized methods of benefit-cost
analysis seldom can be used for making decisions about regulating
chemicals in the environment. Thus the development of such
methods should not have high priority. However, the benefit-cost
and decision frameworks described in this report can be useful in
organizing and summarizing relevant data on regulatory alterna-
tives which the decision maker must review.
It is true that /'highly formalized methods of benefit-
cost analysis" cannot, in themselves, mechanistically make the
decisions. However, the sense of the text, in apparent con-
trast to the first sentence of the recommendation, is that
such methods can, and should, in fact be used within the
decision-making process to identify and evaluate relevant data
This comes through clearly in the approbative definition of
"benefit-cost analysis" provided on page 39, together with
the "highly formalized method" recommended on pp. 44-46.
In this regard, the agency is considering ways to
develop and use methods which would, as is recommended in the
text of the Report, ensure that the evaluation of costs and
benefits implied in most decisions are made explicit and dealt
with in equivalent terms.
Recommendation 31; Value judgments about noncommensurate factors
in a decision such as life, health, aesthetics, equity, and
risk aversion should be explicitly dealt with by the politically
responsible decision makers and not hidden in supposedly objec-
tive data and analysis.
This is an excellent principle. It is desirable to
reduce the implied values used in reaching the decision
concerning the benefits and hazards to common terms when-
ever possible. As noted above, EPA plans to consider ways
of ensuring that this occurs in the course of its decision-
making process.
-------
- 29 -
Recommendation 32; The decision process should require the
agency's technical staff to present a full set of options with
a corresponding range of cost-benefit-hazard data and explicit
statements on the confidence limits of each analysis.
This recommendation helps to clarify the intended meaning
of recommendation 30. Clearly, such analyses should be
provided to the decision-maker by the agency staff. The Agency
is exploring ways of implementing this recommendation.
Recommendation 33; EPA should adopt, whenever scientifically
possible/ a generic approach, as opposed to. an ad hoc procedure,
for the regulation of chemicals.
If "generic approach" simply means grouping chemicals
in classes and, when attention is focused on any one in the
class, looking at the others to determine if they also
require (not necessarily the same) attention, then the
recommendation seems sound. If it means automatically
applying the same regulation to all chemicals in a class,
the approach is inappropriate.
-------
- 30 -
Recommendation 34; EPA should ask the American Bar Association
to cooperate in a study of the way health and environmental
information and economic analyses should be introduced and
considered in judicial proceedings.
In a judicial proceeding to review an Agency regulation
or adjudicatory decision, data and evidence in the record
are examined to determine if an adequate case has been pre-
sented to justify the action. The Courts do not judge the
technical or scientific appropriateness of an Agency action,
but rather determine only if it is supportable as a matter of
administrative law i.e., were the proper procedures
followed and was there sufficient evidence to make the
decision. As Federal Courts have exercised this responsibility
for-many years, the value of the study as proposed is
questionable.
However, the Agency does have a responsibility to hear
and evaluate all testimony and evidence prior to taking
action such that decisions are well-founded and, if need be,
legally defensible. EPA's fundamental task and goal is, to
the maximum extent provided under applicable statutes, to
develop standards and regulations which are environmentally.
sound and effect the optimum balance among the relevant
costs, risks and benefits. This requires a knowledge of
what health and environmental information, economic analy-
ses, assessments of social, international and intergene-
rational impacts, etc. are required under relevant authorities;
what burden-of-proof responsibilities are applicable; and
how such information may be obtained from other agencies,
outside experts and other interested parties, coordinated
and presented so as to be useful to the decision-maker,
intelligible to the public and sustainable before the Courts.
Therefore, studies to ensure that the necessary evidence is
adequately developed and justified during and within the
decision-making process could be useful.
------- |