[Senate Report 106-110]
[From the U.S. Government Publishing Office]
Calendar No. 219
_______________________________________________________________________
106th Congress Report
SENATE
1st Session 106-110
_______________________________________________________________________
REGULATORY IMPROVEMENT ACT OF 1999
__________
R E P O R T
of the
COMMITTEE ON GOVERNMENTAL AFFAIRS
UNITED STATES SENATE
together with
MINORITY VIEWS
to accompany
S. 746
TO PROVIDE FOR ANALYSIS OF MAJOR RULES, TO PROMOTE THE PUBLIC'S RIGHT
TO KNOW THE COSTS AND BENEFITS OF MAJOR RULES, AND TO INCREASE THE
ACCOUNTABILITY AND QUALITY OF GOVERNMENT
July 20, 1999.--Ordered to be printed
__________
U.S. GOVERNMENT PRINTING OFFICE
WASHINGTON : 1999
COMMITTEE ON GOVERNMENTAL AFFAIRS
FRED THOMPSON, Tennessee, Chairman
WILLIAM V. ROTH, Jr., Delaware JOSEPH I. LIEBERMAN, Connecticut
TED STEVENS, Alaska CARL LEVIN, Michigan
SUSAN M. COLLINS, Maine DANIEL K. AKAKA, Hawaii
GEORGE VOINOVICH, Ohio RICHARD J. DURBIN, Illinois
PETE V. DOMENICI, New Mexico ROBERT G. TORRICELLI, New Jersey
THAD COCHRAN, Mississippi MAX CLELAND, Georgia
ARLEN SPECTER, Pennsylvania JOHN EDWARDS, North Carolina
JUDD GREGG, New Hampshire
Hannah S. Sistare, Staff Director and Counsel
Paul R. Noe, Senior Counsel
Joyce A. Rechtschaffen, Minority Staff Director and Counsel
Lawrence B. Novey, Minority Counsel
Linda J. Gustitus, Minority Staff Director and Chief Counsel,
Permanent Subcommittee on Investigations
Carl Gold, Minority Congressional Fellow, Permanent
Subcommittee on Investigations
Darla D. Cassell, Administrative Clerk
C O N T E N T S
----------
Page
I. Purpose and summary..............................................1
II. Background and need for legislation..............................4
III. Legislative history and committee consideration.................17
IV. Administration views............................................19
V. Section-by-section analysis.....................................19
VI. Regulatory impact statement.....................................60
VII. CBO cost estimate...............................................61
VIII.Minority views..................................................66
IX. Changes in existing law.........................................83
106th Congress Report
SENATE
1st Session 106-110
======================================================================
REGULATORY IMPROVEMENT ACT OF 1999
_______
July 20, 1999.--Ordered to be printed
_______
Mr. Thompson, from the Committee on Governmental Affairs, submitted the
following
R E P O R T
[To accompany S. 746]
The Committee on Governmental Affairs, to which was
referred the bill (S. 746) to provide for the analysis of major
rules, to make the regulatory process more efficient and
effective, and for other purposes, having considered the same,
reports favorably thereon with amendments and recommends by a
vote of 11-5 that the bill as amended do pass.
I. Purpose and Summary
S. 746 is a bipartisan effort to achieve meaningful and
lasting improvements to the federal regulatory process through
important changes in the procedural requirements for issuing
federal regulations. S. 746 would subject all ``major rules''
to rigorous economic and scientific analysis before being
issued. By elevating the use of modern decisionmaking tools
such as cost-benefit analysis and risk assessment, the
legislation would promote more open, better-informed, and more
accountable regulatory decisions. Upon introduction of S. 746,
Senator Levin stated:
Those of us who believe in the benefits of regulation
to protect health and safety have a particular
responsibility to make sure that regulations are
sensible and cost effective. . . . I feel strongly that
this bill will improve the regulatory process, will
build confidence in the regulatory programs that are so
important to this society's well-being, and will result
in better, more protective regulations because we will
be directing our resources in more cost-effective ways.
1
---------------------------------------------------------------------------
\1\ 145 Cong. Rec. S 3481-82 (daily ed. March 25, 1999).
---------------------------------------------------------------------------
In the same vein, Chairman Thompson stated:
The Regulatory Improvement Act is an effort by many
of us who want to improve the quality of government to
find a common solution. . . . The supporters of this
bill represent a real diversity of political
viewpoints, but we share the same goals. We want an
effective government that protects public health, well-
being and the environment. . . . in the most sensible
and efficient way possible.
The Regulatory Improvement Act is based on a simple
premise: people have a right to know how and why
government agencies make their most important and
expensive regulatory decisions. This legislation also
will improve the quality of government decision
making--which will lead to a more effective Federal
government. And it will make government more
accountable to the people it serves. 2
---------------------------------------------------------------------------
\2\ Id. at S 3482.
Senator Voinovich, an original co-sponsor of S. 746,
---------------------------------------------------------------------------
stated:
The challenge facing public officials today is in
determining how best to protect the health of our
citizens and environment with limited resources. . . .
we need to do a better job of setting priorities and
spending our resources wisely. I believe that the
Regulatory Improvement Act will achieve these goals.
3
---------------------------------------------------------------------------
\3\ Id. at S 3485.
A brief synopsis of the major provisions of the bill
follows:
A. COST-BENEFIT ANALYSIS
Federal agencies would be required to perform a cost-
benefit analysis for major rules (rules imposing costs over
$100 million or having other material adverse effects). The
cost-benefit analysis would be done at the proposed and final
rulemaking stages and would include: An estimate of the
anticipated benefits of the rule (quantifiable and
nonquantifiable); An estimate of the anticipated costs of the
rule (quantifiable and nonquantifiable); An analysis of a
reasonable number of regulatory alternatives, including
flexible regulatory options; A reasonable determination: (1)
whether the benefits of the rule are likely to justify the
costs; (2) whether the rule is likely to achieve the rule
making objectives in a more cost- effective manner, or with
greater net benefits, than the other alternatives; and (3)
whether the rule adopts a flexible regulatory option.
If the agency determines that the rule is not likely to
satisfy these conditions, the agency shall explain the reasons
for selecting the rule notwithstanding such determination,
including identifying any statutory provision that required the
agency to select such rule, and describe any reasonable
alternative that would satisfy such conditions.
B. RISK ASSESSMENT
Agencies would be required to follow risk assessment
principles for: (1) major rules with the primary purpose of
addressing risks to health, safety, or the environment; and (2)
any risk assessment not related to a rule making that the OMB
Director anticipates is likely to have an annual effect on the
economy of $100 million or more.
To promote transparent and scientifically objective risk
assessment, agencies would be required to: identify and explain
significant assumptions made when estimating risks; notify the
public when the agency is conducting a risk assessment and
allow the public to submit relevant and reliable information;
and disclose relevant information about the risk, including the
range and distribution of the risk, including central and high-
end estimates, and the corresponding exposure scenarios for the
potentially exposed population and for highly exposed
subpopulations. When appropriate scientific information is
reasonably available, the agency would be required to compare
the risk being analyzed with other reasonably comparable risks
familiar to and routinely encountered by the public.
C. PEER REVIEW
Cost-benefit analyses for major rules that are anticipated
to have an annual effect on the economy of $500 million, and
risk assessments required by the Act, would be subject to
independent peer review, prior to issuance of a notice of
proposed rulemaking, if feasible. Only one peer review would be
required during a rule making.
D. JUDICIAL REVIEW
The legislation would provide for judicial review to ensure
that agencies conduct required regulatory analyses. The
regulatory analysis, including the cost-benefit analysis, cost-
benefit determination, and risk assessment, would be included
in the rulemaking record for purposes of judicial review, and
would, to the extent relevant, be considered by the court in
determining whether the final rule is arbitrary or capricious.
E. GUIDELINES, INTERAGENCY COORDINATION, AND RESEARCH
The Director of the Office of Management and Budget
(``OMB'') would consult with the President's Council of
Economic Advisors, the Director of the Office of Science and
Technology Policy (``OSTP''), and the relevant agencies to:
develop guidelines for cost-benefit analysis, risk assessment,
and peer review; improve agency analytical practices; and
arrange for research to improve regulatory analysis.
F. COMPARATIVE RISK ANALYSIS
OMB, in consultation with OSTP, would arrange for a study
to compare and rank health, safety, and environmental risks; to
improve methodologies for comparing various risks; and to make
recommendations on using comparative risk analysis to set
agency priorities for reducing such risks. Each relevant agency
would use the results of the study to inform the agency in the
preparation of its budget and strategic plans and performance
plans under the Government Performance and Results Act.
G. EXECUTIVE OVERSIGHT
OIRA would supervise and oversee implementation of the
requirements of this legislation and would systematically
review agencies' regulatory proposals, subject to public
disclosure requirements.
II. Background and Need for Legislation
Since 1946, the federal regulatory process has been guided
by the Administrative Procedure Act (``APA''), 5 U.S.C.
Sec. Sec. 551-558. The APA was enacted following the dramatic
delegation of discretionary authority to Executive Branch
agencies stemming from the New Deal. It has served for over 50
years as the blueprint for how agencies issue regulations.
With the rapid growth of complex and wide-ranging
regulatory programs since the late 1960s, the limited
procedures of the APA have been faced with new challenges. This
has moved the Committee over the years to review the adequacy
of the regulatory process. Since 1981, four comprehensive bills
have been reported by the Committee, though none of these has
been enacted into law. S. 746, the ``Regulatory Improvement Act
of 1999,'' is the latest product of the Committee's work and
experience in this area.
A. EXECUTIVE BRANCH ACTION ON REGULATORY REFORM
The Committee's concern about the adequacy and
effectiveness of the federal regulatory process has paralleled
a growing interest in centralized control and review by the
President. The assertion of presidential authority over the
rulemaking process began in 1971, when President Richard Nixon
established ``Quality of Life Reviews'' for certain U.S.
Environmental Protection Agency (``EPA'') regulations. Every
President since Richard Nixon has implemented executive
oversight of the regulatory process. President Gerald Ford
required agencies to conduct an inflationary impact analysis
for major rules. President Jimmy Carter established the
Regulatory Analysis Review Group to review important
regulations. He also required an economic impact analysis for
major rules under Executive Order 12044.
President Ronald Reagan implemented the most dramatic
reform over the rulemaking process when he issued Executive
Order 12291 in 1981. This was a significant extension of an
evolving centralized review process, and it required that all
rules be reviewed by the Office of Information and Regulatory
Affairs in the Office of Management and Budget before being
issued in proposed or final form. It also required that each
agency analyze the costs and benefits of each major rule and
that, to the extent permitted by law, agencies issue rules only
if the potential benefits of the rule outweighed the potential
costs. President Reagan also issued E.O. 12498 in March 1985,
directing agencies to prepare a yearly agenda of all
significant regulatory actions for the coming year. When he
took office in 1989, President George Bush continued President
Reagan's Executive Orders.
In 1993, President Bill Clinton replaced E.O. 12291 with
E.O. 12866, which continues the requirement for centralized
review of rules. E.O. 12866 applies only to ``significant
rules,'' not all rules, but it maintains the requirement for a
cost-benefit analysis of significant rules--primarily those
that have an annual effect on the economy of $100 million or
more--and it requires that, to the extent permitted by law,
agencies issue rules ``only upon a reasoned determination that
the benefits of the intended regulation justify its costs.''
Centralized regulatory review by the President, using OMB, is
critical to achieving the goals of this legislation: thorough
analyses of regulatory proposals, balanced consideration of
diverse viewpoints, effective coordination among agencies, and
a cost-effective regulatory system.
B. THE NEED FOR REGULATORY REFORM LEGISLATION
By any measure, federal agencies have long engaged in, and
continue to engage in, an enormous volume of regulatory
activity. In a 1997 report to Congress, OMB reported that there
are over 130,000 pages of federal regulations, ``with about 60
federal agencies issuing regulations at a rate of about 4,000
per year. . . . Federal regulations now affect virtually all
individuals, businesses, State, local and tribal governments,
and other organizations in virtually every aspect of their
lives or operations.'' 4 In recent reports, GAO
noted that the November 1998 edition of the Unified Agenda of
Federal Regulations contained 4,560 entries describing planned
or ongoing federal regulatory actions, 5 and that
federal agencies issued more than 11,000 final rules between
April 1996 and December 1998.6
---------------------------------------------------------------------------
\4\ Office of Management and Budget, Office of Information and
Regulatory Affairs, Report to Congress on the Costs and Benefits of
Federal Regulations (Sept. 30, 1997). The OMB report was required by
the Regulatory Accounting Amendment of Senator Ted Stevens, who was
then the Chairman of the Governmental Affairs Committee. The Stevens
Regulatory Accounting Amendment was modeled on the earlier and more
detailed regulatory accounting provision in S. 291, the ``Regulatory
Reform Act of 1995.'' The Stevens Amendment was contained in Section
645 of the Treasury, Postal Services and General Government
Appropriations Act, 1997 (Pub. L. 104-208), 1996 U.S.S.C.A.N. (110
Stat. 3009): 1088-89.
\5\ GAO, Regulatory Flexibility Act: Agencies' Interpretations of
Review Requirements Vary, GAO/GGD-99-55, at 19, April 2, 1999.
\6\ Statement for the Record of L. Nye Stevens, Director, Federal
Management and Workforce Issues, General Government Division, GAO,
before the Senate Governmental Affairs Committee, Federalism:
Implementation of Executive Order 12612 in the Rulemaking Process, GAO/
T-GGD-99-93, May 5, 1999.
---------------------------------------------------------------------------
The Committee is well aware of the importance of sensible
regulation in improving the quality of life for the American
people. Regulation can help achieve important social and
economic goals such as a clean environment, safe products, a
safe workplace, and reliable economic markets. Over the past 25
years, the nation has made tremendous progress protecting
public health, safety, and the environment and improving our
quality of life. We no longer have rivers catching fire. Our
air is cleaner.7 And American technology and
expertise is in demand around the world. But more challenges
lie ahead.
---------------------------------------------------------------------------
\7\ See Testimony of Carol M. Browner, Administrator, U.S.
Environmental Protection Agency, before the Senate Committee on
Governmental Affairs, S. Hearing 104-419, March 8, 1995.
---------------------------------------------------------------------------
Achieving the benefits of regulatory programs does not come
without cost. In recent reports, the annual cost of regulation
of the environment, health, safety and the economy has been
estimated on the order of several hundred billion dollars, with
the cost of ``social regulations'' (i.e., environmental,
health, and safety rules) making up approximately 75 percent of
the total.8 These costs are often passed on
indirectly to the American consumer and taxpayer through higher
prices, diminished wages, increased taxes, or reduced
government services.\9\ Although deregulation in the 1970s and
1980s reduced the burden of economic regulation, the total cost
of noneconomic or ``social'' regulation has been rising
substantially. At the same time, there is strong public support
for the benefits that well-designed regulations can produce.
---------------------------------------------------------------------------
\8\ See Office of Management and Budget, Office of Information and
Regulatory Affairs, Report to Congress on the Costs and Benefits of
Federal Regulations at 17-19 (1998); see also, Office of Management and
Budget, Office of Information and Regulatory Affairs, Report to
Congress on the Costs and Benefits of Federal Regulations (Sept. 30,
1997). Other studies, which include the full costs of paperwork and
economic transfers, estimate that regulation costs about $700 billion
annually. See, e.g., U.S. Small Business Administration, The Changing
Burden of Regulation, Paperwork, and Tax Compliance on Small Business:
A Report to Congress (Oct. 1995).
\9\ See, e.g., Resources for the Future, Public Policies for
Environmental Protection (Paul R. Portney, ed. 1990); Thomas D.
Hopkins, ``Cost of Federal Regulation'' 3, reprinted in Regulatory
Policy in Canada and the United States, Rochester Inst. Tech., (1992).
---------------------------------------------------------------------------
As the public demands better results while the costs of
regulation rise, the need for a smarter, more cost-effective
approach to regulation is more important than ever. The depth
of this need is not widely appreciated because the costs of
regulation are not as obvious as many other costs of
government, such as taxes, and the benefits of regulation often
are diffuse. But there is substantial evidence that the current
regulatory system often misses opportunities for greater
benefits and lower costs. As noted by the President's then-
chief spokesperson on regulatory policy:
Regrettably, the regulatory system that has been
built up over the past five decades * * * is subject to
serious criticism * * * [on the grounds] that there are
too many regulations, that many are excessively
burdensome, [and] that many do not ultimately provide
the intended benefits. \10\
---------------------------------------------------------------------------
\10\ Testimony of Sally Katzen, Administrator of OIRA, before the
Senate Committee on Governmental Affairs, S. Hearing 104-372, February
22, 1995.
The new challenges facing the regulatory system were not
envisioned by the drafters of the Administrative Procedure Act
some 50 years ago. While the APA has successfully adapted to
many changes in the regulatory process, it was not designed to
address the current regulatory landscape. Since the APA was
passed, the goal of much federal regulation has changed from
curbing monopoly power to addressing risks to the environment,
health, and safety; the form of most federal regulation has
changed from adjudication to informal rulemaking; and the scope
of federal regulation has vastly expanded from single
industries to economy-wide activity.
These dramatic changes have been accompanied by growing
problems that must be solved: agencies may fail to balance the
benefits and costs of regulations, fail to find flexible and
cost-effective solutions, or fail to consider unintended harms.
Moreover, the rulemaking process is not sufficiently
understandable to the public, nor is it as accountable as it
should be.
To date, cost-benefit analysis, so important to the
development of economically significant rules, has been
generally required only through executive order and not through
a statutory framework. There are no government-wide
requirements for conducting risk assessments. Much of the
analytical work of a rulemaking agency is done before the
public has the opportunity to comment, and both the policy and
scientific basis for the agency's choices are often unclear to
the public, through obscure and hard-to-read rulemaking files
or through the failure of the agency to make its thinking clear
and readily available to the public.\11\ While Executive Order
12866 sets out procedures intended to result in better rules
and enhanced public confidence, compliance is not uniform or
complete.\12\
---------------------------------------------------------------------------
\11\ See, e.g., Testimony of L. Nye Stevens, Director, Federal
Management and Workforce Issues, General Government Division, GAO,
before the Senate Committee on Governmental Affairs, S. Hearing 105-
335, September 12, 1997; Statement of L. Nye Stevens, Director, Federal
Management and Workforce Issues, General Government Division, GAO,
before the Senate Committee on Governmental Affairs, February 24, 1998;
GAO, Regulatory Reform: Changes Made to Agencies' Rules Are Not Always
Clearly Documented, GAO/GGD-98-31 (Jan. 1998); GAO, Cost-Benefit
Analysis Can Be Useful in Assessing Environmental Regulations, Despite
Limitations, GAO/RCED-84-62 (April 6, 1984).
\12\ See, e.g., GAO, Regulatory Reform: Agencies Could Improve
Development, Documentation, and Clarity of Regulatory Economic
Analyses, RCED-142 (May 1998); GAO, Regulatory Reform: Changes Made to
Agencies' Rules Are Not Always Clearly Documented, GAO/GGD-98-31 (Jan.
1998); Hearing before the Senate Committee on Governmental Affairs,
Subcommittee on Financial Management and Accountability, ``Oversight of
Regulatory Review Activities of the Office of Information and
Regulatory Affairs,'' S. Hrg. 104-825, 104th Cong., 2d Sess. (Sept. 25,
1996).
---------------------------------------------------------------------------
S. 746 seeks to address these problems. Central to that
effort is the use of accurate and thoughtful cost-benefit
analysis and risk assessment. We know that analyzing the costs
and benefits of regulatory proposals is no longer an
intellectual curiosity or academic exercise: it is a necessity.
In its 1997 Report to Congress on the Costs and Benefits of
Federal Regulations, OMB concluded:
[R]egulations (like other instruments of government
policy) have enormous potential for both good and harm.
Well-chosen and carefully crafted regulations can
protect consumers from dangerous products and ensure
they have information to make informed choices. Such
regulations can limit pollution, increase worker
safety, discourage unfair business practices, and
contribute in many other ways to a safer, healthier,
more productive, and more equitable society. Excessive
or poorly designed regulations, by contrast, can cause
confusion and delay, give rise to unreasonable
compliance costs in the form of capital investments,
labor and on-going paperwork, retard innovation, reduce
productivity, and accidentally distort private
incentives.
The only way we know how to distinguish between the
regulations that do good and those that cause harm is
through careful assessment and evaluation of their
benefits and costs. Such analysis can also often be
used to redesign harmful regulations so they produce
more good than harm and redesign good regulations so
they produce even more net benefits. \13\
---------------------------------------------------------------------------
\13\ Office of Management and Budget, Office of Information and
Regulatory Affairs, Report to Congress on the Costs and Benefits of
Federal Regulations (Sept. 30, 1997), at 10.
Current practices in this regard need significant
improvement. In 1996, Robert Hahn of the American Enterprise
Institute published one of the most comprehensive analyses of
the benefits and costs of recent environmental, health, and
safety regulation.\14\ Hahn concluded that about half the final
rules analyzed in the study would not pass a cost-benefit test.
Hahn's study also showed that the quality of federal agency
cost-benefit analyses varies widely ``from very poor to very
good'' and that we could ``realize significant gains by more
carefully targeting regulations.'' In 1997, Richard
Morgenstern, an EPA official on leave with Resources for the
Future, published a thorough analysis of 12 major rules from
EPA subject to economic analysis.\15\ Morgenstern concluded
that the economic analyses helped reduce the costs of all 12 of
the rules and, at the same time, helped increase the benefits
of five of them. Studies by the U.S. General Accounting Office
(``GAO'') have echoed such findings.\16\
---------------------------------------------------------------------------
\14\ Robert W. Hahn, ``Regulatory Reform: What Do the Government's
Numbers Tell Us?,'' in Risks, Costs, and Lives Saved, (Robert W. Hahn,
ed. 1996). See also, Testimony of Robert W. Hahn before the
Subcommittee on Financial Management and Accountability, Senate
Committee on Governmental Affairs, S. Hearing 104-825, September 25,
1996.
\15\ Resources for the Future, Economic Analyses at EPA (Richard D.
Morgenstern, ed. 1997).
\16\ See, e.g., Statement of L. Nye Stevens, Director, Federal
Management and Workforce Issues, General Government Division, GAO,
before the Senate Committee on Governmental Affairs, February 24, 1998;
Testimony of L. Nye Stevens, Director, Federal Management and Workforce
Issues, General Government Division, GAO, before the Senate Committee
on Governmental Affairs, September 12, 1997; GAO, Cost-Benefit Analysis
Can Be Useful in Assessing Environmental Regulations, Despite
Limitations, GAO/RCED-84-62 (April 6, 1984).
---------------------------------------------------------------------------
There is broad support for reforming the regulatory process
and the tools to accomplish that goal, including cost-benefit
analysis, market-based mechanisms, risk-assessment, and
comparative risk analysis. This support comes from diverse
sources, such as the National Research Council,\17\ the Harvard
Center for Risk Analysis,\18\ the American Enterprise
Institute,\19\ the Brookings Institution,\20\ the Clinton
Administration,\21\ Justice Stephen Breyer,\22\ the Carnegie
Commission,\23\ Resources for the Future,\24\ and other think
tanks, commissions, and independent scholars throughout the
country.\25\ The strong record on the need for regulatory
reform and the tools to achieve it has contributed to this
legislation.
---------------------------------------------------------------------------
\17\ See, e.g., National Research Council, Understanding Risk:
Informing Decisions in a Democratic Society (1996); National Research
Council, Science and Judgement in Risk Assessment, National Academy
Press, Washington, D.C. (1994); National Research Council, Issues in
Risk Management, National Academy Press, Washington, D.C. (1993);
National Research Council, Valuing Health Risks, Costs, and Benefits
for Environmental Decision Making, National Academy Press, Washington,
D.C. (1990); National Research Council, Improving Risk Communication,
National Academy Press, Washington, D.C. (1989); National Research
Council, Risk Assessment in the Federal Government: Managing the
Process, National Academy Press, Washington, D.C. (1983).
\18\ See, e.g., Harvard Center for Risk Analysis, Reform of Risk
Regulation: Achieving More Protection at Less Cost (March 1995); John
D. Graham, ``Making Sense of Risk: An Agenda for Congress,'' in Risks,
Costs, and Lives Saved (Robert W. Hahn, ed. 1996); The Greening of
Industry: A Risk Management Approach, Harvard University Press (John D.
Graham & Jennifer Kassalow Hartwell, eds. 1997).
\19\ See, e.g., American Enterprise Institute & Brookings
Institution, An Agenda for Regulatory Reform (Robert W. Hahn & Robert
E. Litan, eds. 1997); American Enterprise Institute & Brookings
Institution, Improving Regulatory Accountability (Robert W. Hahn &
Robert E. Litan, eds. 1997); American Enterprise Institute, The
Annapolis Center & Resources for the Future, Benefit-Cost Analysis in
Environmental, Health, and Safety Regulation (1996); American
Enterprise Institute, Benefit-Cost Analysis of Social Regulation: Case
Studies from the Council on Wage and Price Stability, Washington, D.C.,
(James C. Miller & Bruce Yandle, eds. 1979); M.J. Bailey, Reducing
Risks to Life: Measurement of Benefits, American Enterprise Institute,
Washington, D.C. (1980); Robert W. Hahn & J.A. Hird, ``The Costs and
Benefits of Regulation,'' 8 Yale J. on Reg. 233 (Winter 1991).
\20\ See, e.g., Lester Lave, The Strategy of Social Regulation,
Brookings Institution, Washington, D.C. (1981); Lester Lave,
Quantitative Risk Assessment in Regulation, Brookings Institution,
Washington, D.C. (1982); Robert W. Crandall, Controlling Industrial
Pollution: The Economics and Politics of the Clean Air Act, Brookings
Institution, Washington, D.C. (1983).
\21\ Office of Management and Budget, Office of Information and
Regulatory Affairs, Report to Congress on the Costs and Benefits of
Federal Regulations (Sept. 30, 1997), at 2 (cost-benefit analysis
significantly enhances the consideration of alternative approaches to
achieving regulatory goals, ultimately producing more benefits and
fewer costs); National Performance Review, Creating a Government that
Works Better and Costs Less, Washington, D.C. (1993); National
Performance Review, Improving Regulatory Systems, Washington, DC (Sept.
1993).
\22\ Stephen Breyer, Breaking the Vicious Circle: Toward Effective
Risk Regulation, Harv. Univ. Press, Cambridge, MA (1993); Stephen
Breyer, Regulation and its Reform (1982).
\23\ Carnegie Commission on Science, Technology, and Government,
Risk and the Environment: Improving Regulatory Decisionmaking,
Washington, D.C. (June 1993).
\24\ J. Clarence Davies & Jan Mazurek, Pollution Control in the
United States, Resources for the Future (1998); Paul R. Portney, Public
Policies for Environmental Protection, Resources for the Future (1990):
Paul R. Portney, ``Economics and the Clean Air Act,'' 4 J. Econ.
Perspectives 173 (Fall 1990); Resources for the Future, Worst Things
First?: The Debate Over Risk-Based National Environmental Priorities,
Washington, D.C. (Adam N. Finkel and Dominic Golding, eds. 1994).
\25\ See, e.g., Cass R. Sunstein, ``Congress, Constitutional
Moments, and the Cost-Benefit State, 2 Stan. L. Rev. 247 (1996); Cass
R. Sunstein, ``Health-Health Tradeoffs,'' 63 U. Chi. L. Rev. 1533
(1996); Cass R. Sunstein, After the Rights Revolution, Harv. Univ.
Press, Cambridge, MA (1990); National Academy of Public Administration,
Resolving the Paradox of Environmental Protection: An Agenda for
Congress, EPA & the States, (Sept. 1997); Enterprise for the
Environment, The Environmental Protection System in Transition: Toward
a More Desirable Future (Jan. 1998); Marian R. Chertow & Daniel C.
Esty, Thinking Ecologically: The Next Generation of Environmental
Policy (1997); Murray L. Weidenbaum, Business and Government in the
Global Marketplace, Prentice Hall, Englewood Cliffs, NJ (5th ed. 1995);
W. Kip Viscusi, Fatal Tradeoffs: Public and Private Responsibilities
for Risk, Oxford Univ. Press, NY (1990), See also, Administrative
Conference of the United States, ACUS Recommendation 85-2, ``Agency
Procedures for Performing Regulatory Analysis of Rules'' (1985); ACUS
Recommendation 88-9, ``Presidential Review of Agency Rulemaking (1988);
ACUS Recommendation 93-4, ``Improving the Environment for Agency
Rulemaking'' (1993).
---------------------------------------------------------------------------
c. governmental affairs committee action on regulatory reform through
the 105th congress
The Committee has been involved in overseeing the
regulatory decisionmaking process for over two decades. Through
a variety of studies, hearings, legislative proposals, and
oversight of the regulatory process, the Committee has
developed a broad expertise on the strengths and weaknesses of
the regulatory process and proposals for reform. This expertise
has contributed to the development of S. 746.
In 1975, the Senate passed a resolution, S. Res. 71,
directing the Governmental Affairs Committee to conduct a
comprehensive study of Federal regulations, to assess the
impact of regulatory programs, and to analyze the need for
change. The Committee spent almost two years carrying out that
mandate and concluded with a six-volume report on various
aspects of the regulatory system, from public participation in
the regulatory process, to the role of congressional
oversight.26 These volumes constitute the most
thorough review of the regulatory process ever conducted by the
Congress. The problems identified and solutions proposed have
substantially informed subsequent debates on regulatory reform,
both within and outside of the Committee, and have influenced
the drafting of this legislation. The Study emphasizes, for
example, that poor, costly, and burdensome agency regulations
often are a product of defective preliminary analyses which
fail adequately to account for costs, the possibility of
alternative regulatory solutions, or no regulation at
all.27
---------------------------------------------------------------------------
\26\ The Governmental Affairs Committee published the following six
volumes of the Study on Federal Regulation between January 1977 and
December 1978:
1. Senate Committee on Government Operations, 95th Cong., 1st
Sess., 1 Study on Federal Regulation, ``The Regulatory Appointments
Process'' (Comm. Print 1977).
2. Senate Committee on Government Operations, 95th Cong., 1st
Sess., 2 Study on Federal Regulation, ``Congressional Oversight of
Regulatory Agencies'' (Comm. Print 1977).
3. Senate Committee on Governmental Affairs, S. Doc. 95-71, 95th
Cong., 1st Sess., 3 Study on Federal Regulation, ``Public Participation
in Regulatory Agency Proceedings'' (Comm. Print 1977).
4. Senate Committee on Governmental Affairs, S. Doc. 95-72, 95th
Cong., 1st Sess., 4 Study on Federal Regulation, ``Delay in the
Regulatory Process'' (Comm. Print 1977).
5. Senate Committee on Governmental Affairs, S. Doc. 95-91, 95th
Cong., 2d Sess., 5 Study on Federal Regulation, ``Regulatory
Organization'' (Comm. Print 1977).
6. Senate Committee on Governmental Affairs, S. Doc. 96-13, 96th
Cong., 1st Sess., 6 Study on Federal Regulation, ``Framework for
Regulation'' (Comm. Print 1978).
\27\ The following conclusion from the 1978 Study rings true today:
---------------------------------------------------------------------------
The report finds that decisions when and how to regulate all
too often are based on insufficient analysis and consideration
of alternatives. Simply because a problem exists and, in theory
is remediable, does not mean that regulation or other
government intervention is desirable. Controls should only be
undertaken where there is a clearly identified problem that
cannot otherwise be solved, and where the anticipated
achievements are significant and vitiated by projected adverse
consequences.
We believe that before Congress or the agency adopts any
proposed regulatory scheme, the possible economic
justifications for regulation should be scrutinized. The
discipline inherent in that procedure is a key element in
helping to insure good regulatory decisions.
6 Study on Federal Regulation, pp. xi-xii.
The Committee's Study provided the foundation for extensive
hearings in the 96th 28 and 97th 29
Congresses. These led to the introduction of S. 1080, the
``Regulatory Reform Act of 1981,'' which was jointly referred
to the Governmental Affairs Committee and the Judiciary
Committee. After receiving unanimous support in Committee, S.
1080 passed the full Senate in 1982 by a vote of 94-0. S. 1080
reflected the increasing concern that the costs of federal
regulation in too many cases do not justify the benefits and
that the scientific and policy assumptions underlying
regulatory decisions often are questionable. Although S. 1080
was overwhelmingly endorsed by the Senate, it was not acted on
in the House of Representatives and died there.
---------------------------------------------------------------------------
\28\ Hearings on Regulatory Legislation, Senate Committee on
Governmental Affairs, 96th Cong., 1st Sess. (1979) (2 parts). These
hearings, encompassing 11 days of testimony from 80 witnesses, are
summarized in S. Rep. No. 96-1018, part 1, 52-55, 96th Cong., 2d Sess.
(1980).
\29\ Hearings on Regulatory Reform Legislation of 1981, Senate
Committee on Governmental Affairs, 97th Cong. 1st Sess. (1981). See
also, S. Rep. No. 97-305, 97th Cong., 1st Sess. 1981. The development
of the reform legislation was in close cooperation with the Senate
Judiciary Committee. See S. Rep. No. 96-1018, Part 2, 96th Cong., 2d
Sess. (1980) (joint report of the Senate Governmental Affairs and
Judiciary Committees).
---------------------------------------------------------------------------
Early in the 104th Congress, Chairman Bill Roth introduced
legislation to improve the regulatory process, S. 291, the
``Regulatory Reform Act of 1995.'' S. 291 contained some of the
basic elements of S. 1080, such as cost-benefit analysis,
centralized regulatory review, and periodic review of existing
rules. S. 291 added other requirements, such as risk assessment
of major environmental, health and safety rules, regulatory
accounting, and comparative risk analysis for setting more
rational regulatory priorities. S. 291 was reported unanimously
by the Committee.30
---------------------------------------------------------------------------
\30\ Many current members of the Governmental Affairs Committee
voted for S. 291, including Senators Roth, Thompson, Stevens, Cochran,
Lieberman, Levin and Akaka.
---------------------------------------------------------------------------
Another regulatory reform bill, S. 343, the ``Comprehensive
Regulatory Reform Act of 1995,'' was introduced early in the
104th Congress. S. 343 covered many of the same issues as S.
1080 and S. 291, but differed in some significant respects. For
example, the cost-benefit requirements were ``decisional
criteria'' that would have amended the substantive standards of
the statutes authorizing the regulations. The decisional
criteria would have required agencies to select, as a matter of
law, the regulatory alternative with the greatest net benefits.
S. 343 also contained a process to allow parties to petition
agencies to review existing rules. S. 343 was jointly referred
to the Governmental Affairs Committee and the Judiciary
Committee.
After the Governmental Affairs Committee unanimously passed
S. 291, the Judiciary Committee reported out S. 343. S. 343
became the subject of extensive negotiations before it was
brought to the floor for consideration during the summer of
1995. The long floor debate ended after three unsuccessful
cloture votes on S. 343 and a close vote defeating the Glenn-
Chafee substitute amendment, which was based on S. 291.
Following the contentious regulatory reform debate of the
104th Congress, Senators Levin and Thompson agreed to work
together to develop bipartisan reform legislation. This led to
the introduction in the 105th Congress of S. 981, the
Regulatory Improvement Act, which is the predecessor to S. 746.
Like S. 746, S. 981 was rooted in past Committee initiatives,
particularly S. 291, but S. 746 was significantly streamlined
and modified. The legislation is grounded in a philosophy of
greater transparency, better informed decision making, and
increased accountability. This philosophy was supported by the
growing Committee record on the shortcomings of the regulatory
process.
In 1996, Senator Thompson, then Chairman of the
Subcommittee on Financial Management and Accountability,
initiated oversight on the implementation of the
Administration's Executive Order 12866 and other initiatives to
reinvent regulation. The Committee heard testimony from many
witnesses and reviewed investigations of the GAO indicating
that E.O. 12866 and the Administration's ``Cutting Red Tape''
initiative were not performing as well as intended.
When Senator Thompson became Chairman of the Committee in
1997, he initiated a series of GAO investigations of the
regulatory process with Ranking Member John Glenn. These
investigations reviewed implementation of Title II of the
Unfunded Mandates Reform Act of 1995; agency efforts to
eliminate and revise existing regulations; agency documentation
of changes made to regulatory proposals during the OMB review
process; and agency use of cost-benefit analysis. All of these
investigations indicated that the current regulatory process is
inadequate.31
---------------------------------------------------------------------------
\31\ See GAO, Unfunded Mandates: Reform Act Has Little Effect on
Agencies' Rulemaking Actions, GAO/GGD-98-30 (Feb. 1998); GAO,
Regulatory Reform: Changes Made to Agencies' Rules Are Not Always
Clearly Documented, GAO/GGD-98-31 (Jan. 1998); GAO, Regulatory Reform:
Agencies' Efforts to Eliminate and Revise Rules Yield Mixed Results,
GAO/GGD-98-3 (Oct. 1997).
---------------------------------------------------------------------------
The Committee held two hearings on S. 981, in September
1997 and February 1998. S. 981 was supported by diverse groups
and individuals representing, among others, State and local
governments, agricultural interests, scientists, policy
institutes, small businesses, and educators. Supporters
testified that S. 981 would foster better federal protection of
public health and safety and the environment; would expedite
the development and issuance of rules; and lead to more
reasoned decisionmaking. Others, including representatives of
environmental, public safety, and labor groups, opposed the
bill as restrictive of agency authority, likely to cause delays
in issuing rules, and leading to agencies placing cost
justification above other factors in determining the regulatory
approach to follow.
The Committee considered at length the concerns raised by
the witnesses opposed to S. 981. Many of the issues raised by
the bill's opponents were addressed in the substitute amendment
offered by Senator Levin and Chairman Thompson on February 4,
1998, and adopted at a markup on March 10, 1998. Others were
addressed by the Committee as amendments during the markup of
S. 981. S. 981 was voted out of the Committee 8-4 on March 10,
1998 32 but did not receive floor consideration.
---------------------------------------------------------------------------
\32\ S. Rep. 105-188, 105th Cong., 2d Sess. (1998).
---------------------------------------------------------------------------
D. CHANGES IN THE REGULATORY IMPROVEMENT ACT FROM THE 105TH CONGRESS
(S. 981) TO THE 106TH CONGRESS (S. 746)
Following the Committee's reporting out of S. 981, Senator
Levin and Chairman Thompson engaged in a series of discussions
with the Administration. The starting point for these
discussions was the letter from then-OMB Director Franklin
Raines prior to the markup of S. 981 expressing the
Administration's concerns with a number of provisions in the
substitute to S. 981. On July 1, 1998, Senator Levin and
Chairman Thompson responded to Director Raines, accepting many
of the changes proposed by the Administration. By letter dated
July 15, 1998, Acting OMB Director Jacob Lew informed Senators
Levin and Thompson that ``if the bill emerges from the House
and Senate as you now propose, with no changes, the President
would find it acceptable and sign it.'' 33
---------------------------------------------------------------------------
\33\ In his written statement submitted into the record of the
Committee's April 21, 1999 hearing on S. 746, OMB Director Lew
reiterated the Administration's commitment that, as proposed, the
President would sign S. 746.
---------------------------------------------------------------------------
As introduced, S. 746 was identical to S. 981 as reported
with the changes accepted by Senator Levin and Chairman
Thompson in their July 1 letter. The key changes from S. 981 as
reported and S. 746 as introduced follow:
1. Judicial review: S. 746 modifies S. 981 by adding
language to clarify that the court shall consider the cost-
benefit analysis, cost-benefit determination, and risk
assessment only in determining under the statute granting the
rule making authority whether the final rule is arbitrary and
capricious. S. 746 also gives a court discretion on whether to
remand or invalidate a rule if the agency fails to perform the
analysis, assessment or determination, or provide for peer
review, and requires a court to order the agency to perform any
or all of these actions if the court allows the rule to take
effect without one or more of them having been performed.
2. Relationship of Regulatory Improvement Act to other
statutes: S. 746 adds two additional provisions to reiterate
that S. 981 did not contain a ``supermandate'' that would
override or alter substantive standards of authorizing
statutes. The revisions confirm that S. 746 does not override
the substantive standards under the statute authorizing the
rule and that the agencies must consider the full range of
regulatory options available under the authorizing statute. S.
746 also confirms that agencies are still entitled to the
deference accorded to them by reviewing courts under the
Chevron doctrine.
3. Review of Rules: S. 746 deletes S. 981's provisions for
the review of existing rules.
4. Risk Assessment: Like S. 981, S. 746 requires a risk
assessment to be conducted for all major rules that have the
primary purpose of addressing risks to health, safety, or the
environment. S. 981 would also have applied to any risk
assessment that is not the basis of a rulemaking if the OMB
Director reasonably determined that the risk assessment might
have a substantial impact on public policy or the economy. S.
746 changes the bill with respect to risk assessments that are
not the basis of a rulemaking by making its risk assessment
requirements applicable if the Director anticipates that the
risk assessment could have an annual effect on the economy of
$100 million or more.
5. Peer review: S. 746 clarifies that members of agency
advisory boards required by statute, and persons who serve as
contractors or grantees to the agency are not precluded from
serving as peer reviewers solely because the bill requires peer
reviewers to be independent of the agency. S. 746 also raises
the threshold for requiring agencies to conduct peer review of
cost-benefit analyses for rules that are anticipated to have a
$500 million annual effect (as opposed to $100 million under S.
981), and clarifies that an agency need conduct only one peer
review of the cost- benefit analysis and the risk assessment.
6. Net benefits: S. 746 clarifies that application of the
net benefits analysis is to include consideration of
nonquantifiable as well as quantifiable benefits.
7. Substitution risk: S. 981 defined ``substitution risk''
as a significant risk to health, safety, or the environment
reasonably likely to result from a regulatory option. S. 746
clarifies that agencies are expected to consider ``reasonably
identifiable'' risks, and that risks attributable to the effect
of a regulatory option on the income of individuals are not to
be considered.
8. Exemptions: S. 981 as reported exempted from coverage
``a rule or agency action that authorizes the introduction into
commerce, or recognizes the marketable status of, a product.''
S. 746 amends this exemption to include removal of a product as
well as introduction into commerce, and it limits the exemption
only to rules promulgated under the Federal Food, Drug and
Cosmetic Act.
E. COMMITTEE CONSIDERATION OF S. 746
As with its predecessor bill, S. 981, many individuals and
organizations representing a wide range of segments of American
society have strongly supported S. 746. These include
representatives of many State and local governments, including
officials responsible for environmental protection and safety;
small business owners; the National Academy of Sciences; many
educational organizations; the GAO; John Graham, Director of
the Harvard Center for Risk Analysis; the United States Chamber
of Commerce; former Federal regulators; and many other
scholars, officials, and experts on the regulatory process.
Testifying in favor of the bill, John Graham, Director of
the Harvard Center for Risk Analysis,cited a study that found
that:
[R]eallocation of lifesaving resources to cost-
effective programs could save 60,000 more lives per
year than we are currently saving, at no increased cost
to taxpayers or the private sector! In short, a smarter
regulatory system can provide the public with more
protection against hazards at less cost than we are
achieving today.34
---------------------------------------------------------------------------
\34\ Testimony of John D. Graham before the Senate Governmental
Affairs Committee, April 21, 1999.
Witnesses testifying in favor of the bill included State
and local government officials whose ability to protect their
constituents' health, safety, and environmental surroundings is
affected on a daily basis by federal regulations. These
witnesses acknowledged the benefits of federal regulation, but
testified that they believe that S. 746 will promote better
prioritization and coordination between the state and local
governments with the federal government, resulting in better
use of limited resources. Robert E. Roberts is the Executive
Director of the Environmental Council of States, which is
comprised of the state and territorial officials who are
responsible for environmental safety in their respective
---------------------------------------------------------------------------
jurisdictions. Mr. Roberts told the Committee that:
We support the consideration of cost benefit
analysis, because to do otherwise is to risk
misapplication of limited resources. We support risk
analysis because to do otherwise may be to attack the
wrong programs. Expanding the participation of state
and local government officials in the development of
national environmental requirements can only strengthen
the final products.35
\35\ Testimony of Robert E. Roberts before the Senate Governmental
Affairs Committee, April 21, 1999.
Similarly, in a letter filed concurrent with the hearing on
S. 746, the leaders of the ``Big 7'' organizations
36 which represent the nation's state, county, and
municipal government organizations, stated that:
---------------------------------------------------------------------------
\36\ The National Governor's Association, the National Conference
of State Legislatures, the Council of State Governments, the National
Association of Counties, the National League of Cities, the U.S.
Conference of Mayors, and the International City/County Management
Association.
The proposed bipartisan legislation would greatly
assist state and local governments in assessing the
costs and benefits of major regulations. This bill
would lead to improved quality of federal regulatory
programs and rules, increase federal government
accountability, and encourage open communication among
federal agencies, state and local governments, the
public and Congress regarding federal regulatory
priorities.37
---------------------------------------------------------------------------
\37\ Letter from Governor Thomas R. Carper of Delaware et al., to
Chairman Thompson and Senator Levin (April 21, 1999).
Scott Holman, testifying on behalf of the U.S. Chamber of
Commerce, is owner and President of a small business, a
manufacturer of custom steel castings for the automotive
tooling, machine tool, steel mill, and construction industries.
---------------------------------------------------------------------------
Mr. Holman testified that:
Information is power. This has never been as true as
it is in today's ``information age.'' S. 746 is about
ensuring a healthy exchange of information on
governmental decisions between the People and their
government. One of the founding principles of our
Nation was the ability of People to question their
government. The Regulatory Improvement Act of 1999
provides power to the American people through greater
information.38
---------------------------------------------------------------------------
\38\ Testimony of Scott Holman before the Senate Governmental
Affairs Committee, April 21, 1999.
Ronald A. Cass, Dean of the Boston University School of
---------------------------------------------------------------------------
Law, told the Committee that:
S. 746 generally should make agency rulemaking
correspond more closely to public interest. The changes
S. 746 would effect primarily ask that agencies attend
to considerations that should be relevant to regulatory
rulemaking, that agencies assess critically information
pertinent to their rulemaking decisions, and that
agencies allow these assessments to be open to the sort
of comparative evaluation common in other venues for
similar analysis.39
\39\ Testimony of Dean Ronald A. Cass before the Senate
Governmental Affairs Committee, April 21, 1999.
Dean Cass further refuted concerns raised by some that S.
746 would drive agencies to always select the least costly
regulatory approach and promote a ``one size fits all''
regulatory approach. He stated that ``S. 746 does not make
formal cost-benefit analysis the sole input to agency decision-
making, and the bill properly cautions attention to
nonquantifiable as well as quantifiable variables''
40 and that ``the risk assessment principles in S.
746 . . . do not handcuff regulatory agencies but merely
promote better informed decision-making. . . . the requirement
of thoughtful risk assessments, including explanations of the
agency's analysis of scientific evidence, is designed to
improve the information relied on by the agency and the
communication of agency decisions to the interested
public.41
---------------------------------------------------------------------------
\40\ Id.
\41\ Id.
---------------------------------------------------------------------------
Some expressed concern that the bill would alter the
current mode of judicial review of regulations by enabling
courts to review the validity of a cost-benefit analysis or
risk assessment without regard to whether the rule itself is
supportable by the facts and law. Dean Cass rebutted such
concerns, noting that:
The judicial review provision in S. 746 seems well-
tailored, neither insulating considerations that make
regulatory analysis sound or unsound from review nor
allowing judicial review to become a strategic tool of
interests opposed to agency action.42
---------------------------------------------------------------------------
\42\ Id. See also Testimony of Warren Belmar, Chair, Section of
Administrative Law and Regulatory Practice, American Bar Association,
before the Senate Governmental Affairs Committee, S. Hrg. 105-468 (Feb.
24, 1998) (expressing ABA support for the judicial review provision of
the Regulatory Improvement Act); Testimony of Ernest Gellhorn,
Professor of Law, George Mason University School of Law, before the
Senate Governmental Affairs Committee, S. Hrg. 105-335 (Sept. 12, 1997)
(supporting the judicial review provision of the Regulatory Improvement
Act).
Dr. Lester M. Crawford, a former federal regulator, also
testified in support of S. 746, including the cost-benefit
analysis, risk assessment, and peer review requirements. In
regard to peer review, Dr. Crawford testified that it has been
used effectively by the FDA, among other agencies, and stated
that ``peer review can and does broaden the expertise available
to the government and it makes the process more open and
democratic.'' 43
---------------------------------------------------------------------------
\43\ Testimony of Dr. Lester M. Crawford, Director, Georgetown
University Center for Food and Nutrition Policy before the Senate
Governmental Affairs Committee, April 21, 1999.
---------------------------------------------------------------------------
Dr. Milton Russell, former Assistant Administrator of the
EPA, told the Committee:
In contrast to previous proposals, which I did not
support, I believe that [the Regulatory Improvement
Act] casts the correct balance in encouraging
appropriate analysis to assure effective and efficient
regulation, in avoiding counterproductive, excessive
review by the courts, and in ensuring that regulation
moves swiftly to implementation to protect the health
and safety of the American people and the
environment.44
---------------------------------------------------------------------------
\44\ Testimony of Dr. Milton Russell, Senior Fellow, Joint
Institute for Energy and Environment, Professor Emeritus of Economics
at the University of Tennessee, before the Senate Governmental Affairs
Committee, S. Hrg. 105-468 (Feb. 24, 1998).
L. Nye Stevens, Director, Federal Management and Workforce
Issues of the General Government Division of GAO, who testified
in the 105th Congress in favor of S. 981, submitted a statement
for the record expressing similar support for S. 746. Mr.
Stevens cited a 1998 GAO report 45 in which GAO
found that cost-benefit analyses prepared under Executive Order
12866 did not incorporate the ``best practices'' elements
recommended by OMB, lacking sufficient, if any, discussion of
alternative regulatory approaches or explanation of
assumptions, limitations, and uncertainties in cost-benefit
analyses. Analyses also lacked executive summaries that could
help Congress, decisionmakers, or the public quickly identify
key information addressed in the agency analyses. Mr. Stevens
told the Committee that:
---------------------------------------------------------------------------
\45\ Regulatory Reform: Agencies Could Improve Development,
Documentation, and Clarity of Regulatory Economic Analyses, GAO/RCED-
98-142 (May 26, 1998).
S. 746 addresses many of these areas of concern . . .
Enactment of the analytical, transparency, and
executive summary requirements in S. 746 would extend
and underscore Congress' previous statutory
requirements that agencies identify how regulatory
decisions are made. We believe that Congress and the
public have a right to know what alternatives the
agencies considered and what assumptions they made in
deciding to regulate. . . . Passage of S. 746 would
provide a statutory foundation for such principles as
openness, accountability, and sound science in
rulemaking.46
---------------------------------------------------------------------------
\46\ Statement for the Record of L. Nye Stevens before the Senate
Committee on Governmental Affairs, April 21, 1999.
In his testimony on S. 981, the predecessor bill to S. 746,
Dr. Bruce Alberts, President of the National Academy of
---------------------------------------------------------------------------
Sciences, told the Committee:
[M]any scientists and engineers who have devoted
their careers to working on environmental problems are
puzzled as to why anyone might oppose [the Regulatory
Improvement Act]. 47
---------------------------------------------------------------------------
\47\ Testimony of Dr. Bruce Alberts, President, National Academy of
Sciences, before the Senate Committee on Governmental Affairs, February
24, 1998.
Others, including representatives of environmental, public
safety, and labor groups, opposed the bill. Despite changes
made to the Regulatory Improvement Act following the reporting
out of S. 981 last year, witnesses expressed concern about the
requirement that agencies state whether the proposed rule the
agency selected is
likely to have benefits that justify the costs or is likely to
be more cost-effective or have greater net benefits than the
other regulatory alternatives considered by the agency. One
witness stated that ``the take home message of S. 746 to
agencies is to optimize the economic benefits of regulation
relative to costs.'' 48 Some witnesses also
testified that S. 746 would require counterproductive analysis
and would ``result in extensive delays in the time it takes for
regulatory decisions to be made'' 49 to the already
lengthy process of proposing and issuing rules.50
They also argued against the peer review requirements of the
bill, claiming that, among other things, they will give an
undue advantage to industry representatives in shaping health
and safety rules.51
---------------------------------------------------------------------------
\48\ Testimony of David C. Vladeck, Director, Public Citizen
Litigation Group before the Senate Governmental Affairs Committee,
April 21, 1999.
\49\ Testimony of Patricia G. Kenworthy for the National
Environmental Trust before the Senate Governmental Affairs Committee,
April 21, 1999.
\50\ Id.
\51\ See Testimony of Franklin E. Mirer, Director of Health and
Safety, United Auto Workers, before the Senate Governmental Affairs
Committee, April 21, 1999.
---------------------------------------------------------------------------
After careful consideration of all views, the Committee
disagrees with the analysis of the organizations opposing the
bill for the reasons identified and explained throughout this
report.
III. Legislative History and Committee Consideration
A. COMMITTEE HEARINGS
On April 21, 1999, the Governmental Affairs Committee held
a hearing on S. 746. This hearing built on the Committee's
extensive hearing record and legislative history on regulatory
reform from the 104th and 105th Congresses. Testifying at this
hearing were: Gregory S. Lashutka, Mayor of Columbus, Ohio, for
the National League of Cities; Robert E. Roberts, Executive
Director, Environmental Council of States; Scott Holman,
Chairman, Regulatory Affairs Committee of the U.S. Chamber of
Commerce; Ronald A. Cass, Dean of Boston University School of
Law; Dr. Lester Crawford, Director, Georgetown University
Center for Food and Nutrition Policy; Patricia G. Kenworthy,
Vice President for Government Affairs, National Environmental
Trust; John D. Graham, Director, Harvard Center for Risk
Analysis; David C. Vladeck, Director, Public Citizen Litigation
Group; and Dr. Franklin E. Mirer, Director of Health and
Safety, International Union, United Automobile, Aerospace, and
Agricultural Workers of America. Jacob J. Lew, the Director of
OMB, and L. Nye Stevens, Director of Federal Management and
Workforce Issues, GAO, submitted statements for the record.
B. AMENDMENTS AND COMMITTEE ACTION
On May 20, 1999, the Committee on Governmental Affairs
marked up and favorably reported S. 746 by a vote of 11 to 5.
Voting in the affirmative were Senators Roth, Stevens, Collins,
Voinovich, Domenici, Cochran, Specter, Gregg, Levin, Cleland,
and Thompson. Voting in the negative were Senators Lieberman,
Akaka, Durbin, Torricelli, and Edwards.
A number of amendments were offered, debated and voted
upon. The following amendment was adopted: Senator Durbin
offered an amendment, which was amended by Senator Levin's
second degree amendment, to require OMB to submit to Congress
in 2002 an accounting statement and report containing an
estimate of the total annual incremental benefits and costs of
complying with the provisions of subchapter II added by this
Act for each agency.
The following amendments were defeated:
(1) Senator Lieberman offered an amendment to require that
the cost-benefit determination required by the bill not be
judicially reviewable and to require that an agency's failure
to conduct a particular requirement for cost-benefit analysis
or risk assessment will not authorize a court to remand the
rule unless the agency ``entirely'' fails to perform the
analysis. The amendment was defeated 6-10. Voting in the
affirmative were Senators Lieberman, Akaka (by proxy), Durbin,
Torricelli (by proxy), Cleland (by proxy), and Edwards. Voting
in the negative were Senators Roth (by proxy), Stevens (by
proxy), Collins, Voinovich, Domenici (by proxy), Cochran (by
proxy), Specter (by proxy), Gregg (by proxy), Levin, and
Thompson.
(2) Senator Lieberman offered an amendment to require that
public hearings in which scientific experts may be cross-
examined would satisfy the bill's requirements for independent
peer review of cost-benefit analyses and risk assessments. The
amendment was defeated 6-10. Voting in the affirmative were
Senators Lieberman, Akaka (by proxy), Durbin, Torricelli (by
proxy), Cleland (by proxy), and Edwards. Voting in the negative
were Senators Roth (by proxy), Stevens (by proxy), Collins,
Voinovich, Domenici (by proxy), Cochran (by proxy), Specter (by
proxy), Gregg (by proxy), Levin, and Thompson.
(3) Senator Durbin offered an amendment to exempt from the
bill's cost-benefit and risk assessment requirements any rule
or agency action to reduce the use of tobacco products by
minors or protect the public from the health risks associated
with tobacco products. The amendment was defeated 6-10. Voting
in the affirmative were Senators Specter (by proxy), Lieberman,
Akaka (by proxy), Durbin, Torricelli (by proxy), and Cleland
(by proxy). Voting in the negative were Senators Roth (by
proxy), Stevens (by proxy), Collins (by proxy), Voinovich,
Domenici (by proxy), Cochran (by proxy), Gregg (by proxy),
Levin, Edwards, and Thompson.
(4) Senator Lieberman, with Senator Akaka, offered an
amendment to exempt any rules from the cost-benefit analysis
and risk assessment requirements if the agency is not required
to base the rule on the outcome of a risk assessment. The
amendment was defeated 7-9. Voting in the affirmative were
Senators Specter (by proxy), Lieberman, Akaka (by proxy),
Durbin, Torricelli (by proxy), Cleland, and Edwards. Voting in
the negative were Senators Roth (by proxy), Stevens (by proxy),
Collins, Voinovich, Domenici (by proxy), Cochran (by proxy),
Gregg, Levin, and Thompson.
(5) Senator Lieberman offered an amendment to require that
the bill's risk assessment requirements would not apply to any
programs for collecting and disseminating information. The
amendment was defeated 7-9. Voting in the affirmative were
Senators Specter (by proxy) Lieberman, Akaka (by proxy),
Durbin, Torricelli, Cleland, and Edwards. Voting in the
negative were Senators Roth (by proxy), Stevens (by proxy),
Collins, Voinovich, Domenici (by proxy), Cochran (by proxy),
Gregg, Levin, and Thompson.
(6) Senator Cleland offered an amendment to exclude from a
peer review panel any person who has, or is employed by an
entity which has, a significant direct financial interest in
the outcome of a rulemaking. Chairman Thompson noted that S.
746 specifically requires peer reviewers to be subject to
current conflict of interest standards, and the amendment was
defeated 6-9. Voting in the affirmative were Senators
Lieberman, Akaka (by proxy), Durbin, Torricelli, Cleland, and
Edwards. Voting in the negative were Senators Roth (by proxy),
Stevens (by proxy), Collins, Voinovich, Domenici (by proxy),
Cochran (by proxy), Gregg, Levin, and Thompson.
(7) Senator Torricelli offered an amendment to exempt from
the bill's cost-benefit analysis and risk assessment
requirements any rules relating to the protection of children's
health, food safety, worker and workplace safety, environmental
protection, firefighter safety, or civil rights. The amendment
was defeated 6-10. Voting in the affirmative were Senators
Specter (by proxy), Lieberman, Akaka (by proxy), Durbin (by
proxy), Torricelli, and Cleland (by proxy). Voting inthe
negative were Senators Roth (by proxy), Stevens (by proxy), Collins,
Voinovich, Domenici (by proxy), Cochran (by proxy), Gregg (by proxy),
Levin, Edwards, and Thompson.
IV. Administration Views
OMB Director Jacob Lew submitted a statement to the
Committee dated April 21, 1999, expressing the Administration's
views on S. 746. Director Lew noted that the Administration had
offered suggestions concerning S. 981, that those concerns had
been taken seriously by the sponsors, and that S. 746 includes
changes suggested by the Administration. Director Lew stated
that the Administration's view remains the same as in the July
15, 1998 letter to Senator Levin and Chairman Thompson and
reiterated that ``if S. 746 emerges from the Senate and House
as you now propose, the President would sign it.''
V. Section-By-Section Analysis
SECTION 1. SHORT TITLE
The name of this legislation is the ``Regulatory
Improvement Act of 1999''.
SECTION 2. FINDINGS
Section 2 lays out eight basic findings by the Committee.
These findings underscore both the strengths and limitations of
regulatory analysis and review. The findings also reflect the
experience and expertise of the Committee, as informed by
scores of experts, government officials, and stakeholders in
the regulatory process.52 These findings are as
follows:
---------------------------------------------------------------------------
\52\ See, e.g., Letter of Baruch Fishoff, Professor of Social and
Decision Sciences and Professor of Engineering and Public Policy,
Carnegie Mellon University, to Chairman Fred Thompson, July 15, 1997,
S. Hearing 105-535, at 294 (``The Findings are a remarkably succinct
summary of what we have learned over the past 20 years regarding the
role of analysis in regulation. We would be much better off as a
society were the wisdom in them more widely understood and accepted.'')
---------------------------------------------------------------------------
First, effective regulatory programs provide important
benefits to the public, including improving the environment,
worker safety, and public health. Regulatory programs also
impose significant costs on the public, including individuals,
businesses, and State, local, and tribal governments.
Second, improving the ability of Federal agencies to use
scientific and economic analysis in developing regulations
should yield increased benefits and more effective protections
while minimizing costs.53
---------------------------------------------------------------------------
\53\ See, e.g., Testimony of Paul R. Portney, President, Resources
for the Future, before the Senate Committee on Governmental Affairs,
September 12, 1997 (Under this legislation, ``we might be able to shave
off a chunk of the nearly $300 billion OIRA estimates we spend each
year on environmental, health and safety regulation . . . without
compromising the benefits we get from regulations. . . . Even a cynical
public ought to warm to a $30 billion ``free lunch'' each year that
does not compromise the quality of their environment or safety of their
food and other products they consume each year.''); Testimony of John
D. Graham, Director, Harvard Center for Risk Analysis, before the
Senate Committee on Governmental Affairs, September 12, 1997.
---------------------------------------------------------------------------
Third, cost-benefit analysis and risk assessment are useful
tools to better inform agencies in developing regulations,
although they do not replace the need for good judgment and
consideration of values.
Fourth, the evaluation of costs and benefits must involve
the consideration of the relevant information, whether
expressed in quantitative or qualitative terms, including
factors such as social values, distributional effects, and
equity.
Fifth, cost-benefit analysis and risk assessment should be
presented with a clear statement of the analytical assumptions
and uncertainties, including an explanation of what is known
and not known and what the implications of alternative
assumptions might be.
Sixth, the public has a right to know about the costs and
benefits of regulations, the risks addressed, the risks
reduced, and the quality of scientific and economic analysis
used to support decisions. Such knowledge will promote the
quality, integrity and responsiveness of agency actions.
Seventh, the Administrator of the Office of Information and
Regulatory Affairs should oversee regulatory activities to
raise the quality and consistency of cost-benefit analysis and
risk assessment among all agencies.
Eighth, the Federal Government should develop a better
understanding of the strengths, weaknesses, and uncertainties
of cost-benefit analysis and risk assessment and conduct the
research needed to improve these analytical tools.
This legislation is designed to elevate the use of modern
decisionmaking tools, such as risk assessment and cost-benefit
analysis, to make the regulatory process more transparent, more
efficient and effective, and more accountable to the public.
SECTION 3. REGULATORY ANALYSIS
Section 3(a) substantially amends chapter 6 of title 5,
United States Code. Section 3(a) creates two new subchapters.
Subchapter II requires analysis of agency rules, including
cost-benefit analysis, risk assessment, peer review, and
guidelines, as well as a comparative risk analysis study.
Subchapter III requires executive oversight of the rule making
process.
Section 3(b) is a savings clause, stating that the current
legislation does not limit any of the President's
constitutional duties and authorities, including the authority
to review regulatory actions not covered by this legislation.
Finally, section 3(c) provides the technical and conforming
amendments necessary to reorganize chapter 6 into subchapters,
including, for example, moving the Regulatory Flexibility Act
to subchapter I of chapter 6.
In amending title 5, United States Code, the Committee-
passed bill applies the definition of ``agency'' under section
551 to subchapters II and III of the bill--the regulatory
analysis and executive oversight requirements. This definition
includes the independent regulatory agencies within the scope
of this legislation. Thus, the requirements to identify major
rules and to performcost-benefit analyses and risk assessments
would apply not only to departments and other executive agencies, but
also to the independent regulatory agencies, such as the Federal Energy
Regulatory Commission, the Nuclear Regulatory Commission, and the
Consumer Product Safety Commission.
This legislation also would require independent regulatory
agencies, like all other Executive Branch agencies, to be
subject to Presidential oversight for compliance with the
requirements of this legislation. Such Presidential oversight
includes the review of proposed and final major rules by OMB's
Office of Information and Regulatory Affairs. Since 1981,
OIRA's regulatory review authority under Presidential executive
order (E.O. 12291, 12498, and 12866) has explicitly exempted
independent regulatory agencies and made their participation in
the regulatory review process voluntary. The Committee believes
that the provisions of this legislation should apply to all
Executive Branch agencies, including the independent regulatory
agencies. The growing need for more efficient and effective
government regulation, as well as for more coherent management
of the Executive Branch, supports lowering some walls that have
separated the independent agencies from other Executive Branch
agencies.\54\ Specific exemptions are provided within the
definition of ``rule'' where this Committee or other
authorizing Committees determined that there would not be
significant benefits from regulatory analysis or OIRA review.
---------------------------------------------------------------------------
\54\ See, e.g., ACUS Recommendation 95-3: ``Review of Existing
Agency Regulations'' (1995); Administrative Law Conference of the
United States: Recommendation 88-9: ``Presidential Review of Agency
Rulemaking'' (1988); American Bar Association, Commission on Law and
the Economy, Federal Regulation: Roads to Reform (1979), at 108;
American Bar Association, Administrative Law Report No. 110 (1986);
Peter L. Strauss & Cass R. Sunstein, ``The Role of the President and
OMB in Informal Rulemaking,'' 38 Admin. L. Rev. 181, 205 (1986).
---------------------------------------------------------------------------
Section 3(a). In General
Section 3(a) creates new subchapters II and III in chapter
6, title 5, United States Code.
Subchapter II. Regulatory Analysis
Subchapter II establishes provisions for new definitions
(sec. 621); applicability and effect (sec. 622); regulatory
analysis (sec. 623); principles for risk assessments (sec.
624); peer review (sec. 625); deadlines for rule making (sec.
626); judicial review (sec. 627); guidelines, interagency
coordination, and research (sec. 628); and risk-based
priorities study (sec. 629).
Sec. 621. Definitions
This section defines certain terms used in regulatory
analysis. Many of these definitions are used not only in the
new subchapter II, but also are referred to and incorporated
into subchapter III.
(1) The term ``Administrator'' means the Administrator of
the Office of Information and Regulatory Affairs of the Office
of Management and Budget.
(2) The term ``benefit'' means the reasonably identifiable
significant favorable effects, quantifiable and
nonquantifiable, including social, health, safety,
environmental, economic, and distributional effects, that are
expected to result from implementation of, or compliance with,
a rule.
The term ``benefit'' has broad meaning. Benefits are the
favorable effects that are causally related to the rule. In
other words, benefits are the improvements upon the status quo
as a result of the rule. Federal agencies issue regulations to
implement laws passed by Congress. As such, the value of a
regulation is the extent to which it provides the public
benefits envisioned by the underlying law. Regulations
addressing health, safety, or environmental risks, for example,
provide benefits from reducing risk, and the evaluation of
those risk-reduction benefits would be based on the risk
assessment performed under section 624 of this Act.
Benefits can be readily apparent, as in economic benefits
obtained from hazardous material transportation rules or in the
regained safety of a locality's drinking water supply. Benefits
also can be very broad, as in the growth of an economic sector
or improved nation-wide employment rates. Finally, regulatory
benefits can be significant but difficult to quantify, such as
the value of increased visibility over the Grand Canyon.
This wide variety of possible benefits must be recognized
in the rulemaking process. However, merely because benefits may
be varied or difficult to quantify should not relieve agencies
from identifying the specific benefits of a rule. The
identification and evaluation of regulatory benefits should
enable agencies to improve the effectiveness and efficiency of
the regulatory process and to best serve the goals of the
enabling statute.
As a part of this broad meaning of ``benefit,'' the
Committee intends for agencies to consider direct as well as
the indirect benefits. Many benefits can be clearly attributed
to a regulatory action. Many, however, flow in more tangential
ways. The Committee expects agencies to make a reasonably
thorough effort at identifying and analyzing all significant
benefits that flow from a regulatory action. At the same time,
the Committee cautions agencies against speculative attribution
of distant outcomes to a regulatory action.
The definition of benefits is not limited to favorable
effects that can be quantified. They may include, for example,
identifiable and significant but potentially nonquantifiable
benefits, such as increased freedom of choice for consumers or
enhanced opportunities for public enjoyment of the environment.
In other words, benefits that cannot be monetarily quantified,
or even numerically measured, also should be considered and
explained by the agency.
At the same time, the definition of benefits is limited to
those that are ``significant.'' Benefits should be more than
trivial or de minimis. The Committee does not anticipate that
agencies will spend valuable resources trying to assess every
small, remote benefit of a rule; during the cost-benefit
analysis, only significant benefits need be addressed.
(3) The term ``cost'' means the reasonably identifiable
significant adverse effects, quantifiable or nonquantifiable,
including social, health, safety, environmental, economic, and
distributional effects, that are expected to result from
implementation of, or compliance with, a rule. The definition
of ``cost'' parallels that of ``benefit,'' and the concerns
expressed above regarding ``benefit'' apply equally here.
As in the case of ``benefits,'' the Committee intends to
give broad meaning to the term ``cost.'' Agencies must be
sensitive to all of the significant costs regulation can
impose. While compliance costs often comprise a substantial
portion of total costs, there are other costs of regulation. To
name a few, these costs include adverse effects on health,
safety or the environment; such adverse effects increase the
net cost of a regulatory alternative. Costs also include
adverse impacts on consumer choice, technological innovation,
wages, productivity, economic growth, and lower employment.
Again, agencies should eschew unreliable speculation about
costs, as with benefits, but they should try to responsibly
identify all ``significant'' costs imposed by a regulatory
action. The concept of ``cost'' for cost-benefit analysis
includes opportunity costs. \55\ Accordingly, agencies should
be more sophisticated in cost estimation than only summing up
compliance costs.
---------------------------------------------------------------------------
\55\ As Paul Portney told the Committee:
[T]he sum total of out-of-pocket expenditures is not
identical to ``costs'' as economists think of them for a
benefit-cost analysis. [Costs] include the value of time that
people must spend waiting in line for permits, car inspections,
etc. It includes the adverse health effects they incur because
of the time involved to bring a potentially effective new
therapeutic drug to market. It includes the inconvenience they
suffer when a product becomes less effective on account of a
regulation, or disappears from the market altogether. None of
these ``costs'' involves any out-of-pocket expenditure, but
---------------------------------------------------------------------------
they must all be counted in any serious benefit-cost analysis.
Testimony of Paul R. Portney, Resources for the Future, before the
Senate Committee on Governmental Affairs, February 8, 1995.
Finally, agencies must identify and evaluate direct and
indirect costs, as well as quantitative and nonquantitative
costs. If a rule sets in motion a series of legally required
actions that result in costs, even if those actions will be
taken by entities other than the regulatory agency, the agency
should consider such adverse effects as ``costs'' under this
Act. \56\
---------------------------------------------------------------------------
\56\ For example, EPA recently issued a major rule under the Clean
Air Act to reduce particulate matter and ozone and performed a cost-
benefit analysis for the rule under Executive Order 12866. This rule
would have required states to change their State Implementation Plans
(``SIPs'') to satisfy the tighter standards. These SIP revisions would
impose costs that are attributable to the EPA rule and such costs would
be included in a cost-benefit analysis under this legislation.
---------------------------------------------------------------------------
(4) The term ``cost-benefit analysis'' means an evaluation
of the costs and benefits of a rule, quantified to the extent
feasible and appropriate and otherwise qualitatively described,
that is prepared in accordance with the requirements of this
subchapter at the level of detail practical for reasoned
decisionmaking on the matter involved, taking into
consideration uncertainties, the significance and complexity of
the decision, and the need to adequately inform the public.
This definition includes the minimum essential features of
cost-benefit analysis.
The Committee intends that the agencies use the best
available techniques for these analyses and tailor the
specificity and rigor of the analysis to the consequences of
the decision to be made and the need to inform stakeholders and
the public. This provides the agency with reasonable
flexibility in the level of detail and rigor that should be
employed. However, the Committee expects that the analysis will
follow the essential requirements of this legislation.
(5) The term ``Director'' means the Director of the Office
of Management and Budget, acting through the Administrator of
the Office of Information and Regulatory Affairs. The reason
for this definition is two-fold. First, the Committee expects
the Director of OMB, not just the Administrator of OIRA, to be
directly accountable for the prompt and effective
implementation of this legislation. Second, the Committee at
the same time intends to recognize the important role and
responsibility of OIRA in the regulatory process. Since 1980,
when the Committee passed the Paperwork Reduction Act, the
Committee has viewed the Administrator of OIRA as an important
partner in ensuring that the regulatory process is efficient,
effective, and accountable. This legislation will further this
critical role of OIRA.
(6) The term ``flexible regulatory options'' means
regulatory options that permit flexibility to regulated persons
in achieving the objective of the statute as addressed by the
rule making, including market-based mechanisms, outcome-
oriented performance-based standards, or other options that
promote flexibility. The Committee believes that flexible
regulatory options have the potential to be more efficient and
effective than command-and-control regulation.
``Market-based mechanisms'' include regulatory programs or
requirements that impose legal accountability for achieving the
regulatory objective on each regulated entity, afford maximum
flexibility to each regulated entity in meeting mandatory
regulatory objectives, and allow those regulated entities to
respond freely to changes in pertinent economic conditions and
circumstances without undermining the achievement of the
program's regulatory mandate or requiring a new rulemaking.
The Committee has heard testimony that some of our greatest
regulatory successes have been achieved through market-based
mechanisms. 57 One such example is the program for
reducing nationwide sulfur dioxide emissions, established under
Title IV of the Clean Air Act. There, Congress imposed directly
on sources of emissions explicit pollution reduction
requirements. The sources were allowed to meet those
requirements through any means they chose, including purchasing
credits representing the performance of needed reductions by
other sources. This program is achieving greater emissions
reductions at a small fraction of the anticipated costs of
command-and-control regulation and is far ahead of the
statutory schedule. 58
---------------------------------------------------------------------------
\57\ See, e.g., Statement of Paul R. Portney, President, Resources
for the Future, before the Senate Committee on Governmental Affairs,
September 12, 1997; Testimony of C. Boyden Gray, Partner, Wilmer,
Cutler & Pickering and Chairman, Citizens for a Sound Economy, before
the Senate Committee on Governmental Affairs, September 12, 1997;
Testimony of Thomas F. Walton, Director of Economic Policy, General
Motors Corporation, before the Senate Committee on Governmental
Affairs, September 12, 1997; Testimony of Joseph Goffman, Senior
Attorney, Environmental Defense Fund, before the Senate Committee on
Governmental Affairs, March 8, 1995; Testimony of Alan J. Krupnik,
Senior Fellow, Resources for the Future, before the Senate Committee on
Governmental Affairs, March 8, 1995; Testimony of Jonathan B. Wiener,
Associate Professor, Duke University School of Law and Duke University
School of Environment, before the Senate Committee on Governmental
Affairs, March 8, 1995; Testimony of C. Boyden Gray, Partner, Wilmer,
Cutler & Pickering and Chairman, Citizens for a Sound Economy, before
the Senate Committee on Governmental Affairs, March 8, 1995; Testimony
of Carol M. Browner, Administrator, U.S. Environmental Protection
Agency, before the Senate Committee on Governmental Affairs, March 8,
1995.
\58\ See Testimony of Joseph Goffman, Senior Attorney,
Environmental Defense Fund, before the Senate Committee on Governmental
Affairs, March 8, 1995; Testimony of Alan J. Krupnik, Senior Fellow,
Resources for the Future, before the Senate Committee on Governmental
Affairs, March 8, 1995; Testimony of Jonathan B. Weiner, Associate
Professor, School of Law, Duke University, before the Senate Committee
on Governmental Affairs, March 8, 1995; Testimony of C. Boyden Gray,
Partner, Wilmer, Cutler & Pickering and Chairman, Citizens for a Sound
Economy, before the Senate Committee on Governmental Affairs, September
12, 1997.
---------------------------------------------------------------------------
``Performance-based standards'' include requirements,
expressed in terms of outcomes or goals instead of prescriptive
command-and-control measures, that permit discretion and the
use of market-based mechanisms in determining how best to meet
specific requirements in particular circumstances. In contrast
to command-and-control regulation, performance-based standards
simply establish the ultimate regulatory goal and free
regulated parties to meet or exceed that goal as they choose.
Like market-based mechanisms, the Committee requires agencies
to seriously consider performance-based standards because they
have similar elements of flexibility, cost-effectiveness, and
accountability.
(7) The term ``major rule'' is defined to include two
categories of significant rules: economically significant and
other significant rules designated by the Director of OMB.
The first category of ``major rule'' is defined in
subsection 621(7)(A) as a rule that the relevant agency or the
Director of OMB reasonably determines is likely to have an
annual effect on the economy of $100,000,000 or more in
reasonably quantifiable costs.'' To be classified as ``major,''
such a rule should be reasonably likely to have such an effect
in any one year following its adoption.
The Committee's decision to set the threshold for major
rules at $100 million follows the longstanding tradition under
centralized executive review of rules. Since President Ford,
every President has required by executive order the review of
regulations which impose annual costs on the economy of $100
million or more. The bill maintains the traditional $100
million threshold because the Committee believes that it will
not unduly increase the analytical burden of the agencies and
that rules of such significance can benefit greatly from
thorough analysis. All costs of a rule should be considered in
determining whether a rule is ``major'' under subsection
621(7)(A).
Subsection 621(7)(B) provides a second prong to the major
rule definition. This allows the President, through the OMB
Director, to subject to cost-benefit analysis those rules
which, while not imposing costs of $100 million on the economy,
still have a substantial impact. This category includes rules
likely to adversely affect, in a material way, the economy, a
sector of the economy (including small business), productivity,
competition, jobs, the environment, public health or safety, or
State, local or tribal governments, or communities.
Regulatory agencies and the OMB Director should be mindful
of the disproportionate impact their actions can have on
certain groups or sectors of the economy, even if the impact on
the country as a whole is not substantial. This is particularly
true of small business. 59 The Committee encourages
the Director and the agencies to be sensitive to these
concerns.
---------------------------------------------------------------------------
\59\ See Statement of Karen Kerrigan, President, Small Business
Survival Committee, before the Senate Committee on Governmental
Affairs, September 12, 1997 (citing Small Business Administration study
showing that the annual regulatory cost per worker for companies with
less than 20 employees is $5,532).
---------------------------------------------------------------------------
(8) The term ``reasonable alternative'' means a reasonable
regulatory option that would achieve the objective of the
statute as addressed by the rule making and that the agency has
authority to adopt under the statute granting rule making
authority, including flexible regulatory options.
Reasonable alternatives embrace the range of options that
the agency has discretion to consider under the statute
authorizing the rulemaking. The Committee included flexible
regulatory options in the definition of ``reasonable
alternative'' to encourage agencies to seek out such
alternatives. The agency should consider the range of options
authorized by the underlying statute to best achieve the
objective being addressed by the rulemaking. ``Reasonable
alternatives'' do not include alternatives prohibited by the
statute authorizing the rule.
(9) The term ``risk assessment'' means the systematic,
objective process of organizing hazard and exposure information
and, based on a careful analysis of the weight of the
scientific evidence, estimating the potential for specific harm
to an exposed population, subpopulation, or natural resource
including, to the extent feasible, a characterization of the
distribution of risk as well as an analysis of uncertainties,
variabilities, conflicting information, and inferences and
assumptions.
Like the definition of ``cost-benefit analysis,'' the
definition of ``risk assessment'' includes specific qualitative
factors which the Committee views as minimum essential features
of a risk assessment. Specifically, the risk assessment should
be scientifically ``objective'' 60 and ``based on a
careful analysis of the weight of the scientific evidence.''
61 Full consideration of the weight of the evidence
often involves balancing positive and negative findings. The
definition further requires that the risk estimate, to the
extent feasible, must contain a characterization of the
distribution of risk as well as an analysis of uncertainties,
variabilities, conflicting information, and inferences and
assumptions. The Committee believes that this type of
information is necessary to get a complete and meaningful
estimate of the risk.
---------------------------------------------------------------------------
\60\ Agency risk assessments should be scientifically objective to
the extent possible, neither minimizing nor exaggerating the nature and
magnitude of the risks. Such risk assessments should be more
transparent and credible, leading to less contentious risk management
decisions. Such assessments should lead to a more risk-based regulatory
system, offering the opportunity for greater overall protection with
the available resources. See Testimony of John D. Graham before the
Senate Committee on Governmental Affairs, September 12, 1997; Safe
Drinking Water Act of 1996, Section 103, 42 U.S.C. Sec. 300g-1(b)(3).
\61\ See Presidential/Congressional Commission on Risk Assessment
and Risk Management, Risk Assessment and Risk Management in Regulatory
Decision-Making (``Risk Commission Report''), Vol. 1, at 4, 23, 38.
---------------------------------------------------------------------------
The Committee recognizes that risk assessment is a flexible
process by which complextechnical data are combined and
analyzed to provide decision makers with useful information to make
policy decisions. In some decision contexts, such as evaluating food
additives, it is useful to distinguish four steps in the risk
assessment: hazard identification, dose-response analysis (which
together comprise ``hazard assessment''), exposure assessment, and risk
characterization. In other contexts, such as transportation safety, one
or another of the first three steps may not be relevant. The Committee
believes that the definition adopted by this legislation is
sufficiently generic to apply to the wide variety of risks covered by
this legislation. The Committee encourages advances in state-of-the-art
risk assessment practices.
(10) The term ``rule'' has the same meaning as such term is
defined in section 551(4) of title 5, United States Code, with
a number of exclusions.
First, subparagraph (A) exempts from the definition of
``rule'' any rule that is exempt from notice and public comment
procedures under section 553 of title 5 of the United States
Code. These include: rules relating to a military or foreign
affairs functions; interpretative rules; rules relating to
grants, benefits, or loans; rules relating to agency management
or personnel; and rules relating to the acquisition, management
or disposal of federal property. In some cases, these rules
could have a significant impact on the economy. Yet, the
Committee decided to minimize the burdens on the agencies;
where notice and comment pursuant to section 553 is not
required, a cost-benefit analysis will not be required either.
However, the Committee cautions the agencies that any
statement of general applicability that actually alters or
creates rights or obligations of persons outside the agency is
included in this definition. While informal agency guidance is
encouraged, agencies should not attempt to evade the
requirements of this legislation through mischaracterizations
of such materials.
Subparagraph (B) excludes rules involving the internal
revenue laws. The Committee was concerned that the enormous
economic impact of such rules might make an overwhelming number
of tax regulations major rules. While many IRS rules have a
major economic impact or are otherwise significant, they have
this impact because their goal is to raise revenue, pursuant to
the explicit mandates of the underlying statute with little or
no agency discretion.
Subparagraph (C) excludes rules of particular applicability
that approve or prescribe for the future rates, wages, prices,
services, corporate or financial structures, reorganizations,
mergers, acquisitions, accounting practices, or disclosures
bearing on any of the foregoing.
This exemption applies to rules ``of particular
applicability'' as that phrase is understood in section 551(4)
of title 5. These are rules which, while technically within the
definition of ``rule,'' are more properly considered as
licenses or orders because they apply only to a small group or
a single individual. The Committee believes that such rules
would not greatly benefit from the analytical requirements of
this legislation because they are generally developed through
complex and lengthy proceedings, which often involve
sophisticated economic analysis.
Subparagraphs (D) and (E) exclude from the legislation's
scope certain rules relating to monetary policy or to the
safety or soundness of federally insured depository
institutions.
Subparagraph (F) excludes certain rules relating to the
integrity of the securities or commodities futures markets or
to the protection of investors in those markets.
Subparagraph (G) excludes certain rules issued by the
Federal Election Commission and the Federal Communications
Commission.
Subparagraph (H) excludes certain rules required to be
promulgated at least annually pursuant to statute. This
exemption would include certain rules that establish, modify,
open, close, or conduct a regulatory program for a commercial,
recreational, or subsistence activity related tohunting,
fishing, or camping.
Subparagraph (I) excludes certain rules or agency actions
relating to the public debt or fiscal policy of the United
States.
In all of these instances, the Committee believes that the
analytic requirements of the legislation would not enhance the
efficiency or effectiveness of these rules.
Subparagraph (J) exempts from ``rule'' a rule that
authorizes or bars the introduction into or removal from
commerce, or recognizes or cancels recognition of the
marketable status, of a product under the Federal Food, Drug
and Cosmetic Act.
(11) The term ``substitution risk'' means a reasonably
identifiable significant risk to health, safety, or the
environment expected to result from a regulatory option and
does not include risks attributable to the effect of an option
on the income of individuals. A regulatory option designed to
decrease certain risks may sometimes actually increase other
risks.62 A substitution risk is an unintended
adverse consequence. The provisions of S. 746 are intended to
focus greater attention on the possibility of such adverse
consequences, including addressing the likelihood of their
occurrence, estimating the nature and magnitude of their
impacts, and systematically considering substitution risks as a
part of sound regulatory policy-making. The agency should
identify, describe, and evaluate any substitution risks in the
regulatory analysis. The agency should integrate such risks in
its analyses and in making the determinations required under
Section 623(d).
---------------------------------------------------------------------------
\62\ See John D. Graham & Jonathan Baert Wiener, Risk vs. Risk:
Tradeoffs in Protecting Health and the Environment, Harv. Univ. Press
(1995).
---------------------------------------------------------------------------
By ``significant,'' the Committee means that the effect of
the substitution risk should be important. ``Significant'' does
not refer to the magnitude of the increase in risk as the term
``significant risk'' is used or interpreted under various
environmental, health, and safety statutes; 63 it
refers to the relative relationship a risk may have to the
effect of a rule. A risk need not have a likelihood of a
particular level, such as one in ten thousand, to be
significant. For a ``significant increased risk'' to qualify as
a substitution risk, it need not be greater than the original
risk reduction otherwise being achieved by the rule. By
``expected to result,'' the Committee means that the
substitution risk should not be implausible. The Committee does
not intend that attenuated arguments that changes in lifestyle
that could result from changes in income of individuals
potentially attributable to a regulatory option should be
considered a substitution risk under this legislation.
---------------------------------------------------------------------------
\63\ See Occupational Safety and Health Act 29, U.S.C. Sec. 651.
---------------------------------------------------------------------------
Sec. 622. Applicability and effect
Section 622 clarifies the scope and effect of this
legislation. Subsection 622(a) provides that this legislation
applies to all ``major rules'' through the proposed and final
rulemaking stages, except as provided in Subsection 623(f).
Subsection 622(b) clarifies that nothing in Subchapter II
shall be construed to alter or modify: (1) the substantive
standards otherwise applicable to a rulemaking under other
statutes; (2) the range of regulatory options that an agency
has the authority to adopt under the statute authorizing the
agency to promulgate the rule, or deference otherwise accorded
to the agency in construing such statute; or (3) any
opportunity for judicial review made applicable under other
statutes.
This so-called ``savings clause'' clarifies a few important
points: First, this legislation is not intended to override
existing statutory standards. The cost-benefit analysis, risk
assessment, and cost-benefit determination required by this
legislation do not supersede or override the substantive
standards in the statute under which a rule is being issued. In
other words, S. 746 does not contain a so-called
``supermandate.'' S. 746 also does not alter the range of
regulatory options that an agency can consider or the deference
otherwise accorded to the agency in construing the statute
authorizing the rule.
Finally, subsection 622(b) clarifies that S. 746 does not
alter or diminish any opportunities for judicial review
available under other statutes. To the extent that another
Federal statute provides an opportunity for judicial review of
agency action, that opportunity for judicial review continues
to apply. Section 622(b) preserves existing opportunities to
secure judicial review and preserves the nature and scope of
judicial review provided by any other Federal statutes.
Sec. 623. Regulatory analysis
A. Background
This section lays out the requirements for agencies to
conduct regulatory analysis, including cost-benefit analysis,
risk assessment, and substitution risk analysis when issuing
proposed and final major rules. The Committee believes that
better use of these important decisionmaking tools will lead to
a significantly more efficient and effective regulatory
process.
The Committee also recognizes that many of the problems
with the regulatory process can be traced to the failure of
agencies to consider all of the potential effects of their
rules before promulgation. The cost-benefit analysis is
intended to provide a framework for the agency to assess the
impact of its rule on the economy and society as a whole. The
concept of cost-benefit analysis has developed over the past
several administrations to the point where some very
sophisticated analyses have been prepared. The Committee
intends that the analysis be used by agencies to consider
alternative regulatory approaches, to compare the benefits and
costs of such approaches, and to produce better
decisions.64
---------------------------------------------------------------------------
\64\ When well used, cost-benefit analysis is a highly effective
tool to increase the efficiency and effectiveness of regulation. See,
e.g., Resources for the Future, Economic Analysis at EPA (Richard D.
Morgenstern, ed. 1997); The Greening of Industry: A Risk-Management
Approach, Harv. Univ. Press (John D. Graham & Jennifer Hartwell, eds.
1997). One EPA study found that ``the return to society from improved
environmental regulations is more than one thousand times EPA's
investment in cost-benefit analysis.'' See, U.S. Environmental
Protection Agency, ``EPA's Use of Cost-Benefit Analysis: 1981-1986''
(Aug. 1987), at p. 5-2.
---------------------------------------------------------------------------
A satisfactory cost-benefit analysis would enable the
agency to make an informed judgment whether the benefits of the
rule justify its costs, and whether the rule substantially
achieves the statutory objectives in the most cost-effective
manner, or with the greatest net benefits. This determination
is based on the whole rulemaking record.
To fulfill its potential for improving the regulatory
process, the preliminary cost-benefitanalysis must be made
public by the agency to allow comment and criticism by interested
parties. As more information is submitted to support or rebut the
analysis, it and the final rule will be improved. The preliminary cost-
benefit analysis must be summarized in the notice of proposed
rulemaking.
The bill requires the cost-benefit analysis to be developed
by the agency during the development of the rule. The cost-
benefit analysis must guide the agency decision-making process,
not provide a post-hoc rationalization for a decision made
before the analysis was prepared. Once completed, the final
cost-benefit analysis must be made public with the statement of
basis and purpose accompanying the rule. An executive summary
of the analysis must be published with the proposed and final
rules in the Federal Register. If the analysis is properly
performed, it will provide an excellent brief in support of the
agency's factual conclusions and policy choices. The cost-
benefit analysis required by this legislation will help to
identify questions clearly, to describe assumptions made, and
then to clarify the rationale justifying the proposed action so
it is open for public debate. An agency must have this
information before it, along with other relevant information,
in order to make an informed choice.65
---------------------------------------------------------------------------
\65\ See, e.g. Risk Commission Report, Vol. 1, pp. 29-36; Vol. 2,
p. 93.
---------------------------------------------------------------------------
B. Framework for conducting cost-benefit analysis
The first step, outlined in subsection 623(a), is for
agencies, before publishing a notice of proposed rulemaking, to
determine whether the rule is or is not a major rule under
subsection 621(7)(A)--that is, whether the rule is likely to
have a gross annual effect on the economy of $100,000,000 or
more in reasonably quantifiable costs. If the rule does not
fall within subsection 621(7)(A), then the agency must
determine whether the rule is a major rule under subsection
621(7)(B).
If the agency does not determine a rule to be major,
subsection 623(a)(2) allows the Director of OMB to exercise the
same authority not later than 30 days after the close of the
comment period for the rule. This provision is designed to
ensure effective Executive Branch oversight of the requirements
of the Act. A notice of any major rule determination shall be
published in the Federal Register, as a part of the notice of
proposed rulemaking where possible, and such notice shall
include a succinct explanation of the agency's or the
Director's action.
Both the preliminary and final cost-benefit analysis should
address in detail the issues presented by the regulation,
including the need for the rule, the various alternative
approaches (including the potential incremental costs and
benefits of each), the legal basis for agency action, and an
assessment of the benefits and costs of the proposed action.
The analysis should provide an objective, critical, and
impartial discussion of the regulatory problem and of the
potential solutions.
Although basically parallel, the preliminary and final
cost-benefit analyses differ in several important respects. In
most instances, the quality of analysis and data relevant to
the analysis will improve between the time a rule is first
proposed and when it is finally issued. Later estimates
typically apply better data sources more sophisticated
analyses. This tends to improve the accuracy and reliability of
estimates, often substantially. To a large degree, such
additional information will be provided by peer review, public
comments, or other material developed by the agency. Thus, the
later analysis should generally be more complete. In addition,
the final analysis should address significant comments
submitted on the preliminary analysis. The preliminary cost-
benefit and cost-effectiveness evaluations required by
subsection 623(b)(2) will be followed by the formal
determinations required by the final cost-benefit analysis. The
final determinations, of course, should consider any additional
data received by the agency since the publication of the
preliminary cost-benefit analysis and risk assessments.
C. Content of the cost-benefit analysis
Subsection 623(b) requires the agency to place an initial
regulatory analysis 66 in the file of a major rule
and publish in the Federal Register a summary of such analysis.
The agency then must provide an opportunity for interested
persons to comment pursuant to section 553 of title 5, United
States Code. This Subsection reflects the Committee's firm
conviction that sound analysis of the benefits and costs of
various alternative regulatory options before the rule is
proposed is essential to reasoned decision making. An agency
needs this information, along with other relevant information,
to make the best regulatory choice.67
---------------------------------------------------------------------------
\66\ A regulatory analysis under this legislation encompasses a
cost-benefit analysis, cost-benefit determinations, any risk
assessment, and, if applicable, a substitution risk analysis.
\67\ The Risk Commission Report emphasizes the importance of
evaluating the costs and benefits of regulatory options before making a
decision; this is an essential feature of the Commission's framework
for environmental health risk management. See Vol 1, at 29-36; Vol. 2,
at 93-101.
---------------------------------------------------------------------------
According to subsection 623(b)(2), each initial regulatory
analysis must contain three major items: (1) a cost-benefit
analysis; (2) a risk assessment, if required; and (3)
information on any substitution risks.
Under subsection 623(b)(2)(A), each initial cost-benefit
analysis shall contain 5 major components:
(i) An analysis of the benefits of the proposed rule;
(ii) An analysis of the costs;
(iii) An evaluation of the relationship of the incremental
benefits of the proposed rule to its costs, taking into account
the results of any risk assessment, including the
determinations whether the identified benefits of the proposed
rule justify its identified costs; whether the proposed rule is
likely to substantially achieve the rule making objective in a
more cost-effective manner, or with greater net benefits, than
other reasonable alternatives considered by the agency; and
whether the rule adopts a flexible regulatory option.
(iv) An evaluation of the incremental benefits and costs of
a reasonable number of reasonable alternatives reflecting the
range options that would achieve the objectives of the statute
as addressed by the rulemaking, including alternatives that
require no government action; provide flexibility for small
entities under the Regulatory Flexibility Act; provide
flexibility for State, local or tribal agencies delegated to
administer a Federal program; employ flexible regulatory
options; and assure protection of sensitive subpopulations, or
populations exposed to multiple and cumulative risks.
(v) A description of the scientific or economic evaluations
or information on which the agency substantially relied in the
cost-benefit analysis and risk assessment, and an explanation
of how the agency reached the determinations under subsection
(d).
In addition to the cost-benefit analysis, if the rule
requires a risk assessment under section 624, that assessment
must be incorporated into the regulatory analysis under
subsection 623(b)(2)(B).
Finally, Subsection 623(b)(2)(C) requires the agency to
identify and evaluate substitution risks. The analysis of
substitution risks is an important part of the rational
decisionmaking framework established by this legislation. The
Committee believes that if an agency properly identifies and
evaluates the potentially adverse health, safety, or
environmental effects of a regulatory option, the agency will
be best prepared to make a regulatory decision that accounts
for such substitution risks. The Committee is concerned that
government has not always been sensitive to substitution risks
caused or exacerbated by certain regulatory
actions.68 The agency must explicitly identify a
substitution risk, provided there is reasonably available
scientific information on the risk, such as in the scientific
literature or as provided during the public comment period. The
phrase ``reasonably available to the agency'' connotes that the
agency is expected to engage in an affirmative and reasonably
thorough search for information on potential substitution
risks, but the search does not have to be exhaustive.
---------------------------------------------------------------------------
\68\ Cass R. Sunstein, ``Health-Health Tradeoffs,'' 63 U. Chi. L.
Rev. 1533 (1996). See also, John D. Graham & Jonathan Weiner, Risk
Versus Risk: Tradeoffs in Protecting Health and the Environment, Harv.
Univ. Press (1995). One example of the substitution risk problem is the
asbestos scare in the early 1980s. Government scientists argued that
asbestos exposure could cause thousands of deaths. Public alarm led
Congress to pass a sweeping law that led cities and states to spend
between $15 and $20 billion to remove asbestos from public buildings.
But about three years later, EPA officials confirmed that asbestos
removal had been a very costly mistake. Ripping out asbestos raised the
risk to the public because asbestos fibers became airborne during
removal. Removing the asbestos also delayed the opening of many schools
and other buildings. See Gregg Easterbrook, A Moment on the Earth: The
Coming Age of Environmental Optimism, 250-53 (1995); Stephen Breyer,
Breaking the Vicious Circle: Toward Effective Risk Regulation, 12-13
(1993).
---------------------------------------------------------------------------
1. Identification of the problem
Every cost-benefit analysis, whether preliminary or final,
should begin with a discussion of the nature of the problem.
The agency should identify those persons that the underlying
statute and the regulation is intended to benefit and discuss
the nature of the harm that likely will occur if no action is
taken. The analysis should identify the cause or causes of the
problem and possible solutions.
The agencies should identify the statutory authority relied
upon to promulgate the regulation. The agency should briefly
explain why its proposals are within its statutory jurisdiction
and are consistent with congressional intent. A similar
analysis should be done for each significant alternative.
2. Benefits
The heart of a cost-benefit analysis is a review and
discussion of the benefits and costs of the proposed rule and
the reasonable alternatives considered by the agency, including
an attempt to balance and compare those costs and benefits.
Subsections 623(b)(2)(A)(i), (A)(iii), (A)(iv), and (d) require
the agency to analyze and estimate the incremental benefits of
a rule and its alternatives. Economists have noted that the
valuation and calculation of benefits generally pose the
greatest problem in preparing a cost-benefit analysis, although
cost estimates also can be difficult. The benefits of
regulation--particularly environmental, safety, and health
standards--can be substantial, yet difficult to calculate. The
Committee does not expect all cost-benefit analyses will assign
numerical values to all projected benefits. The agencies should
use a rule of reason. When some aspect of a benefit cannot be
quantified, the agency should describe the benefit in detail,
state what significance it attributes to the nonquantifable
aspects of the benefit, and explain the basis for its
conclusion of this point. Those benefits that cannot be
quantified should be described precisely and succinctly. If the
agency provides a monetary or other quantitative estimate, the
analysis should include the methodological justification. The
ranges of predictions and margins of error also should be
specified. The cost, benefit, or risk assessment information
relied on by the agency, whether quantifiable or
nonquantifiable, should be supported by material that would
allow the public to assess the accuracy, reliability, and
validity of such information.
The agency should bear in mind that, just as markets may
not function perfectly, neither do regulatory programs. When
considering the benefits of regulating, agencies should not
compare imperfect markets or externalities with idealized,
perfectly functioning regulatory programs. Recognizing these
limitations, the agency should make a reasonable attempt to
predict the real-world results of the rule in the cost-benefit
analysis.
3. Costs
Subsections 623(b)(2)(A)(ii), (A)(iii), (A)(iv), and (d)
make clear that the cost-benefit analysis should address
several critical issues in assessing the costs of a regulation.
The cost-benefit analysis should look beyond the immediate
compliance costs of regulation and attempt to quantify, or at
least identify, the significant direct and indirect costs and
adverse effects which may result from the rule.
Agencies should estimate the total costs of compliance and
opportunity costs. Agencies also should estimate costs to
government units, including costs of compliance,
administration, enforcement, or lost tax revenue.
It is conceivable that some agency actions could impose
costs in the form of new risks to public health, safety, or the
environment. These risks should be viewed as increasing the net
cost of the regulatory alternative. Alternatively, reducing the
compliance burden imposed on one group or sector of the economy
may increase the burden on another; those costs also should be
considered.
Agencies should consider lost benefits as a cost.
Opportunity costs can be difficult to project but also can be
among the most significant costs of regulation. The inefficient
use of resources, and investment disincentives, can have a
significant impact on the economy.
4. Alternatives
Subsection 623(b)(2)(A)(iv) requires the preliminary cost-
benefit analysis to contain a briefdescription of alternatives
that reflect the range of the agency's discretion for achieving the
objective of the statute as addressed by the rulemaking. Agencies must
consider alternatives proposed by the public, but they also should take
the initiative to develop alternatives that could achieve the statutory
objective as addressed by the rulemaking in a more cost-effective
manner. In the past, agencies have sometimes adopted rules without
seriously considering alternatives that could more effectively achieve
the statutory goals in a less costly manner. This provision is intended
to compel agencies to seek out and consider a ``reasonable number'' of
such alternative approaches, particularly flexible options. The
legislation focuses the agency's discussion on a `` reasonable number''
of alternatives so that agencies are not forced to engage in limitless
or wasteful discussions of theoretical regulatory alternatives. At the
same time, the Committee cautions the agencies against using this
provision to justify ignoring compelling alternatives or using the
cost-benefit analysis as a post-hoc rationalization for a pre-
determined political decision.
Under this subsection, the agency should evaluate the
incremental benefits and costs of a reasonable number of
reasonable alternatives reflecting the range of the agency's
discretion, including, where feasible,69
alternatives that--(I) require no government action; (II)
provide flexibility for small entities under the Regulatory
Flexibility Act; (III) provide flexibility for State, local or
tribal agencies delegated to administer a Federal program; and
(IV) employ flexible regulatory options; and (V) assure
protection of sensitive subpopulations, or populations exposed
to multiple and cumulative risks.
---------------------------------------------------------------------------
\69\ The qualifier ``where feasible'' in Subsection
623(b)(2)(A)(iv) reflects the Committee's intent that the alternatives
must be both legally feasible, as well as technically feasible.
---------------------------------------------------------------------------
Alternatives should be identified and considered to
determine if such alternatives could reduce the net costs of
the regulation. Alternative levels and methods of compliance
may be appropriate. The alternative of having no regulation
should be a starting point in the analysis. There may be
existing voluntary,70 market, judicial, state, or
local regulatory mechanisms that could adequately resolve the
problem identified by the agency for action.
---------------------------------------------------------------------------
\70\ Some agencies have successfully used voluntary programs, such
as EPA's 33/50 Program, to achieve substantial reductions in pollution
in a cost-effective, flexible manner. See Testimony of Carol M.
Browner, Administrator, U.S. EPA, before the Senate Committee on
Governmental Affairs, March 8, 1995.
---------------------------------------------------------------------------
In recent years, agencies have developed a number of
innovative regulatory techniques to make regulatory programs
less costly and more effective. For example, performance-based
standards can be used instead of design standards to reduce
compliance cost while still meeting regulatory goals. Market-
based mechanisms, such as the sale of marketable permits, have
been used to reduce the costs of pollution control while
meeting or exceeding regulatory goals.
While far from complete, a fundamental shift is taking
place in the way federal regulators go about their business, a
shift that this legislation is intended to encourage. In the
past, agencies too often reached for a single tool, command-
and-control regulation, relying on administrative sanctions
imposed through formal enforcement procedures, to solve any
regulatory problem that arose. Traditional regulation, while
necessary and appropriate in some cases, can be time-consuming
and costly to both stakeholders and governments, and can create
disincentives to innovation. Command-and-control regulation is
frequently less effective and more costly than more flexible
approaches.
5. Analysis of Flexible Regulatory Options
The specific reference in section 623(d)(1)(A)(iii) to
consider flexible regulatory options, such as market-based
mechanisms and performance-based standards, reflects not only
the Committee's belief in the importance of considering these
options to design regulatory programs, but also the specific
steps agencies must follow so that these options will be
consistently considered when formulating major rules. When the
agency is developing a major rule, subsection 623(d)(1)(A)(iii)
requires the agency to determine whether the rule adopts a
flexible regulatory option. Subsection 623(d)(2)(C) requires
the agency to describe any flexible regulatory option
considered by the agency and to explain why that option was not
adopted. If agencies fulfill the requirement of setting forth
the extent to which the designs of proposed regulatory programs
incorporate flexible regulatory options, then each rulemaking
process, as well as the record created therein, necessarily
should reflect discussion and analysis of flexible regulatory
options. Since the Committee believes that such alternatives
have the potential to produce better performing and more cost-
effective regulatory programs, then flexible regulatory options
will be an important standard against which agency design
efforts can be judged.
6. Scientific or Economic Evaluations or Information
Subsection 623(b)(2)(A)(v) has 2 major purposes. First, it
promotes the public's right to know the key information
underlying important regulatory decisions. Second, it helps
protect against the use of invalid scientific or economic
assumptions by requiring an agency to describe what information
the agency relied on in making its cost-benefit determinations
under section 623(d), and to explain how that information
supported the agency's conclusions. This requirement is
intended to help ensure the accuracy and scientific validity of
the data and studies upon which the agency relies.
7. Cost-Benefit Determinations
Subsections 623(b)(2)(A)(iii) and (iv) and 623(d) are the
heart of the cost-benefit requirements of this legislation.
They take the agencies one step beyond the descriptive
exercises of other subsections. They serve the critical goals
of promoting the public's right to know how and why agencies
make important regulatory decisions; enhancing the quality of
information underlying agency decisions; and increasing the
accountability of government to the public it is there to
serve.
Subsection 623(d) requires that, in the final cost-benefit
analysis for a major rule, the agency must make a three-fold
determination based on the whole rulemaking record: (1) whether
the benefits of the rule justify its costs; (2) whether the
rule will achieve the objective in a more cost-effective
manner, or with greater net benefits, than the other
alternatives before the agency; and (3) whether the rule adopts
a flexible regulatory option. This requirement mirrors that in
subsection 623(b)(2)(A)(iii) for the preliminary cost-benefit
analysis issued in connection with the notice of proposed
rulemaking for a major rule.
In the first requirement, the choice of the word
``justify'' is an important one. It conveys two concepts:
first, that precise quantification of costs and benefits is not
mandated where it is not possible; second, that agencies may
bring to bear certain judgmental factors to supplement their
numerical analysis in making the required determination.
The second requirement, that the rule achieve the objective
``in a more cost-effective manner, or with greater net
benefits, than the other reasonable alternatives considered by
the agency'' also is not a purely quantitative exercise that
focuses only on costs. The agency is not required to adopt the
alternative with the lowest compliance costs where another
alternative provides substantially greater benefits. The term
``cost-effective'' implies a balancing and weighing of not only
the cost of each alternative considered, but also the differing
degrees of effectiveness of each such alternative.71
---------------------------------------------------------------------------
\71\ The concept of ``cost-effectiveness'' is fully consistent with
providing protective and responsible regulatory standards. Cost-
effectiveness does not require the smallest incremental ratio of cost
to effectiveness when mutually exclusive alternatives are compared. See
Hearing before the Senate Committee on Governmental Affairs, September
12, 1997, at 300-01 (Letter of John D. Graham, Harvard Center for Risk
Analysis).
---------------------------------------------------------------------------
The third requirement, discussed above, reflects the
Committee's intent to promote flexible regulatory options. Such
options hold great promise to be more efficient and effective
than traditional command-and-control approaches.
The Committee is aware that there may be limits to
quantifying certain benefits, as well as costs. However, this
does not mean that agencies are free to act arbitrarily or in
the absence of appropriate record support in making their
determinations under subsections 623(b)(2)(A)(iii) and (iv) and
623(d). An agency's cost-benefit determinations must be
``reasonable.'' By imposing this requirement of reasonableness,
the Committee intends that the agency will engage in ``reasoned
decision making.'' To satisfy this standard, an agency must
explore a reasonable range of alternatives, apply clearly
articulate and understandable criteria, and explain the reasons
why it has reached the determinations required under
subsections 623(b)(2)(A)(iii) & (iv) and 623(d).
The Committee realizes that in some cases it will not be
possible or desirable to attempt to quantify all of the costs
or benefits of a regulatory proposal or of the reasonable
alternatives to it. Although nonquantifiable, such costs and
benefits are not to be ignored; they must be described in the
cost-benefit analysis, identified in as precise a manner as
possible, and considered in making the determinations required
by section 623(d). Such determinations need not be made
primarily on a numerical or mathematical basis. The Committee
has made clear that net benefits analysis under subsection
623(d) is not limited to quantifiable effects. This is
consistent with the definitions of ``benefit'' and ``cost.''
The Committee recognizes that regulations sometimes
implement Congressional policy choices that are not consistent
with efficiency criteria. For example, Congress may provide an
economic incentive to create networks and infrastructure
facilities available to Americans in both rural and urban
areas. This policy choice may impose minor quantifiable costs
on the entire population in order to provide significant
nonquantifiable benefits to discrete populations and to ensure
that the country benefits from truly national networks,
infrastructure, services, and opportunities therefrom. The
Committee does not intend that the provisions of this
legislation, particularly the cost-benefit analysis
requirements, override Congress' policy choice.
Quantifiable costs and benefits should be made in the most
appropriate units of measurement and specify the ranges of
predictions and explain the margin of error involved in the
quantification methods and in the estimates used.72
For example, a hypothetical cost-benefit analysis might
describe one of the quantifiable benefits of a regulation as
``cases of serious injury reduced.'' The most precise estimate
may be the prediction that actual benefits will be within a
range of ``ten to fifty cases annually'' (this is the ``range
of prediction''). The probability that the number of cases
reduced will actually be within this range may be eighty
percent.
---------------------------------------------------------------------------
\72\ See M. Granger Morgan & Max Herrion, Uncertainty: A Guide to
Dealing with Uncertainty in Quantitative Risk and Policy Analysis,
Cambridge Univ. Press (1990).
---------------------------------------------------------------------------
Reducing costs and benefits to common units of measurement
can make the analytical and evaluative exercise more useful and
understandable. Hence, efforts should be made to translate
costs and benefits into monetary or other concrete terms where
appropriate. For example, benefits that consist of reducing or
controlling adverse effects on health or the environment could
be described in the first instance by estimating, using the
risk assessment procedures of this legislation, the degree to
which the rule would reduce the risk that such effects would
occur.
These requirements recognize that quantification of costs
and benefits is far from an exact science. As stated elsewhere
in this Report, the Committee intends a reasonable analysis and
comparison employing the degree of precision appropriate to
each situation. The requirements also recognize that past
regulatory analyses have not always adequately disclosed the
imprecisions inherent in numerical estimates or the assumptions
built into the methodologies used to arrive at them. The
significant assumptions and uncertainties in the analysis
should be prominently displayed, a requirement paralleling the
directive in subsection 627(d) that the agency's cost-benefit
determination be ``reasonable.''
Subsection 623(e) provides a practical mechanism to provide
the public with better information about regulatory decisions.
That information needs to be provided in a way that is
understandable and accessible to the public. In the past, the
critical information underlying rulemakings often has been
buried in long, technical documents in large agency rulemaking
files. 73 This does not serve the public's interest,
nor does it serve the interests of Congress, stakeholders, or
the President. In fact, it could inhibit communication among
relevant decision makers inside and outside the agency, whether
they be technical experts, legal counsel or policy makers.
Subsection 623(e) addresses this problem by requiring a
succinct executive summary of the regulatory analysis. The
Committee intends that the executive summary be a useful tool
to communicate the important information about the rulemaking
to the public, stakeholders, Congress, the President, and the
relevant decision makers. The minimal information to be
provided includes: (1) the benefits and costs of the rule, and
any determinations required under subsection 623(d); (2) the
expected risk reduced and the key conclusions of any risk
assessment; (3) the benefits and costs of reasonable
alternatives; and (4) the key assumptions and scientific
information upon which the agency relied. In addressing the key
scientific information and assumptions, the agency should
discuss significant uncertainties and the quality of the
science or economics that is the basis of the regulatory
analysis, including whether experts are divided over competing
paradigms.
---------------------------------------------------------------------------
\73\ See Testimony of L. Nye Stevens, Director, Federal Management
and Workforce Issues, General Government Division, GAO, before the
Senate Committee on Governmental Affairs, September 12, 1997; Statement
of L. Nye Stevens, Director, Federal Management and Workforce Issues,
General Government Division, GAO, before the Senate Committee on
Governmental Affairs, February 24, 1998; GAO, Cost-Benefit Analysis Can
Be Useful in Assessing Environmental Regulations, Despite Limitations,
GAO/RCED-84-62 (April 6, 1984) (recommending that regulatory analyses
contain executive summaries that recognize all benefits and costs,
including nonquantifiable; identify a range of values for benefits and
costs subject to uncertainty, as well as sources of uncertainty; and
compare all feasible alternatives); GAO, Air Pollution: Information
Contained in EPA's Regulatory Impact Analyses Can Be Made Clearer, GAO/
RCED-97-38 (April 1997) (finding deficiencies in EPA regulatory
analyses and reiterating 1984 GAO recommendations). See also, GAO,
Regulatory Reform: Changes Made to Agencies' Rules Are Not Always
Clearly Documented, GAO/GGD-98-31 (Jan. 1998) (finding that selected
federal agencies usually did not comply with requirements of E.O. 12866
to identify for the public ``in a complete, clear, and simple manner''
the substantive changes made to regulatory actions while under review
at OMB's OIRA, and to identify the changes made at the suggestion or
recommendation of OIRA).
---------------------------------------------------------------------------
Subsection 623(f)(1) provides a limited exemption from
compliance with the requirements of this legislation prior to
issuance of the rule where: (1) the agency for good cause finds
that conducting the analysis under this legislation before the
rule becomes effective is impracticable or contrary to an
important public interest; and (2) the agency publishes the
rule in the Federal Register with such finding and a succinct
explanation of the reasons for the finding. The Committee
merely intends to provide sufficient flexibility for agencies
to respond to a true emergency when a rule must be promulgated
without awaiting completion of the analysis. This exemption
closely tracks the category of rules exempted from the notice
and comment procedures of the Administrative Procedure Act, and
the Committee does not expect this exemption to be used often.
Subsection 623(f)(2) requires that, if a major rule is
adopted under subsection 623(f)(1) without prior compliance
with the legislation, then the agency shall comply with this
legislation as promptly as possible unless the OMB Director
determines that compliance would be clearly unreasonable. This
is a very narrow exception to avoid clearly unreasonable
situations where a costly analysis would be required for a rule
that would not be in effect when the analysis was completed.
Subsection 623(g) incorporates and extends the consultation
requirements of Section 204 of the Unfunded Mandates Reform Act
of 1995 (2 U.S.C. Sec. 1534). Agencies must develop, maintain
and use effective processes and solicit meaningful and timely
input of State, local and tribal governments (or their
designated employees with authority to act on their behalf)
into the development of any regulatory proposals that contain
significant Federal intergovernmental mandates. Such processes
and consultations shall be consistent with Section 204 of the
Unfunded Mandates Reform Act and therefore shall be exempt from
the Federal Advisory Committee Act. The Committee believes that
federal agency consultation with State, local, and tribal
governments before a decision is made will improve the quality,
fairness, and responsiveness of federal regulations. In many
respects, State, local, and tribal officials are closer to the
public; they also are often burdened with unfunded mandates
imposed by regulations or withimplementing and enforcing them.
The term ``significant regulatory proposal'' is substantially broader
than the term ``major rule,'' which triggers the cost-benefit
requirements of this legislation. Accordingly, the consultation
requirements of this legislation apply to agency actions exempted from
the cost-benefit requirements of this legislation.
Section 624. Risk assessment
Risk assessment is a widely recognized tool to structure
information for regulatory decision making related to the
environment, health and safety. The acceptance of risk
assessment as a standard tool can be traced back to the seminal
report issued by the National Academy of Sciences in 1983: Risk
Assessment in the Federal Government: Managing the Process. The
report presented a conceptually sound and logical approach that
has been widely adopted by federal and state agencies to assess
environmental, health, and safety risks.
Fifteen years after publication of the NAS risk report,
there is general agreement that the risk assessment process
needs to be refined. The process should be better understood
and more accountable. Risk assessment can be most useful when
those who rely on it to inform the risk management process
understand the strengths and limitations of risk assessment,
and use it accordingly. Decision makers should at least
understand that the process must rely on assumptions and cannot
completely be divorced from assessors' values. Decision makers
must understand what assumptions were used in the assessment in
question, and what values they reflect; that the risk estimate
is expressed as a range and distribution; and that variability
is expressed to the degree that it is known, i.e., how many and
what kind of persons (e.g., children) will likely be at
significantly higher or lower risk than the hypothetical
average individual. Risk managers must take all of those
factors into account in making a decision, along with
political, economic, and social factors extrinsic to the risk
assessment.
In recent years, many studies have supported the use of
risk assessment and recommended improvements to the process. In
1993, the Carnegie Commission on Science, Technology, and
Government issued Risk and the Environment: Improving
Regulatory Decision Making. In 1994, the NAS issued Science and
Judgment in Risk Assessment to review and evaluate the risk
assessment methods of EPA. In March 1995, the Harvard Center
for Risk Analysis issued Reform of Risk Regulation: Achieving
More Protection at Less Cost. The OSTP also issued a brief
report entitled, ``Science, Risk, and Public Policy.'' In 1997,
the Presidential/Congressional Commission on Risk Assessment
and Risk Management issued the report entitled, Risk Assessment
and Risk Management in Regulatory Decision-Making. Many of the
risk assessment provisions of this legislation are strongly
supported by findings and recommendations of these and other
reports.
Section 624 defines which agency actions must follow the
basic principles in this legislation. Subsection (a)(1)(A)
states that the risk assessment principles of this legislation
apply to: (i) proposed and final major rules the primary
purpose of which is to address health, safety, or environmental
risk; and (ii) risk assessments not the basis of a rule making
that the OMB Director reasonably anticipates are likely to have
an annual effect on the economy of $100 million or more in
reasonably quantifiable costs and that the Director determines
shall be subject to the requirements of Section 624.
The Committee recognizes that risk assessments are not
necessary for rules that do not have the primary purpose to
address health, safety or environmental risk. At the Committee
hearing on S. 746, the concern was raised that S. 746 would
require a risk assessment for Toxic Release Inventory (``TRI'')
reporting rules issued under the Emergency Planning and
Community Right to Know Act. This law requires that covered
entities report, not control, the levels of certain chemicals
emitted from a facility. The primary purpose of such rules is
not to address risks but to disclose information. At the
Committee hearing and markup, Senators Levin and Thompson agree
that S. 746 does not mandate a risk assessment for TRI
rules.74
---------------------------------------------------------------------------
\74\ Hearing before the Senate Committee on Governmental Affairs,
``S. 746, the Regulatory Improvement Act of 1999,'' April 21, 1999.
---------------------------------------------------------------------------
The Committee also recognizes that some risk assessments
can have a significant effect even though they are not
associated with a major rule. Under Subsection (a)(1)(A)(ii),
such ``stand alone'' risk assessments also would have to comply
with the risk principles of S. 746 if the risk assessment is
likely to have a $100 million effect on the economy. This could
occur, for example, where a risk assessment may establish the
basis for significant regulatory actions at the Federal, state,
or international level.
The Committee intends to promote the most advanced and
scientifically valid techniques for performing the wide variety
of risk assessments covered by this legislation. The Committee
does not intend to deter agencies from using the forms of risk
assessment appropriate to their respective regulatory
decisions. It does intend that the methodology be credible and
understandable, and its limitations be made known to the
public.
Subsection (a)(1)(B) sets out two general principles for
risk assessments. This first principle provides that a risk
assessment shall be conducted in a manner that promotes
rational and informed risk management decisions and informed
public input into and understanding of the process. This
recognizes that risk assessments play an important role as a
tool for regulatory decision making, as well as for
communicating information to the public about risks.
The second general principle provides that in determining
the scope and level of analysis of a risk assessment, the
significance and complexity of the decision must be considered
as well as the need to inform the public adequately; the need
for expedition; and the nature of the risk being assessed.
75 This provision acknowledges that some risk
assessments need to be done with greater rigor than others.
Differently stated, the level of effort required for a risk
assessment depends on what is at stake. In some cases, very
severe risks can be identified and managed with relatively
simple risk assessments because the stakeholders agree that the
danger is great enough that no further analysis is needed.
Often, the risks requiring detailed analysis are those that are
marginal on a cost-benefit scale: in these cases, credible,
detailed analyses can be crucial to satisfying stakeholders.
The Committee cautions the agencies against construing this
provision as excusing noncompliance with the provisions of
section 624 or other provisions of this legislation.
---------------------------------------------------------------------------
\75\ See OSTP report, ``Principles in Devising Risk Policy,'' at 17
(``The level of effort should be commensurate with the severity of the
risks and costs to society.'') The Risk Commission Report also supports
this principle. See Vol. 2, at 63 (``Deciding to go forward with a risk
assessment is a risk-management decision, and scaling the effort to the
importance of the problem, with respect to scientific issues and
regulatory impact, is crucial.''); Vol. 2, at 21 (``The level of detail
considered in a risk assessment and included in the risk
characterization should be commensurate with the problem's importance,
expected health or environmental impact, expected economic or social
impact, urgency, and level of controversy, as well as with the expected
impact and cost of protective measures.'').
---------------------------------------------------------------------------
To avoid unnecessary duplication of effort, Subsection
(a)(2) provides that an agency does not have to prepare a new
risk assessment for a final rule where: (1) the final rule is
substantially similar to the proposed rule with respect to the
risk being addressed; (2) the risk assessment performed for the
proposed rule is consistent with the provisions in Subchapter
II; and (3) a new risk assessment is not necessary to address
comments submitted during the comment period.
Subsection (b) requires each agency to ``consider . . . all
relevant, reliable and reasonably available scientific
information'' and to describe the basis for selecting that
scientific information. This subsection promotes three basic
principles. First, the agency must make a thorough search for
relevant data. The agency should make a reasonable attempt to
gather data from informed parties and may solicit information
through the Federal Register. Second, the agency should assess
whether the data are relevant and reliable. And third, if the
data are relevant and reliable, the agency should consider and
analyze all those data in the risk assessment. Data can be
``reliable'' if they are well understood and generally
supported in the scientific community; come from well
recognized, credible sources; or are of sufficient quality that
the results could be reproduced. 76
---------------------------------------------------------------------------
\76\ See Risk Commission Report, Vol. 1, at 38 (``Because so many
judgments must be based on limited information, it is critical that all
reliable information be considered. Risk assessors and economists are
responsible for providing decision-makers with the best technical
information available or reasonably attainable, including evaluations
of the weight of the evidence that supports different assumptions and
conclusions.'')
---------------------------------------------------------------------------
The Committee understands that even reliable data will vary
in quality, relevancy and applicability. The definition of
``risk assessment'' in Section 621(9) contemplates that an
agency will use a careful analysis of the weight of the
evidence to evaluate the information it has. 77 In
considering the scientific information, the agencies should
evaluate the data and apply the appropriate weight to them in
the risk assessment.
---------------------------------------------------------------------------
\77\ The Risk Commission Report provides examples of the kinds of
considerations entailed in making judgments on the basis of the weight
of the scientific evidence in a toxicity study: quality of the toxicity
study; appropriateness of the toxicity study methods; consistency of
results across studies; biological plausibility of statistical
associations; and similarity of results to responses and effects in
humans. See Vol. 2, at 20.
---------------------------------------------------------------------------
Agencies make assumptions in conducting risk assessments to
overcome a paucity of data or a lack of scientific
understanding about such things as causality or basic
biological mechanisms. As Subsection (b) establishes, the
agency should consider all relevant, reliable and reasonably
available data. If the agency concludes that information is not
relevant or reliable, the agency should explain how and why it
so concluded. When the agency needs to use assumptions in risk
assessment, Subsection (c) sets out the appropriate treatment
of the assumptions.
Subsection (c) does not dictate which assumptions an agency
shall use. Rather, it requires the agency to disclose pertinent
information about the significant assumptions so that anyone
relying on the risk assessment can better evaluate the validity
of the assumptions and their effect on the risk assessment.
Accordingly, for a significant assumption, the agency must: (1)
identify the scientific basis, and the policy basis (if any),
as well as the extent to which the assumption is validated by
or conflicts with empirical data; (2) explain the basis for
choosing among possible assumptions and/or combining an
assumption with other assumptions; and (3) describe reasonable
alternative assumptions that would have had a significant
effect on the results of the risk assessment, and those that
were considered but not selected by the agency for use in the
risk assessment.
Finally, Subsection (c)(2) establishes the agency's
obligation to update the assumptions it uses to reflect new
data or new scientific understandings. 78 It
requires the agency to revise its assumptions to incorporate
all relevant and reliable scientific information as it becomes
reasonably available. Subsection (c)(2) is intended to keep
agency assumptions current. It is not intended to create a
counterproductive and never-ending cycle of revisions. It is
intended to promote credible and reliable risk assessments.
---------------------------------------------------------------------------
\78\ The Committee supports the conclusions of Risk Commission
Report, which states: ``Agencies should continue to move away from the
hypothetical . . . toward more realistic assumptions based on available
scientific data.'' Vol. 2, at iv. As Science and Judgment in risk
Assessment clearly acknowledges, ``Over time, the choice of defaults
should have decreasing impact on regulatory decision-making. As
scientific knowledge increases, uncertainty diminishes. Better data and
increased understanding of biological mechanisms should enable risk
assessments that are less dependent on default assumptions and more
accurate as predictions of human risk.'' (p. 90).
---------------------------------------------------------------------------
Subsection (d) requires that an agency provide notice to
the public of a risk assessment, and the agency must solicit
relevant and reliable data from the public. The agency must
consider the data in conducting the risk assessment. The
purpose is to make the process more transparent and
accountable. 79
---------------------------------------------------------------------------
\79\ The Committee received comments on the need for a more
transparent risk assessment process that would allow for greater public
input. The Risk Commission Report strongly supports stakeholder
(public) involvement at all stages of risk management. To avoid the
politicization of risk assessments, however, the Commission noted that
``stakeholders play an important role in providing information that
should be used in risk assessments and in identifying specific health
and ecological concerns'' but should not participate directly in the
risk assessment itself. See Vol. 2, at 21 (``Stakeholders play an
important role in providing information that should be used in risk
assessments and in identifying specific health and ecological concerns
they would like to see addressed.''); id., at 185.
---------------------------------------------------------------------------
Subsection (e) mandates some of the basic contents of the
document describing the risk assessment. This subsection and
subsections (c) and (f) are critical to the transparency in the
risk assessment. They will allow the public and agency decision
makers to understand the full scope and dimensions of the
problem that the agency is addressing. Subsection (e) sets out
five pieces of information the agency risk assessment must
disclose:
(1) A description of the hazard of concern--that is,
the problem being addressed.
(2) A description of the populations or natural
resources that are the subject of the risk assessment.
Consistent with subsection (f), ``populations'' would
include the population that could be exposed to the
hazard and, as appropriate, highly exposed or sensitive
subpopulations.
(3) An explanation of the exposure scenarios used in
the risk assessment, an estimate of the population or
natural resource corresponding to each exposure
scenario, and an estimate of the likelihood that the
exposure scenario would actually occur. The Committee
is aware that the concept of ``exposure'' has been more
associated with assessments of risks from pollutants or
disease agents. However, the Committee believes that it
also is applicable to risks from harmful events. For
example, passengers in a car are exposed to passenger
side airbag injuries; workers who work around
electrical machinery are exposed to injuries from
inadvertent start-ups during repairs; and vehicle
passengers or downstream residents may be exposed to
the potential harm from the collapse of a bridge. The
Committee broadly interprets the term ``exposure.''
(4) A description of the nature and severity of the
harm that could occur as a result of exposure to the
hazard. By ``nature'' the Committee means the type of
adverse affect, such as disease, physical harm or
ecosystem damage, that could be attributed to the
hazard. By ``severity'' the Committee means the
seriousness of the harm--not the likelihood--including
whether the harm is reversible.
(5) A description of the major uncertainties in each
component of the risk assessment and their influence on
the results of the assessment. This requirement will
help inform the public and agency decision maker how
certain the risk is. It also will help identify areas
where additional research or data could significantly
improve the quality and reliability of the risk
assessment. 80
\80\ In ``Science, Risk and Public Policy,'' OSTP emphasized the
importance of describing the uncertainties inherent in risk
assessments, stating ``Variation in risk estimates also arises from
choices of assumptions and methods to address and treat uncertainty in
available scientific data. Risk assessors may develop different
estimates of risk because they employ different (but equally
justifiable) assumptions.'' (p. 9).
---------------------------------------------------------------------------
The final product of a risk assessment should be a set of
numeric estimates which, along with the information required
under Subsection (e), constitutes the risk characterization.
Traditionally, agency regulatory decisions have been based on
the estimate of the risk. Subsection (f) describes the form the
risk estimate shall take. In the past, risk assessments
resulted in risk estimates that were a single value, such as
one-in-ten-thousand, or for some toxicological assessments, a
``safe'' dose or exposure level. The Committee believes that
reliance on single point estimates may conceal important
information from the public and the decision maker, such as the
degree of uncertainty about the estimate, how different
populations might be affected differently, or what policy
judgments are embodied in the estimate. For example, to be
protective, agencies routinely have used conservative
assumptions where there were uncertainties or suspected
variability in exposed individuals. The decision to be
protective may well be the correct one, but embedding this
important policy decision in the risk estimate (the
``science'') is not transparent to the public or agency
decision makers. 81
---------------------------------------------------------------------------
\81\ See Risk and the Environment: Improving Regulatory Decision
Making: ``Regulatory agencies should report a range of risk estimates
when assessing risk and communicating it to the public. How risk
estimates, whether derived from an inventory or not, are conveyed to
the public significantly affects the way citizens perceive those risks.
Single-value risk estimates reported to the public do not provide an
indication of the degree of uncertainty associated with the estimate.
Such numbers do not convey the conservative nature of some risk
estimates.'' (p. 87); see also Science and Judgment in Risk Assessment:
``EPA should make uncertainties explicit and present them as accurately
and fully as is feasible and needed for risk management decision-
making. To the greatest extent feasible, EPA should present
quantitative, as opposed to qualitative, representations of
uncertainty.'' (p. 185).
---------------------------------------------------------------------------
The tools of probabilistic risk assessments are now
sufficiently well-developed that agencies often can supply a
multidimensional descriptive estimate of the risk--one that
fully conveys both the range and likely distribution of the
risk. The risk manager should have as complete a picture of the
risk as possible, avoiding, for example, the simple
presentation of a single-point risk estimate that could
overstate or understate the true risk. Accordingly, Subsection
(f) requires that ``to the extent scientifically appropriate,''
which should be typical, agencies must provide such estimates.
Specifically, agencies are required to provide:
(1) The estimate of risk as one or more reasonable
ranges and, if feasible, probability distributions,
reflecting variabilities and uncertainties. By
``reasonable'' the Committee intends that the ranges
and distributions convey a reasonably accurate picture
of the risk, one that neither overstates nor
understates the risk. The reasonable ranges and
distributions would incorporate all of the data and
alternative assumptions used in the risk assessment.
One of the underlying premises of this legislation is
that more information leads to better decisions. Risk
information should at least be presented as a range,
but this bill reflects the preference that agencies
should strive to obtain sufficient information to
provide probability distributions. Such distributions,
when accurately reflecting variability and uncertainty,
give decision makers and the public a more complete
picture of the risks. Accordingly, the bill requires
the agency to provide a probability distribution where
feasible. The reference to ``one or more'' ranges and
distributions reflects that more than one distribution
may be needed to demonstrate fundamental uncertainties
or to provide specialized information for relevant
subpopulations, as described in subsection (f)(2).
(2) The central 82 and high end estimates
for each range and distribution and a description of
the relevant exposure scenario for the potentially
exposed population to which the range and distribution
estimate applies.83 The Committee believes
that the public and the agency decision maker will make
more informed decisions if they know about the central
and high- end estimates of each range and distribution
and the exposure of particularly affected populations.
---------------------------------------------------------------------------
\82\ A ``central estimate of risk'' is: the mean or average of the
distribution; or a number which contains multiple estimates of risk
based on different assumptions, weighted by their relative
plausibility; or any estimate judged to be most representative of the
distribution. See, e.g., Charles A. Holloway, Decision Making Under
Uncertainty: Models and Choices (1979), at 76, 214, 91-127; Theodore
Colton, Statistics in Medicine (1974), at 28-31. The central estimate
should neither understate nor overstate the risk, but rather, should
provide the risk manager and the public with the expected risk. See
Science and Judgment in Risk Assessment, at 170-75.
\83\ See EPA, Policy for Risk Characterization (March 21, 1995), at
2 (``Information should be presented on the range of exposures derived
from exposure scenarios and on the use of multiple risk descriptors
(e.g., central tendency, high-end of individual risk, population risk,
important subgroups (if known) . . .'').
---------------------------------------------------------------------------
(3) A description of qualitative factors that
influenced the ranges, distribution and likelihood of
the risk. Such qualitative factors may include: choice
of data sets; choice of extrapolation models; choice of
statistical cutoff point for validity; choice of end
point; choice of default assumptions, and so on. This
paragraph promotes the core philosophy of this
legislation--namely, that more information and greater
transparency will improve the quality of agency
decision making.
To help the public and the agency decision maker to better
understand the nature and magnitude of the risks that are the
subject of a risk assessment, Subsection (g) requires agencies
to compare the risk to other risks ``familiar to and routinely
encountered by the general public.'' The agency should disclose
the critical features of the compared risks, including whether
they are voluntary or involuntary, newly discovered or well
understood, and reversible or irreversible.84
---------------------------------------------------------------------------
\84\ See, e.g., National Research Council, Improving Risk
Communication, 165-79 (1989).
---------------------------------------------------------------------------
Comparing risks in this manner helps the agency understand
whether it is addressing the right problems in the most
effective way. It also helps the public understand the
dimensions of the risk and whether the agency is focusing its
efforts on the right problems.85 The Committee
intended to underscore the public communication value of risk
comparisons and therefore required that the comparison be
familiar to and routinely encountered risks. The Committee
expects the agencies to select appropriate comparisons that
provide the best contextual information to the public.
---------------------------------------------------------------------------
\85\ One of the key recommendations of the Commission Report was
that the problems a regulation is intended to address should be placed
in their ``public health and ecological context.'' Vol. 1, at 4. For
example, in the environmental area the Report suggests four questions
for an agency to ask and answer:
---------------------------------------------------------------------------
Is the population exposed to the same pollutant from other
sources?
Is exposure to the pollutant also occurring from other
environmental media?
Do other pollutants from the same sources pose additional
risks to the population of concern?
How great a risk does the problem pose compared to other
similar risks that the community?
Vol 1, at 9-10.
Sec. 625. Peer review
This section specifies that agency heads must develop a
systematic program for independent peer review of all risk
assessments covered by S. 746 and of cost-benefit analyses
conducted for major rules costing $500 million or
more.86 Central to the peer review program should be
review by an adequate number of individual experts from
relevant scientific and technical disciplines, through formal
or informal devices. Peer reviewers must be selected on the
basis of their expertise in the sciences or economics relevant
to the regulatory decision. The participants must be broadly
representative of the scientific and technical views relevant
to the decision at hand and independent 87 of the
agency.88
---------------------------------------------------------------------------
\86\ Peer review is a widely endorsed component of risk assessment
and cost-benefit analysis. See, e.g., Risk Commission Report, Vol 2, at
103 (``Peer review of economic and social science information should
have as high a priority as peer review of health, ecological, and
engineering information.''); National Research Council, Valuing Health
Risks, Costs, and Benefits for Environmental Decision Making (1990), at
207 (``benefit-cost analysis should be subject to systematic,
consistent, formal peer review''); American Enterprise Institute &
Brookings Institution, An Agenda for Regulatory Reform (1997), at 13
(the president and Congress should adopt procedures to peer review
regulatory analyses); John D. Graham, ``Making Sense of Risk: An Agenda
for Congress,'' in Risks, Costs, and Lives Saved (Robert W. Hahn, ed.
1996); John D. Graham, Harnessing Science for Environmental Regulation
(1991); Shelia Jasanoff, The Fifth Branch: Science Advisors as
Policymakers, Harv. Univ. Press (1990). As stated in the OSTP
Principles in Devising Risk Policy, ``Appropriate scientific peer
review and guidance are essential to the risk assessment process.'' (p.
17). The Carnegie Commission Report also highlights the importance of
external peer review. The report states, ``A key element in setting
risk-based priorities is science advice, both internal (within the
agency) and external (through science advisory boards and other
mechanisms). External science advisory boards serve a critically
important function in providing regulatory agencies with expert advice
on a range of issues.'' (p. 90).
\87\ Independence from the agency is not intended to preclude use
of established advisory committees like the Science Advisory Board at
EPA. The charter of EPA's Science Advisory Board states that its
objective is to provide ``independent advice to EPA's Administrator on
the scientific and technical aspects of environmental problems and
issues.'' Its membership consists of persons from the private sector
who serve for two year terms. No full time federal employee is
permitted to be on the Science Advisory Board, although most members do
serve as special government employees and are eligible by statute to be
compensated for their services. Permanent advisory committees and the
members of such committees, even though they serve as special
government employees, are not intended to be precluded from serving as
peer reviewers under S. 746.
\88\ See Statement of Dr. Bruce Alberts, President of the National
Academy of Sciences, in Response to Senator Levin's Questions following
February 24, 1998 Hearing on S. 981.
---------------------------------------------------------------------------
At the same time, the bill allows for a variety of
approaches to peer review, including the use of informal
methods. For example, the National Science Foundation (``NSF'')
uses two principal methods for peer review of proposals, by
mail and by panel. In its report to the National Science Board
on the Merit Review System for FY 1997, the NSF reported that
``In `mail only' reviews, peers are sent proposals and asked to
submit written comments to NSF by postal mail,
facsimile,electronic mail, or through FastLane, NSF's Web-based system
for electronic proposal submission and review.'' Many proposals peer
reviewed by the National Science Foundation are done so using a
combination of both mail and panel methods. The peer review
requirements of S. 746 are intended to allow agencies to use peer
review procedures that are commensurate with the significance and
complexity of the subject matter.
The Committee considered in some depth how to draw the line
with respect to possible conflicts of interest of peer
reviewers. S. 981 as introduced provided specifically that
persons with a financial conflict of interest in a rulemaking
could serve as peer reviewers so long as the conflicts were
disclosed to the agency. Many persons who commented on the bill
were not satisfied with that approach as a universal
requirement. After consulting with individuals with expertise
on the practices and conflicts standards of leading agencies
that use peer review widely, including the National Institutes
of Health, the National Academy of Sciences, the National
Science Foundation, and EPA, the Committee concluded that
agencies themselves 89 can adequately address
potential conflicts in a fair and impartial manner, which is
their responsibility today. The Committee is not aware of any
problems with the current conflict of interest standards being
used by federal agencies with respect to peer review, and
expects that agencies new to peer review under S. 746 will seek
guidance from OMB, OSTP and agencies with expertise in the
field.
---------------------------------------------------------------------------
\89\ For example, EPA's approach for addressing possible conflicts
of interest is contained in EPA's recently issued Science Policy
Council Handbook on peer review. It presents alternative approaches to
identifying and resolving potential conflicts, depending upon the
specific situation. The EPA handbook recognizes that ``It is important
that peer reviewers be selected for independence and scientific/
technical expertise.'' U.S. Environmental Protection Agency, Science
Policy Council Handbook, EPA 100-B-98-001 (Jan. 1998), at p. 45. Yet
EPA also acknowledges that ``experts with a stake in the outcome--and
therefore a potential conflict--may be some of the most knowledgeable
and up-to-date experts because they have concrete reasons to maintain
their expertise. Such experts could be used provided the potential
conflicts of interest are disclosed and the peer review panel or group
being used as whole is balanced.'' Id., at p. 48.
---------------------------------------------------------------------------
S. 746 requires that agency peer review programs ensure
that reviews are conducted on a timely basis and that they
contain balanced presentations of all considerations, including
minority reports and an agency response to all significant
comments. In addition, adequate protection must be provided to
ensure that confidential business information and trade secrets
are protected.
Subsection (b)(2) requires the agency to respond in writing
to all significant peer review comments. The agency response
must be made available to the public and be part of the
rulemaking record for purposes of judicial review of any final
agency action.
Subsection (b)(3) provides that where the agency head and
the OMB Director both determine that a cost-benefit analysis,
risk assessment, or any component thereof, previously has been
subjected to adequate peer review, they can exempt them from
the peer review requirements.
Subsection (c) provides for a neutral referee who can
attest to the independence and quality of the peer review. For
each peer review under this section, the agency head shall
include in the rulemaking record a statement by a Federal
officer or employee who is not an employee of the rulemaking
office or program: (1) whether the peer review participants
reflect the independence and expertise required under
subsection (b)(1)(A), and (2) whether the agency has adequately
responded to the peer review comments as required under
subsection (b)(2).
Subsection (d) provides that the formality of the peer
review shall be commensurate with the significance and
complexity of the subject matter.
Subsection (e) provides that the peer reviews required by
this section shall not be subject to the Federal Advisory
Committee Act. With the input of respected scientific and
technical experts, the Committee determined that a FACA
exemption would help expedite peer reviews as well as enhance
their technical rigor. Peer review is not intended to provide
policy advice or analysis to an agency, and it is not a
political debate among interested parties.90
Moreover, the Committee believes that the FACA exemption will
reduce the potential rigidity, time, and expense of peer
reviews.
---------------------------------------------------------------------------
\90\ See, e.g., Statement of Dr. Bruce Alberts, President of the
National Academy of Sciences, in Response to Senator Levin's Questions
following February 24, 1998 Hearing on S. 981. As defined by EPA,
``Peer review is a documented critical review of a specific agency
major scientific and/or technical work product. . . . It is usually
characterized by a one-time interaction or a limited number of
interactions by independent peer reviewers.'' EPA, Science Policy
Council Handbook, at p. 10.
---------------------------------------------------------------------------
Subsection (f) makes clear that statutorily created agency
advisory boards may be considered ``independent of the agency''
under subsection 625(b)(1)(A)(ii). Subsection (g) clarifies
that the status of a person as a contractor or grantee of the
agency shall not by itself exclude such person from serving as
a peer reviewer for such agency because of the requirement
under subsection 625(b)(1)(A)(ii).
Finally, Subsection (h) makes clear that Section 625 does
not mandate more than one peer review of the cost-benefit
analysis or the risk assessment during a rule making. To the
extent feasible, peer reviews under Section 625 shall occur
prior to the notice of proposed rule making.
Sec. 626. Deadlines for rulemaking
For a 2-year period after the effective date of the
legislation, this section extends certain rulemaking deadlines
for up to six months to allow agencies time needed to comply
with the analytical requirements of the legislation. The
affected deadlines include statutory and judicial deadlines for
rulemakings, as well as rulemaking deadlines that would create
an obligation to regulate through individual adjudications. To
avoid any constitutional concerns about extending judicial
deadlines by legislation, subsection (b) authorizes and directs
the United States to ask the relevant court to extend any
deadlines imposed by the court.
The sole purpose of section 626 is to give agencies time to
make a reasonable effort to faithfully fulfill the requirements
of this legislation. The Committee understands that the
legislation asks for better quality and greater openness in
many analyses already done, and in some cases, creates new
obligations. The Committee intends that agencies be given a
reasonable opportunity to develop policies and procedures
adequate to comply with the law. The Committee does not intend
this grace period to be used otherwise to delay decisions or to
compromise the implementation of legal requirements.
Sec. 627. Judicial review
Section 627 establishes the framework for judicial review
of agency compliance with the regulatory analysis, risk
assessment, and peer review requirements of this legislation.
Specifically, Section 627 is addressed solely to judicial
review of ``[c]ompliance by an agency with the provisions of
[Subchapter II].'' To the extent that an agency action is being
challenged on grounds other than alleged noncompliance with the
provisions of Subchapter II, Section 627 would not
apply.91
---------------------------------------------------------------------------
\91\ This point is underscored by the savings clause in Section
622(b), which states: ``Nothing in this subchapter shall be construed
to alter or modify . . . any opportunity for judicial review made
applicable under other statutes.''
---------------------------------------------------------------------------
Subsection (a) sets three basic conditions for judicial
review of agency compliance with the provisions of Subchapter
II: The judicial review must occur--(1) in connection with
review of final agency action; (2) in accordance with the
provisions of Section 627; and (3) in accordance with the
limitations on timing, venue, and scope of review imposed by
the statute authorizing the review. In setting forth the third
condition, the Committee recognizes that in some cases, the
statute authorizing review may not impose any special
limitations on timing, venue, or scope of review; in other
cases, these matters may be addressed in several different
statutes.
Subsection (b) governs the availability and standard of
review of agency ``major rule'' determinations. An agency's
determination of whether a rule is a major rule--and thus
subject to the regulatory analysis and risk assessment
requirements of Subchapter II--is subject to review only in
connection with review of the final agency action to which it
applies. At that time, a court may set aside the agency's
determination of whether the rule is ``major'' only if it is
shown to be arbitrary or capricious.
In close cases, the Committee would expect that the agency
would err on the side of good analysis and avoid the risk of
remand or invalidation of the rule. As a practical matter, the
agency's major rule determination will be consequential where
the agency wrongly determines that a rule is not ``major'' and
does not bother to perform the cost-benefit analysis, cost-
benefit determination, risk assessment, or peer review that
Subchapter II requires for ``major rules.'' In such a case,
Section 627(e) would require the court to remand or invalidate
the rule, unless the court found that such failure to perform
the analysis or assessment, to make the determination, or to
provide for peer review, was not prejudicial.
By contrast, if the agency incorrectly determines that a
rule is ``major,'' the impact on the rule itself is not likely
to be adverse--since a rule would not be remanded or
invalidated just because an agency performed a cost-benefit
analysis and risk assessment, made a cost-benefit
determination, and provided for peer review in circumstances
where such action was not statutorily mandated. After all, the
Executive Branch is free to undertake such actions today even
where not required to do so by statute. Indeed, that is the
premise of a series of executive orders on regulatory analysis
and review that dates back to the Carter Administration, that
grew in the Reagan Administration, and that is currently
embodied in Executive Order 12866.
Under subsection (c), a designation by the Director of OMB
that a rule is a major rule--or the failure to make such a
designation--is not subject to judicial review. If the Director
has designated a rule as ``major,'' the requirements of
Subchapter II that apply to major rules must be met.
Conversely, if neither the Director nor the agency has
designated a rule as ``major,'' and the rule does not fall
within Subsection 621(7)(A), then the requirements of
Subchapter II would not apply.
Subsection (d) provides that any cost-benefit analysis,
cost-benefit determination, or risk assessment required under
Subchapter II for a rule shall not be subject to judicial
review separate from review of the final rule to which the
analysis or assessment applies. Such a cost-benefit analysis,
cost-benefit determination, or risk assessment, however, would
be part of the rulemaking record, and if the final rule to
which they apply is brought before a court for review, the
court would have to consider the analysis, determination, and
any assessment--to the extent relevant--in determining under
the statute granting rule making authority whether the final
rule is arbitrary, capricious, an abuse of discretion, or
unsupported by substantial evidence.92
---------------------------------------------------------------------------
\92\ The ``substantial evidence'' standard would apply in those
cases where a ``substantial evidence'' standard of review is provided
by the enabling statute--such as under the Occupational Safety and
Health Act, 29 U.S.C. Sec. 655(f), or the Toxic Substances Control Act,
15 U.S.C. Sec. 2618(c)--or where it is required by the Administrative
Procedure Act, 5 U.S.C. Sec. 706(2)(E).
The phrase ``under the statute granting the rule making authority''
clarifies that a rule should not be set aside where the action alleged
to be arbitrary, capricious, or an abuse of discretion involves a
matter that cannot be relevant to promulgating the rule under the
authorizing statute.
---------------------------------------------------------------------------
Section 627(e) states that if an agency fails to perform
the cost-benefit analysis, cost-benefit determination, or risk
assessment, or to provide for peer review as required under
Subchapter II, the court ``may, giving due regard to
prejudicial error, remand or invalidate the rule.'' The
adequacy of compliance with the specific requirements of the
subchapter shall not otherwise be grounds for remanding or
invalidating a rule under the subchapter. If the court allows
the rule to take effect, the court shall order the agency to
promptly perform such analysis, determination, or assessment or
to provide for peer review. If an agency fails to perform the
cost-benefit analysis, cost-benefit determination, risk
assessment, or peer review, the court may, with due regard to
the principle of prejudicial error, invalidate or remand the
rule. In this respect, S. 746 expands the role of a reviewing
court by directing that a rule may be invalidated in
circumstances where it might not be invalidated under current
law.
Under Section 627, an agency's failure to comply with a
specific requirement of S. 746 regarding how to perform a risk
assessment or cost-benefit analysis would not, in and of
itself, be grounds for invalidating a rule. That is, a rule
could not be invalidated simply because a ``how to''
requirement of Section 623 (governing cost-benefit analyses) or
624 (governing risk assessments) was not met, unless the
statute granting the rule making authority imposes such a
requirement. At the same time, however, in determining whether
the final rule is arbitrary or capricious, the court would be
free to consider the effect that the agency's failure to comply
with any such requirement (e.g., a failure to consider reliable
and reasonably available scientific information) had on the
rulemaking. In addition, of course, the cost-benefit and risk
assessment information would be available to the court and
could be considered in determining whether the final rule is
arbitrary or capricious.
The following three scenarios illustrate how the judicial
review provision of S. 746 is intended to operate.
Scenario (1): S. 746 requires an agency to identify and
evaluate reasonably identifiable substitution risks. Suppose
that during a rulemaking, a person submitted information to the
agency on the possibility of a substitution risk and the agency
ignored it. Could that person later argue in a lawsuit
challenging the rule that the agency action in adopting the
final rule is arbitrary or capricious simply because the agency
violated a requirement of S. 746 when it failed to consider a
legitimate substitution risk?
No. Failure to comply with a specific procedural
requirement of S. 746 regarding how to perform a risk
assessment or cost-benefit analysis would not, in and of
itself, be grounds for invalidating a rule.
However, the person could argue that the agency's failure
to consider the legitimate substitution risk had the effect of
making the resulting rule arbitrary or capricious--whether or
not that failure also violated a specific procedural
requirement of S. 746. Such an argument is available today, and
would continue to be available after S. 746 is enacted.
93
---------------------------------------------------------------------------
\93\ In addition, of course, the failure to consider a substitution
risk could be a ground for invalidating the rule if the statute
granting the rule making authority requires that substitution risks be
considered. See, e.g., the Safe Drinking Water Act of 1996,
Sec. 1412(b)(3)(C)(i), 42 U.S.C. Sec. 300g-1(b)(3)(C)(i).
---------------------------------------------------------------------------
Scenario (2): S. 746 requires agencies, when doing a risk
assessment, to consider ``reliable and reasonably available
scientific information.'' If an agency fails to consider such
information which we know the agency had access to through the
public comment period, can a person argue that the rule should
be remanded or invalidated just because the agency violated a
specific procedural requirement of S. 746 when it failed to
consider such information?
No. As indicated in Scenario (1), failure to comply with
the procedural requirements of S. 746 regarding how to conduct
a risk assessment is not independent grounds for remanding or
invalidating a rule.
On the other hand, the fact that Congress directed agencies
to follow this requirement is an indication that it is
important to the development of a risk estimate on which a
rational and well-informed rulemaking decision can be based.
Depending on the circumstances of the particular case, a court
today might conclude that a rule is arbitrary or capricious
where it is based on a risk assessment that did not consider
reliable and reasonably available scientific information.
Nothing in S. 746 is intended to preclude a court from reaching
the same result in the future. To the contrary, S. 746
specifically directs agencies to consider ``reliable and
reasonably available scientific information'' in conducting
risk assessments, so it does not prevent a court from finding a
rule to be arbitrary or capricious when such information is
ignored.
Scenario (3): S. 746 requires the agency to make a
determination as to whether the benefits of the rule justify
the costs. The agency doesn't make that determination. Can a
person challenge the rule for the failure of the agency to make
that determination based on the requirement of S. 746?
Yes. The bill explicitly states that the failure to make
the determination allows the court to remand or invalidate the
rule.
As the foregoing scenarios illustrate, an agency's failure
to comply with the specific procedural requirements of S. 746
regarding how to conduct a risk assessment or cost-benefit
analysis would not, in and of itself, be grounds for
invalidating a rule. That is, the rule could not be invalidated
under section 627(d) simply because a procedure required by S.
746 had been violated. At the same time, the court could
consider the content of the cost-benefit analysis and risk
assessment, any omissions in such analyses (such as those
discussed in the above scenarios), or the arbitrary treatment
of the content of those analyses, in determining whether the
final rule is arbitrary or capricious. This is true under
current law and would continue to be true once S. 746 is
enacted.
In addition, if an agency fails to perform a required cost-
benefit analysis or risk assessment, does not make a cost-
benefit determination, or does not provide for peer review, a
court could remand or invalidate the rule. In this respect, S.
746 changes the role of a reviewing court by providing that a
rule be remanded or invalidated in circumstances where it might
not be remanded or invalidated under current law.
In sum, in determining whether a rule is arbitrary or
capricious, a court would remain free under S. 746--as it is
under current law--to consider both what the agency did do, as
reflected in the cost-benefit analysis and risk assessment, and
what it did not do, such as failing to consider relevant,
reliable, and reasonably available scientific information. But,
with the exception of cases covered by Section 627(e)--where
remand or invalidation of the rule is allowed--a court would
not remand or invalidate a rule solely on the ground that the
agency had not complied with a specific procedure of S. 746.
Sec. 628. Guidelines, interagency coordination, and research
Subsection 628(a)(1) requires the Director of the Office of
Management and Budget, in consultation with the Council of
Economic Advisors, the Director of the Office of Science and
Technology Policy, and relevant agency heads, to develop and
issue uniform guidelines to implement the cost-benefit
analysis, risk assessment, and peer review requirements of this
legislation. Such guidelines should embody, and expand upon,
principles required by this legislation. The OMB Director is
responsible for overseeing the implementation of these
guidelines, and periodically revising them as appropriate and
as warranted by advances in risk analysis, cost-benefit
analysis, and related fields.
No later than 18 months after issuance of those uniform
guidelines, each agency subject to section 624 is required to
adopt detailed guidelines under subsection 628(a)(2) for risk
assessments as required by section 624. Such guidelines shall
be consistent with the uniform guidelines issued under
subsection 628(a)(1). The Committee expects each agency to
revise these risk assessment guidelines as appropriate and as
warranted by advances in science and risk assessment
methodology.
Subsection (a)(3) requires that all guidelines developed
under subsection (a) must be developed following notice and
public comment. OMB and the agencies are expected to make
diligent efforts to solicit input from all informed parties.
Agencies are not required, however to develop the guidelines
through the legislative rulemaking process. The Committee was
concerned that the APA rulemaking process may be too rigid and
time-consuming for the expeditious development and updating of
risk assessment guidelines. Accordingly, Subsection (a)(3)
makes clear that the development, issuance, and publication of
risk assessment and risk characterization guidelines developed
under this section are subject only to limited judicial review
under section 706(1) of title 5. The Committee expects the
agencies to develop andmaintain state-of-the-art guidelines.
Subsection (b) is designed to improve the conduct,
application, and practice of cost-benefit analysis and risk
assessment across all relevant agencies. Subsection (b)(1)
requires the OMB Director, in consultation with the Council of
Economic Advisors and the Director of the Office of Science and
Technology Policy, to oversee periodic evaluations of the
manner in which agencies are conducting cost-benefit analyses
and risk assessments. Such a survey will allow for a
determination of the scope and adequacy of cost-benefit
analysis and risk assessment practices of the federal agencies.
It also will promote the injection of new scientific and
technical advances into the analytical practices of the
agencies.
Subsections (b)(3) and (b)(4) require OMB to establish with
CEA and OSTP appropriate interagency mechanisms to promote
coordination between agencies and to ensure consistent use of
state-of-the-art cost-benefit and risk assessment practices.
Subsection (c)(1) requires OMB, in consultation with the
agencies, CEA, and OSTP, to develop and periodically evaluate a
strategy to meet agency needs for research and training in
cost-benefit analysis and risk assessment. This strategy should
address the need for research on modeling, the development of
generic data, use of assumptions, the identification and
quantification of uncertainty and variability, and other areas.
OMB also should identify long-term needs to adequately train
individuals in risk assessment techniques.
Subsection (c)(2) requires the OMB, in consultation with
OSTP, to enter a contract with an accredited scientific
institution, to conduct research to: (1) develop a common basis
to assist risk communication related to both carcinogens and
non-carcinogens; and (2) develop methods to appropriately
incorporate risk assessments into related cost-benefit
analyses.94 The OMB shall enter into the contract no
later than 6 months after enactment of section 628, and the
results of the research shall be submitted to OMB and to
Congress no later than 24 months after the date of enactment.
---------------------------------------------------------------------------
\94\ See Risk Commission Report, Vol 2, at 43, 99.
---------------------------------------------------------------------------
Sec. 629. Risk-based priorities study
The Committee believes that setting risk-based priorities
offers an excellent opportunity to promote better allocation of
resources of both the government and the private sector to
increase the protection of human health, safety and the
environment. The importance of such a risk-based approach has
been advocated in numerous studies and
publications,95 as well as in testimony before the
Governmental Affairs Committee.96 The Committee
believes that the tool of comparative risk analysis can help us
find ways to make our health, safety and environmental
protection dollars go farther and provide greater overall
protection, saving even more lives than the current
system.97 As the blue-ribbon Carnegie Commission
panel noted in its report, Risk and the Environment: Improving
Regulatory Decision Making, ``The economic burden of regulation
is so great and the time and money available to address the
many genuine environmental and health threats so limited, that
hard resource allocation choices are imperative.'' (p. 118).
---------------------------------------------------------------------------
\95\ See, e.g., J. Clarence Davies & Jan Mazurek, Pollution Control
in the United States, Resources for the Future (1998), at 101-22; Cass
R. Sunstein, ``Health-Health Tradeoffs,'' U. Chi. L. Rev. 1533 (1996);
Resources for the Future, Comparing Environmental Risks (J. Clarence
Davies, ed. 1996); John D. Graham, ``Making Sense of Risk: An Agenda
for Congress,'' in Risks, Costs and Lives Saved, (Robert W. Hahn, ed.
1996); National Academy of Public Administration, Setting Priorities,
Getting Results (April 1995); Harvard Center for Risk Analysis, Reform
of Risk Regulation: Achieving More Protection at Less Cost (March
1995); Carnegie Commission on Science, Technology, and Government, Risk
and Environment: Improving Regulatory Decisionmaking, Washington, D.C.
(June 1993); Stephen Breyer, Breaking the Vicious Circle: Toward
Effective Risk Regulation, Harv. Univ. Press (1993).
\96\ See, e.g., Testimony of John D. Graham, Director, Harvard
Center for Risk Analysis, before the Senate Committee on Governmental
Affairs, September 12, 1997.
\97\ The need for a national comparative risk analysis was one of
the chief recommendations of the Report of the Harvard Group on Risk
Management Reform entitled, Reform of Risk Regulation: Achieving More
Protection at Less Cost (March 1995). The Harvard report states that
the purpose of such an analysis would be ``to learn how diverse risks
should be compared, how ordi nary citizens should participate in risk
ranking, what inherent limitations to the process might be, and how
guidelines can be developed to govern a broad-based process of risk-
based priority setting in the federal government.'' (p. 27).
---------------------------------------------------------------------------
The 1995 National Academy of Public Administration
(``NAPA'') report to Congress, entitled Setting Priorities,
Getting Results, recommends that the Environmental Protection
Agency use comparative risk analysis to identify priorities and
use the budget process to allocate resources to the agency's
priorities. The NAPA study commends EPA for having pioneered
risk prioritization studies and comparative risk analyses.
However, the report states that during the budgetary process,
EPA did not push for shifts in resources to the higher-priority
programs. The report recommends that Congress ``could enact
specific legislation that would require risk-ranking reports
every two to three years. Congress should use the information
when it passes environmental statutes or reviews EPA's budget
proposals.'' (p. 49).
The purpose of the analyses required by this section is to
provide Congress and the President with the information to make
more informed choices. The Committee anticipates that, among
other things, these analyses will be useful for identifying
unaddressed risks, risks borne disproportionately by a segment
of the population, and research needs. This will provide better
information for deciding where to focus regulatory efforts and
agency resources. Finally, conducted through an open process,
these analyses are likely to enhance public debate about these
choices and ultimately create greater public confidence in
government policy.
The comparative risk study should compare significant risks
to human health, safety or the environment and make
recommendations on setting priorities to reduce them. The
comparison is limited to ``significant'' risks, and the study
should examine which of those risks are the most serious and
most amenable to cost-effective reduction.
Section 629 furthers the use of comparative risk analysis
to inform planning and budgetary decision making. To begin, it
calls for contracting with an accredited scientific institution
to conduct a study with three components. The first and most
important component is a comparative risk analysis, which is a
process to systematically estimate, compare, and rank the size
and severity of risks to provide a common basis for evaluating
strategies for reducing or preventing those risks.98
---------------------------------------------------------------------------
\98\ See OSTP report, Science, Risk, and Public Policy. The report
defines policy trade-offs, and stakeholder concerns. The goal is to
conduct a broad examination of governmental policies and expenditures
to reduce risk. (p. 11).
---------------------------------------------------------------------------
Since the purpose is to assist the Federal government in
evaluating how to use its resources effectively to address the
most serious problems, to the extent feasible, the comparison
should include all such risks that are, or could reasonably be,
addressed by the various agencies and programs whose purpose is
to protect human health and safety or the environment,
including natural resources. Comparative risk analysis is not
purely a scientific undertaking. The Committee believes that,
while hard data will form the underpinnings of the analysis,
public values must also be incorporated when assessing the
relative seriousness of the risks and when setting priorities.
Scientific data alone cannot tell us which risks should be
addressed first, for example: neurological damage, heart
disease, or birth defects; a plane crash or cancer. The
comparative risk analysis should be conducted in a way that
enables public values to be ascertained and considered. This
will require public input into the comparative risk analysis.
Nevertheless, when the analysis is completed, it should be
clear to the public and policy makers which part of the risk
comparison reflects science and which part reflects values.
The second component is a study of methodologies for using
comparative risk analysis to compare dissimilar risks to
further development and use of this tool. Because comparative
risk analysis is still a relatively new science, particularly
when used to compare dissimilar risks, sub section (a)(2)
requires that, even while the comparative risk analysis is
being conducted, a study be done to improve the methods and use
of comparative risk analysis. The Committee anticipates that
this study will draw on the analyses already conducted by
numerous states. The results of this part of the study should
also facilitate risk comparisons required by Section 624(g).
The third component of the study is a set of
recommendations on the use of comparative risk analysis for
setting priorities. These recommendations should provide
sufficient guidance to enable the President, the agency heads,
and Congress to evaluate how to better allocate resources
across agencies and among programs to achieve the most cost-
effective risk prevention and reduction.
To assure its credibility, the study must be conducted by
an accredited body selected by the Director of OMB in
consultation with the Office of Science and Technology Policy.
Subsection (b) requires that the study provide an opportunity
for public comment and public participation. For the
comparative risk analysis to be reliable and credible, the
Committee thinks it is important that the study be conducted
through an open process, utilizing expertise in appropriate
fields, such as toxicology, biology, engineering, medicine,
industrial hygiene and environmental effects. The Committee
also recognizes that experts in the relevant social sciences
may be needed to help incorporate public values into the
process. The analysis should be conducted consistent with the
risk assessment principles in Section 624. The methodologies
and scientific determinations made in the analysis are to be
subjected to external peer review, in compliance with Section
625, and made available for public comment. The results of the
comparative risk analysis under subsection 629(a)(1) should be
presented in a manner that distinguishes between the scientific
conclusions and any policy or value judgments embodied in the
comparisons.99
---------------------------------------------------------------------------
\99\ The Carnegie Commission report, Risk and the Environment,
recommends that agencies ``experiment with different mechanisms for
integrating societal values into the process of setting risk-based
regulatory priorities.'' (p. 89). The report states that value choices
should not be made covertly by unaccountable ``experts.'' The report
offers that ``One possibility is for the experts to make explicit, to
the extent possible, all value judgments and their relative weights in
the ranking process.'' (p. 89).
The 1995 NAPA report supports the Carnegie Commission
recommendation: ``Because comparing risks is a value-laden process as
well as a technical challenge, EPA should conduct its comparative risk
analyses as policy exercises with the active engagement of the public
or its representatives. Doing so would provide legitimate results that
would become a base for agency priorities and budget proposals.'' (p.
49).
---------------------------------------------------------------------------
The study must be completed within three years following
enactment of this section. Within one year thereafter, agencies
are to use the results of the study to inform the agencies in
the development of their budgets and strategic plans and
performance plans under the Government Performance and Results
Act, which should provide an excellent framework for achieving
more cost-effective risk reduction.
Finally, to implement any lessons learned from the
exercise, Subsection 629(d) directs the President to recommend
legislative changes to assist in setting priorities so that the
federal government can ``more effectively and efficiently''
reduce risks to human health, safety, or the environment. The
Committee views this report to Congress as an important element
in setting the federal government's priorities so that we can
achieve the greatest degree of protection for health, safety
and the environment with our resources. Congress needs this
information to evaluate its agenda.
Subchapter III. Executive Oversight
This subchapter establishes in law the responsibility of
the President to supervise the regulatory process of the
federal agencies. Such responsibility includes coordinating
agency regulatory policies and procedures, including those
required by this legislation; developing a process for the
review of rules; and developing and overseeing an annual
government-wide regulatory planning process.
Oversight of the federal regulatory process by the
President, including review of proposed rules by an office
designated by the President, has been in effect in one form or
another for about twenty years. Since 1981, it has been
conducted in a centralized process by OMB through the Office of
Information and Regulatory Affairs under Executive Order Nos.
12291, 12498, and, most recently, 12866. The bill recognizes
that centralized regulatory review has become an integral part
of the Federal regulatory process and provides an important
double-check on the work of the regulatory agencies in the
effort to achieve cost-effective regulations. The Committee is
mindful that in the past, presidents have argued against
regulatory review legislation because of potential inroads on
presidential prerogatives. The Committee believes,however, that
placing a regulatory review mandate into this legislation will help put
to rest arguments about the fundamental nature or need for effective
and transparent regulatory review. Nonetheless, respectful of
separation of powers, the Committee has placed into statute only a
general framework of executive oversight, with basic guidelines for
regulatory review and public disclosure. This allows the President the
flexibility to craft the details and scope of any regulatory review
scheme, consistent with the requirements of this legislation.
Subchapter III has four sections: Section 631, definitions;
Section 632, presidential regulatory review; Section 633,
public disclosure of information; and Section 634, judicial
review.
Section 631 provides several definitions for Subchapter
III. First, it applies the same definitions in Section 551 of
current law and Section 621 of the bill to the provisions in
Subchapter III. The section also defines the term ``regulatory
action'' to include advance notice of proposed rulemaking,
notice of proposed rulemaking, and final rulemaking, including
interim final rulemaking. These are the activities for which
the Director of OMB, acting through the OIRA Administrator, is
responsible to review and coordinate under this subchapter.
Subsection 632(a) makes clear that Subchapter III applies
to all proposed and final major rules, including interim direct
and interim final rules, and to all other rules designated by
the President, acting through the Director, for review.
Subsection 632(b) requires the President to establish a
process for such review and coordination and requires that the
day-to-day responsibility for that reside in the Director of
OMB, acting through the Administrator of OIRA. Section 632(c)
enumerates specific activities that the Director/Administrator
is required to carry out, namely: the development and oversight
of uniform regulatory policies and procedures throughout the
federal government, including those by which each agency shall
comply with the requirements of chapter 6; the development of
policies and procedures for the review of rulemakings or
regulatory actions by the Director/Administrator; and the
development and oversight of an annual government-wide
regulatory planning process. The planning process in 632(c)(3)
is to include:
A summary of and schedule for the promulgation of
major rules.
Agency specific schedules for the review of existing
rules required under section 610 of title 5, United
States Code, and under other authorities.
A summary of regulatory review actions undertaken in
the prior year.
A list of major rules promulgated in the prior year
for which an agency could not make the determinations
that the benefits of a rule justify the costs under
section 623(d) of this Act.
An identification of significant agency noncompliance
with Chapter 6 of title 5, United States Code, in the
prior year.
Recommendations for improving compliance with this
chapter and increasing the efficiency and effectiveness
of the regulatory process.
Section 632(d)(1) states that the OMB review of regulatory
actions should be conducted as expeditiously as practicable and
should be limited to no more than 90 calendar days. Under
subsection (d)(2), the review may be extended by either the
Administrator of OIRA or at the request of the rulemaking
agency to the Administrator, and such extension must be
published promptly in the Federal Register.
Section 633 mandates important disclosure requirements for
the OMB review process. This has been an area of particular
concern to the Committee for almost 20 years, beginning with
President Reagan's issuance of E.O. 12291. Many in Congress
were concerned about guaranteeing the openness of the
regulatory review process to instill public confidence and
equal access in such review. The Committee held numerous
hearings over the years on OMB's review process, culminating in
an agreement in 1986 with then OIRA Administrator, Wendy Gramm,
over basic disclosure procedures specifically identified in a
Memorandum to all agencies and made available to the public.
This debate reemerged in connection with oversight of the
Council on Competitiveness in 1991-92 and consideration of
legislation to require disclosure in regulatory review. In
1996, Senator Thompson, then Chairman of the Subcommittee on
Financial Management and Accountability, conducted oversight on
President Clinton's E.O. 12866 on regulatory review. That
oversight, and related GAO investigations, showed that agencies
were not complying with the disclosure requirements of E.O.
12866.100
---------------------------------------------------------------------------
\100\ See GAO, Regulatory Reform: Changes Made to Agencies' Rules
Are Not Always Clearly Documented, GAO/GGD-98-31 (Jan. 1998); Hearing
before the Senate Committee on Governmental Affairs, Subcommittee on
Financial Management and Accountability, ``Oversight of Regulatory
Review Activities of the Office of Information and Regulatory
Affairs,'' 104th Cong., 2d Sess. (Sept. 25, 1996).
---------------------------------------------------------------------------
The disclosure procedures in the 1986 Gramm memo were
included in E.O. 12866 when it was issued in 1993. Also
included in E.O. 12866 was the additional requirement that the
public be informed on an ongoing basis as to the status of
regulatory actions undergoing review (a requirement never
resolved in the 1986 Gramm memo). Section 633 would codify
those disclosure procedures developed and agreed to over time.
Generically, Subsection 633(a) requires the Director of OMB,
acting through the OIRA Administrator, to establish procedures
for public and agency access to information concerning the
review of regulatory actions. Specifically it requires that
certain elements must be included in such procedures. These
are:
Disclosure to the public on an ongoing basis of
information regarding the status of regulatory actions
undergoing review.
Disclosure to the public no later than publication of
a regulatory action of--(1) all written correspondence
relating to the substance of a regulatory action
(including the drafts of proposed and final rules and
the associated analyses) between the OIRA Administrator
or employees of the Administrator and the regulatory
agency; (2) all written correspondence relating to the
substance of a regulatory action between the
Administrator and employees of the Administrator and
any person not employed by the executive branch of the
Federal Government; and (3) a list identifying the
dates, names of individuals involved, and subject
matter discussed in significant meetings and telephone
conversations relating to the substance of a regulatory
action between the OIRA Administrator or employees of
the Administrator and any person not employed by the
Executive Branch.
Disclosure to the regulatory agency, on a timely
basis of--(1) all written correspondence relating to
the substance of a regulatory action between the
Administrator or employees of the Administrator and any
person not employed by the executive branch of the
Federal Government; and (2) a list identifying the
dates, names of individuals involved, and subject
matter discussed in significant meetings and telephone
conversations, relating to the substance of a
regulatory action between the Administrator or
employees of the Administrator and any person not
employed by the Executive Branch.
Subsection 633(b) requires the rulemaking agency, before
publication of any proposed or final rule, to include in the
rulemaking record the following--
A document identifying in a complete, clear, and
simple manner, the substantive changes between the
draft submitted to the Administrator for review and the
rule subsequently announced.
A document identifying and describing those
substantive changes in the rule that were made as a
result of the regulatory review and a statement if the
Administrator suggested or recommended no changes.
All written correspondence relating to the substance
of a regulatory action between the Administrator and
the agency during the review of the rule, including
drafts of all proposals and associated analyses.
Finally, Subsection 633(c) requires that a representative
of the agency submitting the regulatory action shall be invited
to any meeting relating to the substance of a regulatory action
under review between the Administrator or employees of the
Administrator and any person not employed by the Executive
Branch.
Section 634 states the exercise of the authority granted
under this Subchapter by the President, the OMB Director, or
the OIRA Administrator shall not be subject to judicial review.
Section 3(b). Presidential authority
Section 3(b) provides that nothing in this Act shall limit
the exercise by the President of the authority and
responsibility that the President otherwise possesses under the
Constitution and other laws of the United States with respect
to regulatory policies, procedures, and programs of
departments, agencies, and offices. The President retains the
authority to extend regulatory analysis and review requirements
beyond those established in this Act.
Section 3(c). Technical and conforming amendments
Section 3(c) provides the technical and conforming
amendments to Chapter 6 of title 5, United States Code. Up to
this point, Chapter 6 consisted of the Regulatory Flexibility
Act. With this legislation, Chapter 6 is substantially amended
to create Subchapter I, which includes the regulatory
flexibility analysis. It also creates two new subchapters:
Subchapter II--Regulatory Analysis--and Subchapter III--
Executive Oversight.
section 4. compliance with the unfunded mandates reform act of 1995
To avoid duplicative cost-benefit analyses under the
Unfunded Mandates Reform Act of 1995, Section 4 states that
compliance with the cost-benefit provisions of the Regulatory
Improvement Act constitutes compliance with the cost-benefit
provisions applicable to the private sector in sections 202,
205(a)(2) and 208 of UMRA (2 U.S.C. Sec. Sec. 1532, 1535(a) and
1538).
section 5. report to congress
Section 5 requires that by February 5, 2002, the President,
acting through the Director of the Office of Management and
Budget, shall prepare and submit to Congress an accounting
statement and report containing an estimate of the total annual
incremental benefits and costs of complying with the provisions
of subchapter II of the Regulatory Improvement Act for each
agency.
section 6. effective date
Except as otherwise provided in this legislation, this Act
shall take effect 180 days after the date of the enactment of
this Act, but shall not apply to any agency rule for which a
notice of proposed rule making is published on or before 60
days before the date of enactment of this Act.
VI. Regulatory Impact Statement
Pursuant to paragraph 11(b), rule XXVI of the Standing
Rules of the Senate, the Committee, after due consideration,
concludes that S. 746 will have a significant regulatory
impact.
VII. CBO Cost Estimate
U.S. Congress,
Congressional Budget Office,
Washington, DC, July 9, 1999.
Hon. Fred Thompson,
Chairman, Committee on Governmental Affairs,
U.S. Senate, Washington, DC.
Dear Mr. Chairman: The Congressional Budget Office has
prepared the enclosed cost estimate for S. 746, the Regulatory
Improvement Act of 1999.
If you wish further details on this estimate, we will be
pleased to provide them. The CBO staff contact is John R.
Righter.
Sincerely,
Barry B. Anderson
(For Dan L. Crippen, Director).
S. 746--Regulatory Improvement Act of 1999
Summary--CBO estimates that implementing S. 746 would, on
average, cost about $6 million a year, assuming appropriation
of the necessary amounts. Enacting the bill would not affect
direct spending or receipts; therefore, pay-as-you-go
procedures would not apply. The bill contains no
intergovernmental or private-sector mandates as defined in the
Unfunded Mandates Reform Act (UMRA) and would impose no costs
on the budgets of state, local, or tribal governments.
S. 746 would amend chapter 6 of title 5, U.S. Code, to
require federal agencies to complete specific studies,
including cost-benefit analyses and risk assessments, as part
of the regulatory analysis performed before certain major rules
are issued. The bill would define a major rule as a regulatory
action expected to result in an annual impact on the economy of
$100 million or more in costs, or a rule designated as major by
the Office of Management and Budget (OMB). The bill will exempt
many rules from the new requirements, however, and would
primarily apply to those agencies that issue major rules
governing health, safety, and the environment. In cases where
an agency issues a rule that is expected to have an annual
impact on the economy of $500 million or more in costs, the
legislation would require that the agency submit the rule for
peer review. For major rules covered by S. 746, agency
compliance with the bill's regulatory analysis provisions would
be subject to limited judicial review.
CBO expects that implementing S. 746 would have a small
impact on the federal government's cost to perform regulatory
analyses because the bill would: (1) codify much of existing
practice, (2) generally not apply to so-called minor rules, (3)
exempt most major rules from its review, and (4) allow agencies
to opt out of its requirements in certain situations. Based on
our review of the number and type of major rules issued in
fiscal years 1997 and 1998 and on past costs of regulatory
analyses, CBO estimates that, subject to appropriation of the
necessary amounts, implementing S. 746 would increase the
government's costs to perform regulatory analyses by around $5
million annually. Such costs would result from the additional
documentation and analyses required by S. 746 and from
requiring that independent agencies perform cost-benefit
analyses for certain major rules.
In addition, the bill would require the Office of
Information and Regulatory Affairs (OIRA) within OMB to write
regulations, periodically evaluate training needs at agencies
that perform regulatory analyses, contract for a pair of
studies, submit an accounting statement to the Congress that
contains an estimate of agencies' incremental costs in
complying with the bill's regulatory analysis provisions, and
review applicable major rules issued by independent agencies.
The legislation also would direct agencies that regulate
health, safety, and the environment to devise detailed
guidelines for performing risk-assessment analyses. CBO
estimates that implementing these administrative requirements
would cost federal agencies an average of less than $1 million
a year over the 2000-2004 period.
Under some circumstances, S. 746 could result in additional
costs to federal agencies beyond those in this estimate. OMB
could require that agencies perform risk assessments according
to the bill's detailed procedures for agency actions, other
than major rules, that it anticipates could have an annual
effect on the economy of $100 million or more in costs. CBO
assumes however, that the bill's procedures for conducting risk
assessments would be applied only in the case of major rules.
If OMB required agencies to apply the bill's risk assessment
procedures to other agency actions that include an assessment
of risk, the additional costs would likely be significant. The
estimate also does include costs that might be incurred as the
result of additional judicial review because CBO has no basis
for predicting how many regulatory actions might be challenged
under this bill.
Estimated cost to the Federal Government--CBO estimates
implementing S. 746 would increase the costs of regulatory
analysis at agencies that issue major rules government health,
safety, and the environment, as well as increase federal
reporting and administrative costs. In total, implementing the
bill would require appropriations of about $6 million a year
over the next five years.
Regulatory Analysis
Much of the regulatory analysis and review that would be
required by S. 746 is already required by Executive Order 12866
(``Regulatory Planning and Review'') and the accompanying best
practices for performing economic analyses of significant
regulatory actions (``Economic Analysis of Federal Regulations
Under Executive Order 12866''), as well as title II of UMRA.
In addition, the bill would exempt many federal regulatory
actions from its requirements, including rules that apply to
regulate: (1) military or foreign affairs, (2) federal agency
management or personnel, (3) public property, loans, grants,
benefits, or contracts, (4) governmental receipts, (5) certain
commerce activities, including wages and prices, mergers and
acquisitions, and accounting practices, (6) securities trading,
(7) monetary and federal fiscal policy, (8) banking, and (9)
the removal or introduction of products under the Federal Food,
Drug, and Cosmetic Act.
In addition, the bill would exempt certain regulations of
the Federal Election Commission and the Federal Communications
Commission (FCC) and any rule that an agency must issue at
least annually. Based on a review of the summaries provided by
the General Accounting Office (GAO) of approximately 115 major
rules issued by agencies during fiscal years 1997 and 1998
(GAO's list, which is required by Public Law 104-121, does not
include all major rules issued over the two years), CBO
estimates that at least two-thirds of majors rules would be
exempt from the bill's requirements.
In addition to the specified exemptions, agencies could
exempt rules from the bill's provisions where the more detailed
reviews are either not practical or contrary to an important
public interest. In such cases, the bill would direct the
agency to comply with its provisions as soon as possible after
adopting the rule, unless OMB determines that such compliance
would be unreasonable.
CBO expects that enacting S. 746 would have a small impact
on the cost to perform regulatory analyses for agencies that
issue major rules governing health, safety, and the
environment, such as the Environmental Protection Agency (EPA),
the Occupational Safety and Health Administration, and the
Departments of Health and Human Services, Energy,
Transportation, and Agriculture, as well as certain independent
agencies that are excluded from the requirements of Executive
Order 12866.
Based on our review of the type and number of major rules
issued during fiscal years 1997 and 1998, we expect the bill's
provisions would apply to about 20 rules a year, although the
volume of regulatory activity can fluctuate depending on the
demands on regulatory agencies. On average, we expect that the
EPA would issue about one-third of the major rules covered by
S. 746.
In 1997, CBO published a paper that examined the costs of
85 regulatory impact analyses (RIAs) conducted by selected
agencies (Regulatory Impact Analysis: Costs at Selected
Agencies and Implications for the Legislative Process, March
1997). The cost of these RIAs ranged from $14,000 to $6
million, with the time required to complete them ranging from
six weeks to more than 12 years. (Because the paper did not
attempt to obtain a representative sample of RIAs, it does not
indicate the cost of a typical or average RIA.) Based on our
review of the number and type of rules that would likely be
affected by the provisions of S. 7246, the bill's requirements
for conducting regulatory analyses, and our analysis of the
costs of RIAs, CBO estimates that implementing S. 746 would, on
average, increase regulatory analysis costs for agencies that
issue rules governing health, safety, and the environment by
around $5 million a year. That estimated increase would cover
the costs for health, safety, and environmental agencies to
conduct additional analyses, including assessments of
comparative risks and analysis of substitution risks, as well
as to provide additional documentation of the agencies'
assumptions, models, findings, public comments, and
conclusions. On average, we estimate that the provisions of S.
746 would add a few hundred thousand dollars to the cost of
such rules, although the amount per rule could vary greatly.
Independent agencies, which currently are not required to
prepare a cost-benefit analyses for major rules, would also
need to begin preparing such analyses for a handful of rules
each year. This requirement would predominantly affect the FCC,
although it also would occasionally affect rules issued by the
Nuclear Regulatory Commission (NRC). While implementing the
cost-benefit analysis provisions of S. 746 would increase
regulatory costs at both agencies, particularly the FCC, both
the FCC and NRC are authorized to collect fees to offset the
cost of their regulatory programs. Thus, CBO estimates that
implementing S. 746 would result in no significant net
budgetary effect for independent agencies.
Finally, S. 746 would require that agencies submit for peer
review any cost-benefit analysis or risk assessment developed
for a rule that is covered by the bill's provisions and is
reasonably expected to have an annual impact on the economy of
$500 million or more in costs. The bill would require that the
peer review panel represent all points of view and that
agencies respond in writing to all significant comments from
peer review. With theconcurrence of OMB, an agency could
certify that a cost-benefit analysis or risk assessment has received
adequate peer review outside of the bill's procedures. Based on our
review of GAO's summaries of major rules for fiscal years 1997 and
1998, the bill's requirement for peer review would appear to have
applied to only four of those rules--all issued by EPA, which already
submits its rules for formal peer review. Although the provisions
would, at times, apply to rules issued by other agencies, most or all
of which do not currently submit their rules for peer review, CBO
expects that implementing this provision would result in only a
negligible increase in the annual cost for agencies to issue major
rules.
Reporting, Oversight, and Implementation
S. 746 would impose several reporting and oversight
requirements, which would be performed mostly by OIRA.
Specifically, the bill would require that OIRA:
(1) issue guidelines for cost-benefit analyses, risk
assessments, and peer reviews and periodically evaluate
agency efforts in implementing these guidelines;
(2) develop a strategy to meet agency needs for
research and training in performing regulatory-impact
analyses;
(3) contract with accredited scientific institutions
to study the use of risk assessments and comparative-
risk analysis in performing regulatory analyses;
(4) prepare and submit to the Congress by February 5,
2002, an estimate of the total annual incremental costs
and administrative benefits for each agency of
complying with the bill's provisions; and
(5) review the regulatory analyses of certain major
rules issued by independent agencies.
The bill would require that the results of the research on
risk assessments be forwarded to OMB and the Congress within
two years of enactment and that the results of the research on
comparative-risk analyses be forwarded within three years. In
addition, the bill would require agencies that issue health,
safety, and environmental regulations to adopt within 18 months
detailed guidelines for performing risk assessments as part of
their regulatory impact analyses. In total, CBO estimates that
the bill's reporting, oversight, and implementation
requirements would cost agencies an average of less than $1
million a year over the 2000-2004 period. Such costs would be
about $1 million for each of the next three years, but would
fall below $500,000 in subsequent years.
Pay-as-you-go considerations--None.
Intergovernmental and private-sector impact--S. 746
contains no intergovernmental or private-sector mandates as
defined in UMRA and would impose no costs on the budgets of
state, local, or tribal governments.
Estimate prepared by--John R. Righter.
Estimate approved by--Paul N. Van de Water, Assistant
Director for Budget Analysis.
VIII. Minority Views
We understand that the goal of our colleagues in crafting
S. 746 is to improve the regulatory process. We appreciate the
efforts that the sponsors of this bill have made, including a
number of changes in response to some of the concerns that we
and others expressed about this legislation in the last
Congress. The sponsors and we share the goal of protecting the
health and safety and environment of Americans through
effective and efficient regulation, but how to accomplish this
goal is where there are differences.
S. 746 would require Federal agencies issuing major rules
to take a number of new, complex, and time-consuming analytic
and procedural steps and would authorize more judicial review.
We oppose this bill because of our concerns that the
consequences of this kind of omnibus regulatory reform
legislation will be to threaten the ability of our health and
safety, environmental, and consumer protection agencies to act
in a timely and decisive manner to protect us, our children,
and the natural resources we all cherish.
As elected representatives, we have an obligation to the
people we serve to protect them from harm, and that includes
protecting people from breathing polluted air, drinking
poisonous water, eating contaminated food, working under
hazardous conditions, exposing children to unsafe toys, and
becoming victims of consumer fraud. Americans depend on
regulatory agencies to prevent harm before it occurs. As we
have evaluated and heard testimony on S. 746 and predecessor
legislation in recent years, the risks to these protections of
trying to achieve regulatory reform through one-size-fits-all
requirements has become increasingly evident.
Witnesses testified that the highly prescriptive
requirements for risk assessment, cost-benefit analysis, net
benefit determinations, and peer review could further delay
what is already a slow process of establishing needed public
protections.\1\ For example, the bill would require agencies to
conduct time-consuming risk assessments even where Congress has
decided that regulatory standards should be based on available
technology rather than on estimates of risk. We also heard
concerns that new avenues for judicial review may create
significant new opportunities for opponents of public
protections to challenge them in court. The provisions on peer
review and on OMB review of agency regulations would make the
regulatory process less fair and transparent than it is now.
And perhaps most troubling, witnesses explained that the
emphasis on cost-benefit analysis and net benefits, coupled
with new opportunities for judicial review, could lead agency
officials to choose regulatory alternatives that are less
protective of the public. We believe that the combined effect
of these new hurdles will actually make it more difficult for
the environmental, health, safety, and consumer agencies to
establish the protections Americans want.
---------------------------------------------------------------------------
\1\ Our laws provide a host of procedural protections to make
certain that all individuals have an opportunity to participate in the
rule making process and then to challenge decisions that they believe
are wrong in federal court. Guaranteeing that all citizens have these
procedural due process rights with respect to rule making makes the
process of issuing regulations a lengthy one. It is, therefore, rare
that citizens object that a health or safety agency has acted with too
much speed on their behalf. On the contrary, citizens often complain
that agencies do not act rapidly enough. Sue Doneth, the mother of
hepatitis A victim, and Nancy Donley, whose child died from eating an
E. Coli contaminated hamburger, spoke eloquently to the Committee on
this point. (This testimony was presented at hearing on S. 981 in the
105th Congress, which was a predecessor bill to S. 746.) So did Dr.
Franklin Mirer, Director of the Health and Safety Department of the
United Automobile, Aerospace & Agricultural Implement Workers of
America, who testified that he current standard setting process at OSHA
to protect workers from chemical exposure is stalled and failing to
protect workers.
---------------------------------------------------------------------------
These are not just abstract concerns. The Committee has
heard about a number of specific examples where S. 746 would
adversely affect programs to establish protective regulation--
illustrating how this bill would be the wrong kind of
regulatory reform. Some of these examples will be discussed
below. We have also heard a wide array of environmental
organizations, public health groups, and other public interest
organizations and labor unions express very serious concerns
about the harmful consequences of this legislation.
We offered a number of amendments at Committee markup to
try to fix some of the problems that we identified.
Unfortunately, except for an amendment requiring a report on
the cost of the legislation to agencies, none of the amendments
was adopted. When this legislation comes before the entire
Senate, we will try again to fix the problems that have been
identified by offering a number of amendments.
Furthermore, in considering across-the-board regulatory
reform legislation such as S. 746, which would impose new
analytic and procedural requirements on regulatory agencies
across-the-board, we must also recognize that there will surely
be unforeseen consequences. As far as we know, neither the
proponents of this legislation nor anyone else has produced a
law-by-law analysis showing how this bill would affect
individual programs--whether involving the environment, worker
safety and health, highway and aviation safety, food safety,
protection of nursing home residents, nuclear safety, civil
rights including rights of individuals with a disability, and
all the other areas where we rely on regulation to protect the
public. We are therefore concerned about the unforeseen
consequences--the problems that we may not learn about until
some months or years after the legislation has passed, when the
resulting harm to our efforts to establish essential regulatory
protections will become manifest.
For these reasons we oppose S. 746.
there is a better way to improve regulations
We would all agree on the importance of adopting and
enforcing health and safety and consumer protections in an
equitable, efficient, and fact-based way, that is as open to as
much public understanding and participation as possible. Such
improvements might be called ``regulatory reform.'' But we have
concluded that the better way to enact such regulatory reform
is targeted and in the framework of specific regulatory
statutes, not in an across-the-board omnibus bill such as S.
746.
Statute-by-statute reform does not create problems of over-
inclusiveness, as does omnibus legislation like S. 746. And it
works. The 1996 Safe Drinking Water Act Amendments is an
outstanding example of regulatory reform legislation that was
very targeted and dealt with the problems unique to drinking
water quality. The Congress carefully considered how risk
assessment and cost-benefit analysis could make the statute
more effective and incorporated those principles, based on the
overall objectives and operation of that law. For example, an
issue unique to the Safe Drinking Water Act is the different
capacities of large and small water systems. As a result, the
law specifically tailored the Environmental Protection Agency's
authority to use cost-benefit analysis based on differences in
these systems. We are concerned that such refinement and
targeting will be missed in this type of broad government-wide
proposal.
In another example of specifically tailored regulatory
reform legislation from the 104th Congress, we also passed and
the President signed the 1996 Food Quality Protection Act. In
the enactment of this legislation, like the Safe Drinking Water
Act Amendments, negotiations led to bipartisan agreements for
tailored provisions to increase future cost-effectiveness,
while giving EPA flexibility to address high-priority risks.
The Accountable Pipeline Safety and Partnership Act of 1996 is
yet another example of legislation that included narrowly
targeted regulatory reform provisions. Such refinement and
targeting is impossible in across-the-board regulatory reform
legislation like S. 746, and serious unintended consequences
may result.
Although the statute-by-statute approach may be more time-
consuming and difficult in the short-run than an omnibus bill,
the result is well worth it. The Environment and Public Works
Committee spent three years on the reauthorization process for
the Safe Drinking Water Act, listening to all views on how this
law was or was not working, but the bill passed the Senate
unanimously with the support of virtually every interested
group. The Food Quality Protection Act also passed the Senate
by a wide margin. The importance of that type of consensus
cannot be overstated. Among other advantages, it makes everyone
want to work to implement effectively a law that they supported
and have a stake in.
By contrast, there is no consensus with respect to S. 746.
As we noted, the bill is opposed by a wide array of
environmental, public-health, and other public-interest and
labor organizations. The testimony of one Committee witness,
Patricia Kenworthy on behalf of the National Environmental
Trust, indicates just how far from consensus we are. Ms.
Kenworthy testified: ``it is our belief that this legislation
will result in extensive delays in the time it takes for
regulatory decisions to be made and will thus undermine federal
agencies' ability to protect public health, worker safety and
the environment.'' In a similar vein, Dr. Frank Mirer of the
UAW testified: ``Now our members are asking why legislation is
being considered to make it even more difficult to get new
protections against hazards that put their lives, limbs and
health in danger.''
In addition to the statute-by-statute approach, in recent
years Congress passed and President Clinton signed a number of
more targeted regulatory reform bills to address some of the
concerns raised by the business community and state and local
governments about the regulatory process. The Unfunded Mandates
Reform Act of 1995 includes provisions for cost-benefit
analysis of major rules. The Paperwork Reduction Act, which was
designed, in part, to assure that Federal regulations requiring
the collection of information will minimize the burden on
respondents and maximize the usefulness to agencies, was
reauthorized in 1995, including specific requirements for
paperwork reduction. The Small Business Regulatory Enforcement
Fairness Act (SBREFA), passed in 1996, combines several new
laws intended to ease regulatory burdens on small businesses.
Under this legislation, agencies must write regulations so that
those affected can more easily understand them and know how to
comply, and must establish programs to provide for the
reduction and, in some circumstances, for the waiver of
penalties for violation of requirements by a small entity.
SBREFA also provides for enhanced judicial review for
regulations affecting small businesses and provides for
Congressional review of agency rulemaking whereby Congress
acknowledges and assumes more responsibility for the rules that
agencies issue. Finally, in 1996, 1997, and 1998, Congress
passed regulatory accounting measures requiring OMB to submit a
report on the costs and benefits of regulations.
The Clinton Administration has also undertaken a number of
initiatives to improve the Federal regulatory system. In 1993,
President Clinton issued Executive Order No. 12866 setting
forth a regulatory philosophy that, consistent with existing
law, regulations should be issued only where necessary and be
based on a full assessment of costs and benefits of reasonable
alternatives. This Executive Order is a powerful tool for OMB
to ensure that agencies' regulations both protect public health
and make good economic sense, but the Order does not add new
judicial hurdles for agencies to overcome. The Administration
has also undertaken to improve programs at every regulatory
agency as part of the National Partnership for Reinventing
Government. EPA, for example, has reported that it established
stronger partnerships, especially with States; provided for
greater public access to environmental information; gave more
attention to compliance assistance to help businesses and
communities meet their environmental responsibilities; used
more flexible, tailored approaches to solving environmental
problems; and substantially reduced regulatory paperwork.
Proponents of S. 746 have referred to assertions in GAO
reports that some of these new requirements are not being
implemented as effectively as they could be. We do not know
whether GAO is correct in these conclusions--and we will not
know until this Committee takes the appropriate next step,
which is to conduct oversight hearings to determine how these
laws are working, where the gaps, if any, may be, and whether
more needs to be done. Let us try to make the laws work that we
just passed in the last several years, rather than throwing up
our hands and imposing yet another set of overlapping
requirements on our environmental and public health and safety
agencies.
SPECIFIC CONCERNS WITH S. 746
In addition to our general concerns about the unforeseen
consequences arising from omnibus, across-the-board regulatory
reform legislation, we also have particular concerns about
specific provisions of S. 746.
1. Judicial review
Throughout the years of debate on regulatory reform, many
have expressed their opposition to creating new grounds for
litigation. We fully support the thorough judicial review that
the Administrative Procedure Act (APA) provides for all rules.
Under the APA, an agency's decision will be set aside if it is
arbitrary, capricious, an abuse of discretion, or otherwise not
in accordance with law. Courts find that an agency has passed
these tests if the agency's analyses, assessment and responses
to comments have provided the court with a reasoned discussion
of choices the agency has made and a sufficient explanation of
the reasons for those choices so that the court can conclude
that the agency had a reasonable basis for making its decision.
Any relevant risk assessment or cost-benefit analysis that is
prepared must be sufficient to withstand this APA test.
a. Judicial review of the bill's cost-benefit tests create
new opportunities and arguments for regulated
interests to overturn safeguards in court, and
create incentives for agencies to choose less
protective options.
S. 746 imposes new, more burdensome hurdles that the agency
must overcome, beyond what is required under present law. We
are particularly concerned that the judicial review allowed by
S. 746 could create new opportunities and arguments for
overturning regulations and could result in agencies' selecting
rules that are less protective of the public than those the
agencies would have selected without the bill. S. 746 would do
this, first, by mandating that agencies must justify their
decisions using the specific cost-benefit terminology set forth
in the bill, and then by placing those agency justifications
before the reviewing courts--thereby providing new substantive
arguments that regulated interests could use in challenging the
agency's decision as arbitrary and capricious. In particular,
S. 746 would first require the agency to publish a
determination of: (1) whether the agency's rule is likely to
provide ``benefits that justify the costs,'' (2) whether the
rule is likely to achieve its objectives in ``a more cost-
effective manner'' or with ``greater net benefits'' than
alternatives, and (3) whether the rule ``adopts a flexible
regulatory option.'' If any answer is no, the agency head must:
(4) explain why. The bill then requires that each of these
determinations and explanations by the agency must go into the
record for judicial review. By requiring the agency to make
determinations about whether these demanding new tests are
satisfied, and then by making the agency's determinations
subject to judicial review, S. 746 raises our concern about
unintended consequences: a competent lawyer representing
opponents of the regulation will frequently be able to find
some basis for arguing that the agency's conclusions about
whether the ``benefits justify the costs,'' and whether the
selected rule is ``more cost-effective'' or achieves ``greater
net benefits'' was arbitrary and capricious. We fear that
courts may allow these complex cost-benefit determinations, as
well as the bill's cost-benefit terminology, to be injected
into the argument about whether the rule is arbitrary and
capricious, thereby encouraging new substantive arguments for
challenging an agency's rule.
For example, suppose EPA were setting a standard for
reducing pollution from hazardous industrial waste, and must
choose between a less-stringent standard and a more-stringent
standard. Typically, all of the costs, but only some of the
benefits, would be quantifiable. Let's say, in our example,
that the most significant benefit--avoiding reproductive
problems causing birth defects in children--cannot be fully
quantified, so that the more stringent standard has lower net
quantifiable benefits than the less stringent standard. The law
now generally instructs EPA to develop hazardous-waste
standards ``as may be necessary to protect human health and the
environment.'' Therefore, under current law, if EPA selected
the more stringent alternative, the debate before a reviewing
court would focus on whether it was arbitrary or capricious for
the agency to decide that the standard yielding greater
reduction in reproductive problems and birth defects is
``necessary to protect human health and the environment.''
If S. 746 were enacted, however, EPA would be required to
state the agency's determination whether benefits of the more
stringent standard justify'' costs, and whether the standard is
``more cost-effective'' or has ``greater net benefits'' than
the alternative, and the agency would also have to explain any
decision to adopt a rule that did not pass these tests. Then
the bill instructs that all of these determinations and
explanations by the agency shall be part of the rule making
record for the court to consider in deciding whether the rule
is arbitrary and capricious. S. 746 might thereby direct the
court's attention away from whether the agency's decision
satisfies the standards of applicable environmental law, and
towards evaluating the agency's determinations and explanations
mandated by the economics-based requirements of the bill. Thus,
a court case that under today's law would focus on the mandate
to adopt standards ``as necessary to protect human health and
the environment'' could be transformed into a debate on whether
it was reasonable for the agency to conclude that ``greater net
benefits'' are achieved when the nonquantifiable health and
social values are balanced against the economic costs of the
selected regulatory standard.
We do not believe that the law should grant regulated
entities new opportunities to challenge an EPA decision as
being arbitrary and capricious, by finding fault with the
agency's determination that the chosen standard is ``more cost-
effective'' or would achieve ``greater netbenefits'' than
alternatives, or by arguing that it was unreasonable for the agency to
select a standard that does not pass these new tests. In essence, the
court would be second-guessing whether the agency, for example, was
``correct'' in the quantifiable and nonquantifiable value it assigned
to avoiding birth defects in children and whether the agency properly
balanced that value against economic costs to ascertain the ``net
benefits'' or the ``cost-effectiveness'' of the rule. Current
environmental law does not require the agency to determine whether
those tests are satisfied, and we see no reason to provide litigants
with new opportunities and arguments for overturning good rules. We
oppose giving courts the ability to be the arbiters of fundamental
value decisions such as the value of avoiding birth defects in
children, or the value of preserving a child's IQ, or the value of
seeing a clear Grand Canyon. But this is what could occur under S. 746.
We also oppose creating incentives for agencies to avoid
judicial challenge by adopting less protection for these
values. Again, this is what could occur under S. 746, and is
another concern raised by the judicial review provisions of S.
746--another unintended consequence. The agencies may choose a
less protective option in order to avoid the risks of a court
fight. This could lead to regulations that will be
unnecessarily weakened, resulting in potential dangers to the
public and less protection of the environment.\2\ As Dr. Mirer
testified on behalf of the UAW, the bill would, in the OSHA
context, ``shift the balance in standard setting decisions from
worker protection to industry costs.''
---------------------------------------------------------------------------
\2\ One of the ironies of this bill is that it could actually
discourage use of voluntary, incentive-based programs, despite the
sponsors' clear intention to encourage these programs. One of the
Committee witnesses in hearings last Congress on a predecessor bill to
S. 746, Karen Florini of the Environmental Defense Fund, testified that
the cost-effectiveness or net benefits test, combined with judicial
review, may actually discourage the use of information-based and
incentive-oriented approaches such as the very popular Right-to-Know
laws. She testified: It's typically difficult to predict just how, and
to what extent, incentives will lead to a particular outcome because,
by definition, compliance isn't mandatory. But if you can only
generally describe the benefits, how can you do a `net benefits' or
`cost-effectiveness' determination with enough specificity to withstand
attacks by lawyers seeking to derail the rule?'' Testimony of Karen
Florini, Senior Attorney, Environmental Defense Fund, before the Senate
Committee on Governmental Affairs on February 24, 1998, in Hearing on
S. 981, 105th Cong., 2nd Sess., S. Hrg. 105-486, at page 131.
---------------------------------------------------------------------------
b. Courts might overturn rules if the agency has not
performed the required analyses to the letter of
the statute.
We are also concerned that this bill may authorize the
courts to overturn a rule if the agency fails to perform some
particular requirement in the bill regarding cost-benefit
analysis and risk assessment. This problem is most evident when
the language of S. 746 is compared to provisions considered by
the Committee in prior Congresses. Judicial review language in
the Glenn/Chafee proposal from the 104th Congress, and proposed
by the Administration in a letter on March 6, 1998 from former
Director of OMB Franklin Raines to the Chairman of this
Committee, allowed for judicial remand only if the agency
``entirely fails to perform'' the required cost-benefit
analysis or risk assessment. Earlier Glenn-Chafee language,
introduced in S. 1001 from the 104th Congress, and incorporated
into S. 291 as reported in that Congress by this Committee,
provided for remand only if the required analysis was ``wholly
omitted.''
S. 746 dropped the phrases ``entirely fails to perform''
and ``wholly omitted'' used in these earlier proposals and
instead authorizes judicial remand if the agency ``fails to
perform'' a cost-benefit analysis or risk assessment. Both the
terms ``cost-benefit analysis'' and ``risk assessment'' are
defined in the bill in considerable detail, and an opponent of
an agency's regulation could therefore argue that the agency
failed to perform the ``cost-benefit analysis'' or ``risk
assessment'' based on a failure to perform any one of the
numerous requirements in the bill. One of the Committee
witnesses, David Vladeck, a lawyer with extensive experience
arguing cases in front of courts of appeals, testified that:
``courts are likely to measure whether an agency has
`performed' these analyses against the yardsticks established
in the statute. If the agency has not followed the statute to
the letter, a court might well rule that it has not `performed'
the required analysis and set aside the rule on that basis
alone.''
Senator Lieberman offered an amendment at markup to make
clear that the bill would not give rise to these new bases for
overturning an agency's rule. The amendment would have made
changes in the judicial review provision of the bill like the
Glenn/Chafee judicial review language recommended by the
Administration in the March 6, 1998, letter from former OMB
Director Raines to the Chairman of this Committee.
Unfortunately, the amendment was rejected. The sponsors contend
that their intent is to have limited judicial review.\3\ We
therefore remain troubled by their rejection of an amendment
that would express their intent clearly in the text of the
bill.
---------------------------------------------------------------------------
\3\ ``The [cost-benefit] analysis and [risk] assessment are
included in the rulemaking record, but there is no judicial review of
the content of those items or the procedural steps followed or not
followed by the agency in developing the analysis or assessment. Only
the total failure to actually do the cost-benefit analysis or risk
assessment would allow the court to remand the rule to the agency.''
Statement of Senator Levin upon introducing S. 746, Congressional
Record, page S3482, March 25, 1999 (emphasis added).
---------------------------------------------------------------------------
2. Peer review
We strongly support a process for ensuring that the
agencies' approaches to risk assessment are vetted on a regular
basis with those who are the best in their field and willing to
devote the time to such a review. But we are concerned that
this bill does not achieve such a goal and instead will result
in burdensome new processes for peer review without any
benefits and without appropriate safeguards of fairness and due
process. The reaction to this legislation of many scientists we
have heard from has been strongly negative.\4\
---------------------------------------------------------------------------
\4\ For example, in a letter to the sponsors of S. 981 in the 105th
Congress, a group of scientists, including representatives from across
the country, concluded that the legislation, ``particularly the
provisions governing participation on peer review panels, takes a
peculiar and even damaging approach to science.'' Letter to Senators
Levin and Thompson, signed by 74 scientists, physicians, and public
health professionals, dated March 3, 1998. Most of the problems in S.
981 identified by these scientists remain in S. 746, including: (1)
that the peer review requirements are redundant and wasteful, since the
science and the risk assessments underlying major regulations are
already being vetted routinely by intra-agency review panels and
independent scientific bodies; (2) that the agency's own experts are
excluded from participating on peer review panels, yet employees of the
regulated industry may participate, as long as the federal agency's
procedures permit it; (3) that a requirement in predecessor bills that
the membership of panels be ``balanced'' has been dropped from the
legislation; and (4) that panel deliberations could take place ``behind
closed doors'' and review is made exempt from the Federal Advisory
Committee Act, which guarantees public scrutiny and participation in
established advisory committees. The letter also criticized S. 981
because university researchers might be disqualified from participating
on panels because of research grants received in that university, but
S. 746 has been fixed to make clear that grantees are not disqualified.
---------------------------------------------------------------------------
a. For proposed Occupational Safety and Health standards
(and other rules thoroughly vetted though public
hearing processes), the peer review required by S.
746 would superimpose yet another time-consuming,
burdensome, and completely unnecessary process.
S. 746 would impose new requirements for ``peer review''
applicable to all risk assessments required by the bill and to
cost-benefit analyses of those major rules having a large ($500
million) annual effect on the economy. We are concerned, first,
about the interaction of this peer review requirement with
other established approaches that may not currently be called
peer review, but that serve to provide a similar type of
review--another example where unintended consequences emerge
when we start looking at how this bill would apply to
individual agencies and programs. The purpose of peer review in
S. 746 is to create an opportunity for individual experts from
relevant scientific and technical disciplines to provide their
expert advice to the agency. However, such an opportunity is
already built into the hearing process that the Labor
Department must follow in adopting or amending OSHA standards.
We heard testimony about the OSHA procedures from Dr.
Franklin E. Mirer, Director of the Health and Safety Department
of the UAW. Under current law and procedure, proposed OSHA
standards must be presented in a public hearing if any affected
party requests a hearing. The agency must present evidence
supporting the proposed standard including the health risks,
control measures, cost analyses and other details. Scientific
experts, unions, employers, and individual employees also are
allowed to testify. Any participant in the rule making
proceeding may ask questions of the others, and OSHA staff may
ask questions as well. This OSHA process, which is mandated by
statute and agency regulation, fulfills the same function as
``peer review,'' and the OSHA procedures are better in some
ways than those in S. 746, because the hearing follows public
notice and is open and on the record. To require the agency to
conduct an additional round of peer review to meet the
specifics of the bill would further delay a rule making process
that already takes a very long time. Dr. Mirer testified that
``the extra peer review step mandated by the legislation takes
extra time and extra OSHA and stakeholder resources, which
could be better spent addressing additional hazards.''
Senator Lieberman offered an amendment at markup to
establish that the OSHA hearing procedure and similar
procedures under other laws would be recognized under the bill
as satisfying the peer review requirements. We were
disappointed that the amendment was not accepted.
b. Peer review under S. 746 lacks necessary protections of
due process, fairness, and transparency.
In arguing against Senator Lieberman's amendment at markup,
proponents of S. 746 argued that the OSHA hearing comes too
late in the process because, for peer review to have its
desired effect, it must occur before the proposed rule is
published when agency staff positions remain amenable to
influence. But this argument actually shows how the OSHA
hearing procedure is preferable. Unlike the hearing, peer
review as required by S. 746 is actually contrary to the
principles of due process and public participation established
under the Administrative Procedure Act. The OSHA hearing occurs
only after public notice, is open to the public, and is fully
on the public record. Under the APA, it is at this stage, after
public notice, that interested parties are to submit their
comments and evidence to the agency, and the agency is then
legally obligated to consider these comments and evidence in
deciding whether and how to modify the original proposal. All
interested parties--both the regulated industry and the people
to be protected by the regulation--receive the notice at the
same time and have the same opportunity to submit their views
and information.
By contrast, S. 746 calls for peer review before the public
is even notified of the agency's proposed rule, and authorizes
the peer review to occur without public announcement and
including representatives of the regulated industry. Although
we are convinced that the sponsors of S. 746 advocate this kind
of peer review in a good-faith effort to improve regulations,
we are concerned that the peer review requirements in the bill
would actually foster suspicion that some interested parties
may get unfair access and influence in the rule making; and we
believe it is understandable that Dr. Mirer, on behalf of the
UAW, testified that: ``Quite frankly, this extra step is just
one more foothold for interests who simply want no change and
whose only goal is to stop any new regulation.''
Furthermore, the requirement in the bill that the peer
review panel be independent of the agency could also diminish
the quality of the panel. The federal government has some of
the best scientists in the world and there is no reason to
exclude a scientist working in one office from serving on the
peer review panel reviewing a risk assessment done by another
office.Former OMB Director Raines pointed out in his March 6,
1998 letter to the Chairman of this Committee, that the independence
requirement could mean that in some highly specialized areas, such as
nuclear safety, good peer review would become virtually impossible.
On the other hand, the bill provides no assurance that a
person with a direct financial interest in the outcome of the
rulemaking or employed by an entity with a direct financial
interest in the outcome of the rulemaking will not be allowed
to serve on a peer review panel. This raises serious concerns
about potential conflicts of interest.
Supporters of S. 746 have asserted that decisions about
conflicts of interest are best left to agencies. But agencies
cannot always be relied upon to avoid conflicts of interest
undermining the integrity of peer review. For example, Senator
Lieberman described to the Committee his experience in a Senate
investigation relating to pesticides where there were
allegations that people on the peer review panels, including
some experts from the academic world, were nonetheless also
doing work for the regulated industries.\5\ Furthermore, the
Majority Report gives no assurance that all agencies have any
conflict-of-interest rules, but offers merely the
``expectation'' that agencies new to peer review will seek
guidance from their more experienced sister-agencies.
---------------------------------------------------------------------------
\5\ Hearing on S 981, 105th Cong., 1st Sess., S. Hrg. 105-335,
Sept. 12, 1977, at page 18.
---------------------------------------------------------------------------
Senator Cleland offered an amendment at markup designed to
address some of these concerns by excluding participation in
peer review by individuals with a direct and substantial
conflict of interest. We were disappointed and puzzled that the
sponsors did not accept this amendment.\6\ Furthermore, the
bill provides no safeguards to counteract the potentially
corrosive effects of conflicts of interest in the peer review
panel. The bill does not require that conflicts be disclosed or
that panel meetings be announced and opened to the public or
that the membership of panels be balanced.\7\
---------------------------------------------------------------------------
\6\ The sponsor's argument in response to Senator Cleland's
amendment was puzzling. They quoted from EPA's Peer Review Handbook,
which states that sometimes experts with a ``stake in the outcome--and
therefore a potential conflict'' may be used for peer review because
they may be the most knowledgeable and up-to-date. However, Senator
Cleland's amendment did not exclude those with a ``potential
conflict,'' such as EPA may use in peer review; his amendment would
exclude those with a ``direct'' and ``significant'' conflict of
interest. This limitation in the proposed amendment is similar to the
limitation in this EPA Handbook, which states that in some cases ``the
conflict may be so direct and substantial as to rule out a particular
expert.'' EPA Peer Review Handbook, prepared by the EPA Science Policy
Council, page 48 (January 1998).
\7\ The Glenn/Chafee regulatory reform proposal from the 104th
Congress and S. 981 as introduced in the 105th Congress required that
peer review panels be ``balanced.'' However, this requirement was
dropped from S. 981 as reported by this Committee in the 105th Congress
and, now, from S. 746. Instead, S. 746 requires that the peer review
must ``contain a balanced presentation of all considerations,'' but
does not require that the membership on the panel be balanced.
---------------------------------------------------------------------------
In fact, S. 746 would actually make some peer review less
transparent and balanced than it is under present law. Peer
review carried out by formal and established advisory
committees such as the EPA Science Advisory Board is now
subject the Federal Advisory Committee Act (FACA), which
requires that panel membership be balanced and that meetings be
announced and open to the public. However, if S. 746 were
enacted, even peer review conducted by established advisory
committees would become exempt from these existing guarantees
of fairness and openness.
3. Risk assessments; over-broad and time-consuming analytic
requirements generally
S. 746 would require the agency to conduct a risk
assessment for each proposed and final major rule ``the primary
purpose of which is to address health, safety, or environmental
risk.'' In addition, the Director of OMB would have authority
to impose the prescriptive requirements of the bill on any
agency risk assessment anticipated to have an annual effect on
the economy of $100 million, even if the risk assessment is not
part of a rulemaking.
In the all-too-common situation where adequate scientific
data are not available, the requirements of S. 746 are
ambiguous. It is unclear whether the risk assessment provisions
direct the agency to base its decision on whatever data are
available or to collect additional data. This uncertainty
presents the agency and its scientists with a Hobson's choice.
If they rely on available data, they put the rule in jeopardy
of judicial challenge by regulated interests who will argue
that the agency failed to collect sufficient data to perform a
risk assessment meeting the minimum requirements of S. 746.
However, the only way to limit this risk of judicial remand is
to delay the rule while additional data are collected, even if
the new data are not essential to making any critical
regulatory or policy decision. These serious consequences are
described in a recent report from the Congressional Research
Service (CRS), which concludes: ``Such uncertainty [in S. 746,
as ordered to be reported,] is likely to lead to legal
challenge.'' \8\
---------------------------------------------------------------------------
\8\ The CRS report states:
``Some proposals attempt to ensure scientific objectivity
by mandating it. For example, S. 746, as ordered to be
reported, would require scientists to consider `all
relevant' and `all reliable' scientific data and to perform
an `objective' assessment `based on the weight of the
scientific evidence.' However, the effect of these
legislated mandates on agency behavior is unpredictable due
to the variety of circumstances surrounding risk
assessments and the legal consequences of EPA actions. For
example, the validity of the `weight-of-the-[scientific]-
evidence' approach in practice depends on the quality and
comprehensiveness (or representativeness) of the data.
Therefore, a legal requirement to rely on the approach may
be interpreted by scientists as a directive either to base
decisions on available data even if data are inadequate and
misleading, or to collect additional data to meet minimum
data requirements, even if the aspect of the risk
assessment for which data are unavailable is unimportant to
the risk analysis as a whole or to significant regulatory
or policy decisions. Such uncertainly is likely to lead to
legal challenges.''
CRS Issue Brief, ``The Role of Risk Analysis and Risk Management in
Environmental Protection'' (Order Code IB94036, updated June 9, 1999),
at page CRS-3.
---------------------------------------------------------------------------
a. The risk assessment provisions of S. 746 would impose
burdensome requirements even where those
requirements cannot enhance the rule
In those circumstances where Congress has already mandated
that the regulatory agency must base its standard on best
available performance practices rather than on an assessment of
risk, it makes no sense to require the agency to conduct an
expensive, complex, and time-consuming risk assessment before
the standards can be proposed and adopted. This is a
particularly good demonstration of why one-size-fits-all
regulatory reform may unduly delay critical public protections,
and why it is better to craft targeted reforms appropriate in
the framework of particular regulatory statutes.
In its comments on S. 981 in the last Congress, the
Administration proposed to exclude from the coverage of the
risk-assessment requirements those major rules that are not
premised on the outcome of a risk assessment. (These comments
were presented in the March 6, 1998 letter from former OMB
Director Franklin Raines to the Chairman of this Committee.)
The examples offered in the letter are the technology-based
standards that EPA is required to adopt for toxic air
pollutants and water pollutants.
Let us consider first the example of air toxics regulation.
When amending the Clean Air Act in 1990, Congress recognized
that toxic air pollution was not being adequately controlled.
Literally thousands of pollution sources were releasing
chemicals into the air that were known or suspected causes of
cancer, birth defects, or other serious health problems. Many
of these sources were without controls (despite the
availability of cost-effective technology to control the
pollution), partly because it took too long for the agency to
research and analyze the nature and extent of the risks.
Instead, Congress decided there was already enough evidence of
risk to justify regulating a list of particularly harmful
chemicals and instructed EPA to set standards for sources of
those chemicals based on existing technologies, to reflect the
``maximum achievable control technology'' (MACT).\9\ A second
phase of controls, based on risk, are to be established, only
if necessary, eight years after the MACT standards are
required. Congress made a similar decision in the Clean Water
Act, which requires EPA to set technology-based standards for
water pollutants.
---------------------------------------------------------------------------
\9\ In preparing the MACT standards, the agency is to consider
costs, non-air quality health and environmental impacts, and energy
requirements. See Section 112(d)(2) of the Clean Air Act, 42 U.S.C.
Sec. 7412(d)(2).
---------------------------------------------------------------------------
We worry that if S. 746 applied to the air toxics or water
pollutant programs, EPA could be required to delay issuing
technology-based standards until the agency conducted risk
assessments on the potential for harm posed by each pollutant B
despite Congress's intent to the contrary. We believe that
Congress should not pass omnibus reform legislation that would
effectively overturn earlier decisions enacted by Congress to
require agencies to develop standards without further
consideration of the risk posed by each pollutant. At markup,
Senators Lieberman and Akaka proposed an amendment to exempt an
agency from the risk-assessment requirements insofar as
Congress has not authorized the agency to take risk into
account in developing a regulation. Again, we were disappointed
that this amendment was not adopted.
In arguing against the amendment, a sponsor of S. 746
proposed that risk assessments should be required, even if not
useful to the agency preparing the regulations, because the
assessments will provide useful information to Congress and the
public. However, in situations where Congress decided that the
agency should issue standards to reduce toxic emissions without
evaluating the potential for harm posed by those toxic
emissions, we would oppose holding those public-health
standards hostage while the agency prepares risk assessments
that are irrelevant to the agency's decision and would result
in considerable delay. We believe it is inappropriate for this
Committee, on the record before us, to use this legislation to
effect wholesale changes in the approach enacted by prior
Congresses for regulating air toxics or water pollutants. The
current air toxics provisions, for example, were adopted after
many years of debate. Yet this Committee has not held a single
hearing on whether EPA's technology-based standards are
working. If, on this record, Members of the Committee believe
that further evaluation of the potential for harm posed by
toxic emissions would be valuable not to the agency developing
the rules, but to Congress for legislative purposes or to the
public, then let us consider commissioning the necessary
research in some way that does not delay the issuance of the
essential public-health regulations.
Another example of the harm that could result from over-
broad requirements in S. 746 involves right-to-know laws. At
the hearing on S. 746, Senators Durbin and Lieberman asked
whether the listing of toxic chemicals under the Emergency
Planning and Community Right-to-Know law should be delayed
while complex risk assessments and cost-benefit analyses are
conducted. The purpose of this law is to provide information
about toxic chemicals and releases to local communities and to
encourage voluntary actions by companies to reduce releases of
these chemicals, and the law has been successful.
The sponsors of S. 746 responded (as noted in the Majority
Report) that right-to-know regulations are not covered by the
risk assessment requirements. This seems a highly debatable
point, so Senator Lieberman proposed an amendment at markup to
clarify that the risk assessment requirements would not cover
community right-to-know or similar programs that address risks
to health, safety, and the environment by requiring the
collection, reporting, and dissemination of information. This
amendment would have exempted the community right-to-know law,
OSHA's worker right-to-know requirements, and other
environmental and consumer protection programs that require
reporting, disclosure, and dissemination of information about
health and safety risks and precautions. For example, a number
of programs require labeling of commercial products to provide
consumers accurate information about risks and precautions they
can take.
Unfortunately, this amendment was not adopted. The debate
on the right-to-know law again points to the fundamental
problems with across-the-board regulatory reform; we simply do
not, and cannot, now know the unintended consequences if S. 746
were enacted.
b. Application of S. 746 to ``stand-alone'' risk
assessments could be used to delay scientific
research and analysis important to protecting
public health and the environment and could
engender legal challenge to future regulations
based on that analysis.
The risk assessment requirements of S. 746 apply not only
to a rule, but also to a ``stand-alone'' risk assessment,
provided the OMB Director ``reasonably anticipates'' that the
risk assessment ``is likely to have an annual effect on the
economy of $100,000,000 or more in reasonably quantifiable
costs.'' We fear this provision could be used to significantly
delay important scientific research and analysis needed to
protect human health and the environment.\10\
---------------------------------------------------------------------------
\10\ The Majority Report adds to our concern by stating that the
Director might apply the legislation where a risk assessment ``may
establish the basis for significant regulatory action at the Federal,
state, or international level.'' If this requires our Federal
regulatory agencies and research institutes to demonstrate that their
risk assessments are not likely to be the basis for regulatory actions
aggregating $100 million in annual effect at some future date anywhere
in the world, the legislation could be made applicable to many ``stand-
alone'' risk assessments each year.
---------------------------------------------------------------------------
Moreover, S. 746 imposes no deadline for the OMB Director
to determine whether a ``stand-alone'' risk assessment is
subject to the bill. A Director might assert that he has the
authority to require a Federal regulatory agency or research
institute to comply with the bill's specifications and
procedures even after a risk assessment is well underway or is
even completed. Finally, if regulations are eventually
promulgated based on an earlier risk assessment that the OMB
Director determined was subject to the bill, opponents of the
regulations could attempt to use the judicial review provisions
of S. 746 in seeking to have the regulations overturned. These
provisions of S. 746 are a prescription for disruption and
delay of research, analysis, and regulatory actions by our
public health, safety, and environmental agencies.
Additional concerns about the risk assessment provisions
were expressed in the March 6, 1998 letter from former OMB
Director Franklin Raines, which criticized these provisions for
being too specifically tailored to analysis of cancer risks,
and thus ill-suited to evaluation of other kinds of risks such
as environmental and natural resource protection, worker
safety, or airworthiness. No changes were made in S. 746 to
respond to these concerns.
c. S. 746 does not just provide ``information''; it will
also move agencies towards less protective
decisions and will delay critical rules.
In response to some of our concerns about risk assessments
and other analytic requirements of the bill, proponents contend
that the required analyses are simply designed to provide
``information.'' However, this response is not reassuring, for
several reasons. First, the combination of mandated judicial
review and the new complex regulatory requirements will move
agencies towards making decisions that are less protective of
public health, the environment, wildlife, and consumers. This
is a substantive outcome about how protective our laws will be,
not simply a requirement to provide information. Furthermore,
other Congresses determined in the context of specific laws, in
areas such as clean air, clean water, and wildlife protection,
that some of the analysis required in this bill should not be
done. Congress determined, in some instances, that the agencies
already had enough information on which to proceed, or that
delay was too risky to public health and the environment, or
that the delay resulting from requiring the analyses (such as
risk assessment) had actually proved counter-productive to
Congress's goals, or that the analyses were fundamentally
incompatible with the basic guarantees provided by this nation
(among other reasons). Senator Torricelli highlighted this last
concern at markup by demonstrating the fundamental
inappropriateness of applying cost-benefit analysis to civil
rights protections. Thus, to impose the analytical requirements
of this legislation can both affect the substantive outcome of
a rulemaking and impose undue delay, in some instances in
direct contradiction of decisions made by prior Congresses.
Proponents of S. 746 also respond to our concerns about
delay by referring to testimony of one witness, Dr. Lester
Crawford, expressing his opinion that the bill would have sped
up, rather than delayed, issuance of food-safety regulations
including the U.S. Department of Agriculture's 1996 meat and
Poultry Hazard Analysis Critical Control Point (HACCP) rule. He
observed that agencies' proposed regulatory actions were often
held up during review within the executive departments or at
OMB by arguments over such matters as cost-benefit analysis,
risk assessment, and judicial review; and he expressed his
opinion that requiring these matters to be addressed early
would actually accelerate the development of useful
regulations. However, we believe it is far more likely that the
highly prescriptive requirements for regulatory analysis in S.
746, enforceable by judicial review, will create even more
grist for controversy during administrative review and will
encourage and empower cautious reviewers to return even well-
analyzed regulatory proposals to the agency for yet more
complete analysis. Our view is shared by a number of public
health experts and consumer representatives who have been
active participants in rule making to assure food safety. For
example, on June 9, 1998, in response to similar comments by
Dr. Crawford regarding S. 981 (a predecessor bill in the 105th
Congress), the American Public Health Association (APHA) wrote
to each Senator: ``Food safety regulations are a prime example
of regulations that could be substantially delayed under S.
981.'' \11\
---------------------------------------------------------------------------
\11\ Likewise, on May 19, 1999, the APHA, the Consumer Federation
of America, the Consumers Union, and the United Food and Commercial
Workers, among others, sent a letter to all Senators in opposition to
S. 746, stating: ``The bill's mandates would also delay the agencies'
[FDA and USDA] already slow regulatory process.'' Furthermore, in a
June 24, 1998 letter to all Senators, Nancy Donley, President of
S.T.O.P.--Safe Tables Our Priority, wrote: ``S.T.O.P. was and is
actively involved in USDA rulemaking and implementation for meat and
poultry regulations. . . . [U]nder S. 981's prescription, common sense
and practical science would not have been enough. The 2\1/2\ year meat
and poultry HACCP rulemaking process would have been further delayed.''
---------------------------------------------------------------------------
4. Transparency and accountability in OMB review of agency regulations
Vice President Quayle's Council on Competitiveness in the
1980's was accused of bottling up regulations from agencies
such as EPA and of providing an off-the-record mechanism by
which outside interests could impose pressure on agencies
without public scrutiny. This Committee was in the forefront of
opposing these abuses.
President Clinton, upon taking office in 1993, sought to
end these abuses by issuing Executive Order 12866. This
executive order guaranteed that regulations could not be
bottled up by OMB indefinitely, and it established a number of
procedural guarantees to avoid off-the-record pressure on the
agencies.
Some supporters of S. 746 have said that it only codifies
this executive order. However, the bill actually differs from
the executive order in a number of important ways. Perhaps most
significantly, the implementation of the executive order's
requirements is not subject to judicial review, the executive
order does not require risk assessment, and it does not require
peer review.
Furthermore, specifically in the area of regulatory-review
procedures, S. 746 includes some of the guarantees established
in E.O. 12866, but omits or cuts back on several of the most
important ones:
The executive order provides that OMB may only review
a regulation for 90 days, subject to a single 30-day
extension. S. 981 as introduced in the last Congress
contained the same provisions. However, the language
was modified in the version of S. 981 reported by this
Committee and in S. 746 so that OMB may unilaterally
extend the review period indefinitely.
The executive order requires OMB to provide a written
explanation for all regulations that OMB returns to the
agency. (In fact, this requirement has been in place
since it was established in a 1986 memorandum by former
OIRA Administrator Wendy Gramm.) A similar requirement
was included in S. 981 as introduced and reported by
this Committee last Congress--yet it has now been
dropped from S. 746.
Under the executive order, OMB must log and disclose
to the agency all ``substantive'' oral communications,
such as meetings and telephone conversations, relating
to the regulation under review. (Indeed, OMB has been
committed to disclosing to the agency ``all oral
communications'' since the 1986 memorandum by former
OIRA Administrator Gramm.) A requirement to log and
disclose all ``substantive'' oral communications was
included in S. 981 as introduced and reported by this
Committee last Congress. However, in S. 746 the logging
requirement has now been weakened, so that OMB may
decline to log a substantive communication if OMB deems
it to be not ``significant.''
The logging requirement in S. 746 applies only to
communications with OMB's Office of Information and
Regulatory Analysis. If OMB assigned regulatory review
responsibilities to any other office, the requirements
to log correspondence and communications could be
circumvented.
Under the executive order, if OMB decides to declare
that an agency's proposed regulatory action is a major
rule subject to regulatory review, it must do so at an
early stage, when the disruption to the agency's
rulemaking process will be relatively limited.
Specifically, if an agency decides that its proposed
regulatory action is not a major rule, that decision
will stand unless overruled by OMB within 10 days after
receiving the agency's list of planned regulatory
actions. By contrast, S. 746 would allow OMB to declare
that a proposed action is a major rule as late as 30
days after the end of the comment period for the rule--
giving OMB the power to force the agency to go back and
prepare a cost-benefit analysis and risk assessment at
that stage, when disruption to the agency's rulemaking
process will be far greater.
We are concerned that these provisions would allow a future
Administration to depart from the procedures in E.O. 12866 and
to restore the abuses of Vice President Quayle's Council on
Competitiveness. Specifically, an Administration could bottle
up rules in endless review, could return rules to the agency
without explanation, and could use regulatory review as a back-
channel conduit for opponents of regulations to communicate
their views off the record.
Senators Lieberman and Edwards advocated amendments at the
markup to rectify some of these problems in the bill. The
sponsors did express a willingness to consider these concerns
further, but we were disappointed that the amendments were not
accepted.
One final point--we are concerned that the bill and the
Majority Report may skew the cost-benefit analysis by including
costs that are speculative and often inflated. For example, the
Majority Report states that agencies should include both
compliance costs and ``opportunity costs'' of the regulation.
This concept of ``opportunity costs'' could lead to very
speculative cost estimates, including, for example, forecasts
of how a business project not even in existence might have
worked out had the government regulations not been issued.
Projections of opportunity costs could also require time-
consuming and extensive information-gathering and analyses of
financial and business data collected from companies by the
government, which some might view as intrusive. Difficult
confidentiality issues and claims might also arise in the
context of collecting and analyzing information on opportunity
costs, resulting in potentially more litigation. Moreover, the
Majority Report fails to acknowledge many of the uncertainties
associated with estimating compliance costs. For example,
industries often adopt advanced or innovative control measures
that significantly bring down costs, but are not anticipated at
the time of the rulemaking.\12\
---------------------------------------------------------------------------
\12\ In a 1995 study, the Congressional Office of Technology
Assessment reviewed seven major OSHA regulatory programs and found that
in no case had regulated companies spent significantly more than OSHA
had predicted, and in five to seven they had spent less.
* * * * * * *
---------------------------------------------------------------------------
In conclusion, we do not question that the sponsors of S.
746 seek to improve the regulatory process through this bill.
But we disagree that this result has been achieved. We are
concerned that the consequences of this legislation will be to
threaten the ability of our health and safety, environmental,
and consumer protection agencies to act in a timely and
decisive manner, and we therefore oppose this bill.
Joseph Lieberman.
Daniel K. Akaka.
Dick Durbin.
Robert Torricelli.
John Edwards.
IX. Changes in Existing Law
In compliance with paragraph 12 of rule XXVI of the
Standing Rules of the Senate, changes in existing law made by
S. 746 as reported are shown as follows (existing law proposed
to be omitted is enclosed in brackets, new matter is printed in
italic, and existing law in which no change is proposed is
shown in roman):
UNITED STATES CODE
TITLE 5--GOVERNMENT ORGANIZATION AND EMPLOYEES
PART I--THE AGENCIES GENERALLY
Chapter Sec.
1. Organization 101
* * * * *
7. Judicial Review 701
8. Congressional Review of Agency Rulemaking 801
9. Executive Reorganization 901
CHAPTER 6--THE ANALYSIS OF REGULATORY FUNCTIONS
[Sec.
[601. Definitions.
[602. Regulatory agenda.
[603. Initial regulatory flexibility analysis.
[604. Final regulatory flexibility analysis.
[605. Avoidance of duplicative or unnecessary analyses.
[606. Effect on other law.
[607. Preparation of analyses.
[608. Procedure for waiver or delay of completion.
[609. Procedures for gathering comments.
[610. Periodic review of rules.
[611. Judicial review.
[612. Reports and intervention rights.]
CHAPTER 6_THE ANALYSIS OF REGULATORY FUNCTIONS
Subchapter I_Analysis of Regulatory Flexibility
Sec.
601. Definitions.
602. Regulatory agenda.
603. Initial regulatory flexibility analysis.
604. Final regulatory flexibility analysis.
605. Avoidance of duplicative or unnecessary analyses.
606. Effect on other law.
607. Preparation of analysis.
608. Procedure for waiver or delay of completion.
609. Procedures for gathering comments.
610. Periodic review of rules.
611. Judicial review.
612. Reports and intervention rights.
Subchapter II--Regulatory Analysis
621. Definitions.
622. Applicability and effect.
623. Regulatory analysis.
624. Principles for risk assessments.
625. Peer review.
626. Deadlines for rule making.
627. Judicial review.
628. Guidelines, interagency coordination, and research.
629. Risk based priorities study.
Subchapter III--Executive Oversight
631. Definitions.
632. Presidential regulatory review.
633. Public disclosure of information.
634. Judicial review.
Subchapter I--Analysis of Regulatory Flexibility
Sec. 601. Definitions
For purposes of this chapter--
* * * * * * *
Subchapter II--Regulatory Analysis.
Sec. 621. Definitions
For purposes of this subchapter the definitions under
section 551 shall apply and--
(1) the term ``Administrator'' means the
Administrator of the Office of Information and
Regulatory Affairs of the Office of Management and
Budget;
(2) the term ``benefit'' means the reasonably
identifiable significant favorable effects,
quantifiable and nonquantifiable, including social,
health, safety, environmental, economic, and
distributional effects, that are expected to result
from implementation of, or compliance with, a rule;
(3) the term ``cost'' means the reasonably
identifiable significant adverse effects, quantifiable
and nonquantifiable, including social, health, safety,
environmental, economic, and distributional effects,
that are expected to result from implementation of, or
compliance with, a rule;
(4) the term ``cost-benefit analysis'' means an
evaluation of the costs and benefits of a rule,
quantified to the extent feasible and appropriate and
otherwise qualitatively described, that is prepared in
accordance with the requirements of this subchapter at
the level of detail appropriate and practicable for
reasoned decisionmaking on the matter involved, taking
into consideration uncertainties, the significance and
complexity of the decision, and the need to adequately
inform the public;
(5) the term ``Director'' means the Director of the
Office of Management and Budget, acting through the
Administrator of the Office of Information and
Regulatory Affairs;
(6) the term ``flexible regulatory options'' means
regulatory options that permit flexibility to regulated
persons in achieving the objective of the statute as
addressed by the rule making, including regulatory
options that use market-based mechanisms, outcome
oriented performance-based standards, or other options
that promote flexibility;
(7) the term ``major rule'' means a rule that--
(A) the agency proposing the rule or the
Director reasonably determines is likely to
have an annual effect on the economy of
$100,000,000 or more in reasonably quantifiable
costs; or
(B) is otherwise designated a major rule by
the Director on the ground that the rule is
likely to adversely affect, in a material way,
the economy, a sector of the economy, including
small business, productivity, competition,
jobs, the environment, public health or safety,
or State, local or tribal governments, or
communities;
(8) the term ``reasonable alternative'' means a
reasonable regulatory option that would achieve the
objective of the statute as addressed by the rule
making and that the agency has authority to adopt under
the statute granting rule making authority, including
flexible regulatory options;
(9) the term ``risk assessment'' means the
systematic, objective process of organizing hazard and
exposure information, based on a careful analysis of
the weight of the scientific evidence, to estimate the
potential for specific harm to an exposed population,
subpopulation, or natural resource including, to the
extent feasible, a characterization of the distribution
of risk as well as an analysis of uncertainties,
variabilities, conflicting information, and inferences
and assumptions;
(10) the term ``rule'' has the same meaning as in
section 551(4), and shall not include--
(A) a rule exempt from notice and public
comment procedure under section 553;
(B) a rule that involves the internal revenue
laws of the United States, or the assessment or
collection of taxes, duties, or other debts,
revenue, or receipts;
(C) a rule of particular applicability that
approves or prescribes for the future rates,
wages, prices, services, corporate or financial
structures, reorganizations, mergers,
acquisitions, accounting practices, or
disclosures bearing on any of the foregoing;
(D) a rule relating to monetary policy
proposed or promulgated by the Board of
Governors of the Federal Reserve System or by
the Federal Open Market Committee;
(E) a rule relating to the operations,
safety, or soundness of federally insured
depository institutions or any affiliate of
such an institution (as defined in section 2(k)
of the Bank Holding Company Act of 1956 (12
U.S.C. 1841(k)); credit unions; the Federal
Home Loan Banks; government-sponsored housing
enterprises; a Farm Credit System Institution;
foreign banks, and their branches, agencies,
commercial lending companies or representative
offices that operate in the United States and
any affiliate of such foreign banks (as those
terms are defined in the International Banking
Act of 1978 (12 U.S.C. 3101)); or a rule
relating to the payments system or the
protection of deposit insurance funds or Farm
Credit Insurance Fund;
(F) a rule relating to the integrity of the
securities or commodities futures markets or to
the protection of investors in those markets;
(G) a rule issued by the Federal Election
Commission or a rule issued by the Federal
Communications Commission under sections
312(a)(7) and 315 of the Communications Act of
1934 (47 U.S.C. 312(a)(7) and 315);
(H) a rule required to be promulgated at
least annually pursuant to statute;
(I) a rule or agency action relating to the
public debt or fiscal policy of the United
States; or
(J) a rule or agency action that authorizes
or bars the introduction into or removal from
commerce, or recognizes or cancels recognition
of the marketable status, of a product under
the Federal Food, Drug and Cosmetic Act (21
U.S.C. 301 et seq.); and
(11) the term ``substitution risk''--
(A) means a reasonably identifiable
significant increased risk to health, safety,
or the environment expected to result from a
regulatory option; and
(B) shall not include risks attributable to
the effect of an option on the income of
individuals.
Sec. 622. Applicability and effect
(a) Except as provided in section 623(f), this subchapter
shall apply to all proposed and final major rules.
(b) Nothing in this subchapter shall be construed to alter
or modify--
(1) the substantive standards applicable to a rule
making under other statutes;
(2) (A) the range of regulatory options that an
agency has the authority to adopt under the statute
authorizing the agency to promulgate the rule; or
(B) the deference otherwise accorded to the agency in
construing such statute; or
(3) any opportunity for judicial review made
applicable under other statutes.
Sec. 623. Regulatory analysis
(a)(1) Before publishing a notice of a proposed rule making
for any rule, each agency shall determine whether the rule is
or is not a major rule covered by this subchapter.
(2) The Director may designate any rule to be a major rule
under section 621(7)(B), if the Director--
(A) makes such designation no later than 30 days
after the close of the comment period for the rule; and
(B) publishes such designation in the Federal
Register, together with a succinct statement of the
basis for the designation, within 30 days after such
designation.
(b)(1)(A) When an agency publishes a notice of proposed
rule making for a major rule, the agency shall prepare and
place in the rule making file an initial regulatory analysis,
and shall include a summary of such analysis consistent with
subsection (e) in the notice of proposed rule making.
(B)(i) When the Director has published a designation that a
rule is a major rule after the publication of the notice of
proposed rule making for the rule, the agency shall promptly
prepare and place in the rule making file an initial regulatory
analysis for the rule and shall publish in the Federal Register
a summary of such analysis consistent with subsection (e).
(ii) Following the issuance of an initial regulatory
analysis under clause (i), the agency shall give interested
persons an opportunity to comment under section 553 in the same
manner as if the initial regulatory analysis had been issued
with the notice of proposed rule making.
(2) Each initial regulatory analysis shall contain--
(A) a cost-benefit analysis of the proposed rule that
shall contain--
(i) an analysis of the benefits of the
proposed rule, including any benefits that
cannot be quantified, and an explanation of how
the agency anticipates that such benefits will
be achieved by the proposed rule, including a
description of the persons or classes of
persons likely to receive such benefits;
(ii) an analysis of the costs of the proposed
rule, including any costs that cannot be
quantified, and an explanation of how the
agency anticipates that such costs will result
from the proposed rule, including a description
of the persons or classes of persons likely to
bear such costs;
(iii) an evaluation of the relationship of
the benefits of the proposed rule to its costs,
including the determinations required under
subsection (d), taking into account the results
of any risk assessment;
(iv) an evaluation of the benefits and costs
of a reasonable number of reasonable
alternatives reflecting the range of regulatory
options that would achieve the objective of the
statute as addressed by the rule making,
including, where feasible, alternatives that--
(I) require no government action or
utilize voluntary programs;
(II) provide flexibility for small
entities under subchapter I and for
State, local, or tribal government
agencies delegated to administer a
Federal program;
(III) employ flexible regulatory
options; and
(IV) assure protection of sensitive
subpopulations, or populations exposed
to multiple and cumulative risks; and
(V) a description of the scientific
or economic evaluations or information
upon which the agency substantially
relied in the cost-benefit analysis and
risk assessment required under this
subchapter, and an explanation of how
the agency reached the determinations
under subsection (d);
(B) if required, the risk assessment in accordance
with section 624; and
(C) when scientific information on substitution risks
to health, safety, or the environment is reasonably
available to the agency, an identification and
evaluation of such risks.
(c)(1) When the agency publishes a final major rule, the
agency shall prepare and place in the rule making file a final
regulatory analysis.
(2) Each final regulatory analysis shall address each of
the requirements for the initial regulatory analysis under
subsection (b)(2), revised to reflect--
(A) any material changes made to the proposed rule by
the agency after publication of the notice of proposed
rule making;
(B) any material changes made to the cost-benefit
analysis or risk assessment; and
(C) agency consideration of significant comments
received regarding the proposed rule and the initial
regulatory analysis, including regulatory review
communications under subchapter IV.
(d)(1)(A)The agency shall include in the statement of basis
and purpose for a proposed or final major rule a reasonable
determination, based upon the rule making record considered as
a whole--
(i) whether the rule is likely to provide benefits
that justify the costs of the rule;
(ii) whether the rule is likely to substantially
achieve the rule making objective in a more cost-
effective manner, or with greater net benefits, than
the other reasonable alternatives considered by the
agency; and
(iii) whether the rule adopts a flexible regulatory
option.
(B) Consistent with section 621 (2) and (3), net benefits
analysis shall not be construed to be limited to quantifiable
effects.
(2) If the agency head determines that the rule is not
likely to provide benefits that justify the costs of the rule
or is not likely to substantially achieve the rule making
objective in a more cost-effective manner, or with greater net
benefits, than the other reasonable alternatives considered by
the agency, the agency head shall--
(A) explain the reasons for selecting the rule
notwithstanding such determination, including
identifying any statutory provision that required the
agency to select such rule;
(B) describe any reasonable alternative considered by
the agency that would be likely to provide benefits
that justify the costs of the rule and be likely to
substantially achieve the rule making objective in a
more cost-effective manner, or with greater net
benefits, than the alternative selected by the agency;
and
(C) describe any flexible regulatory option
considered by the agency and explain why that option
was not adopted by the agency if that option was not
adopted.
(e) Each agency shall include an executive summary of the
regulatory analysis, including any risk assessment, in the
regulatory analysis and in the statement of basis and purpose
for the proposed and final major rule. Such executive summary
shall include a succinct presentation of--
(1) the benefits and costs expected to result from
the rule and any determinations required under
subsection (d);
(2) if applicable, the risk addressed by the rule and
the results of any risk assessment;
(3) the benefits and costs of reasonable alternatives
considered by the agency; and
(4) the key assumptions and scientific or economic
information upon which the agency relied.
(f)(1) A major rule may be adopted without prior compliance
with this subchapter if--
(A) the agency for good cause finds that conducting
the regulatory analysis under this subchapter before
the rule becomes effective is impracticable or contrary
to an important public interest; and
(B) the agency publishes the rule in the Federal
Register with such finding and a succinct explanation
of the reasons for the finding.
(2) If a major rule is adopted under paragraph (1), the
agency shall comply with this subchapter as promptly as
possible unless the Director determines that compliance would
beclearly unreasonable.
(g) Each agency shall develop an effective process to
permit elected officers of State, local, and tribal governments
(or their designated employees with authority to act on their
behalf) to provide meaningful and timely input in the
development of regulatory proposals that contain significant
Federal intergovernmental mandates. The process developed under
this subsection shall be consistent with section 204 of the
Unfunded Mandates Reform Act of 1995 (2 U.S.C. 1534).
(h) Not later than February 5, 2002, the Director of the
Office of Management and Budget shall prepare and submit to
Congress an accounting statement and report containing an
estimate of the total annual incremental administrative
benefits and incremental costs of complying with the provisions
of this subchapter for each agency.
Sec. 624. Principles for risk assessments
(a)(1)(A) Subject to paragraph (2), each agency shall
design and conduct risk assessments in accordance with this
subchapter for--
(i) each proposed and final major rule the primary
purpose of which is to address health, safety, or
environmental risk; or
(ii) any risk assessment that is not the basis of a
rule making that the Director--
(I) reasonably anticipates is likely to have
an annual effect on the economy of $100,000,000
or more in reasonably quantifiable costs; and
(II) determines shall be subject to the
requirements of this section.
(B)(i) Risk assessments conducted under this subchapter
shall be conducted in a manner that promotes rational and
informed risk management decisions and informed public input
into and understanding of the process of making agency
decisions.
(ii) The scope and level of analysis of such a risk
assessment shall be commensurate with the significance and
complexity of the decision and the need to adequately inform
the public, consistent with any need for expedition, and
designed for the nature of the risk being assessed.
(2) If a risk assessment under this subchapter is otherwise
required by this section, but the agency determines that--
(A) a final rule subject to this subchapter is
substantially similar to the proposed rule with respect
to the risk being addressed;
(B) a risk assessment for the proposed rule has been
carried out in a manner consistent with this
subchapter; and
(C) a new risk assessment for the final rule is not
required in order to respond to comments received
during the period for comment on the proposed rule,
the agency may publish such determination along with the final
rule in lieu of preparing a new risk assessment for the final
rule.
(b) Each agency shall consider in each risk assessment all
relevant, reliable, and reasonably available scientific
information and shall describe the basis for selecting such
scientific information.
(c)(1) When a risk assessment involves a choice of
assumptions, the agency shall, with respect to significant
assumptions--
(A) identify the assumption and its scientific and
policy basis, including the extent to which the
assumption has been validated by, or conflicts with,
empirical data;
(B) explain the basis for any choices among
assumptions and, where applicable, the basis for
combining multiple assumptions; and
(C) describe reasonable alternative assumptions
that--
(i) would have had a significant effect on
the results of the risk assessment; and
(ii) were considered but not selected by the
agency for use in the risk assessment.
(2) Significant assumptions used in a risk assessment shall
incorporate all reasonably available, relevant and reliable
scientific information.
(d) The agency shall inform the public when the agency is
conducting a risk assessment subject to this section and, to
the extent practicable, shall solicit relevant and reliable
data from the public. The agency shall consider such data in
conducting the risk assessment.
(e) Each risk assessment under this subchapter shall
include, as appropriate, each of the following:
(1) A description of the hazard of concern.
(2) A description of the populations or natural
resources that are the subject of the risk assessment.
(3) An explanation of the exposure scenarios used in
the risk assessment, including an estimate of the
corresponding population or natural resource at risk
and the likelihood of such exposure scenarios.
(4) A description of the nature and severity of the
harm that could reasonably occur as a result of
exposure to the hazard.
(5) A description of the major uncertainties in each
component of the risk assessment and their influence on
the results of the assessment.
(f) To the extent scientifically appropriate, each agency
shall--
(1) express the estimate of risk as 1 or more
reasonable ranges and, if feasible, probability
distributions that reflects variabilities,
uncertainties, and lack of data in the analysis;
(2) provide the ranges and distributions of risks,
including central and high end estimates of the risks,
and their corresponding exposure scenarios for the
potentially exposed population and, as appropriate, for
more highly exposed or sensitive subpopulations; and
(3) describe the qualitative factors influencing the
ranges, distributions, and likelihood of possible
risks.
(g) When scientific information that permits relevant
comparisons of risk is reasonably available, each agency shall
use the information to place the nature and magnitude of a risk
to health, safety, or the environment being analyzed in
relationship to other reasonably comparable risks familiar to
and routinely encountered by the general public. Such
comparisons should consider relevant distinctions among risks,
such as the voluntary or involuntary nature of risks, well
understood or newly discovered risks, and reversible or
irreversible risks.
Sec. 625. Peer review
(a) Each agency shall provide for an independent peer
review in accordance with this section of--
(1) a cost-benefit analysis of a major rule that the
agency or Director reasonably anticipates is likely to
have an annual effect on the economy of $500,000,000 in
reasonably quantifiable costs; and
(2) a risk assessment required by this subchapter.
(b)(1) Peer review required under subsection (a) shall--
(A) be conducted through panels, expert bodies, or
other formal or informal devices that are broadly
representative and involve participants--
(i) with expertise relevant to the sciences,
or analyses involved in the regulatory
decisions; and
(ii) who are independent of the agency;
(B) be governed by agency standards and practices
governing conflicts of interest of nongovernmental
agency advisors;
(C) provide for the timely completion of the peer
review including meeting agency deadlines;
(D) contain a balanced presentation of all
considerations, including minority reports and an
agency response to all significant peer review
comments; and
(E) provide adequate protections for confidential
business information and trade secrets, including
requiring panel members or participants to enter into
confidentiality agreements.
(2) Each agency shall provide a written response to all
significant peer review comments. All peer review comments and
any responses shall be made--
(A) available to the public; and
(B) part of the rule making record for purposes of
judicial review of any final agency action.
(3) If the head of an agency, with the concurrence of the
Director, publishes a determination in the rule making file
that a cost-benefit analysis or risk assessment, or any
component thereof, has been previously subjected to adequate
peer review, no further peer review shall be required under
this section for such analysis, assessment, or component.
(c) For each peer review conducted by an agency under this
section, the agency head shall include in the rule making
record a statement by a Federal officer or employee who is not
an employee of the agency rule making office or program--
(1) whether the peer review participants reflect the
independence and expertise required under subsection
(b)(1)(A); and
(2) whether the agency has adequately responded to
the peer review comments as required under subsection
(b)(2).
(d) The formality of the peer review conducted under this
section shall be commensurate with the significance and
complexity of the subject matter.
(e) The peer review required by this section shall not be
subject to the Federal Advisory Committee Act (5 U.S.C. App.).
(f) A member of an agency advisory board (or comparable
organization) established by statute shall be considered
independent of the agency for purposes of subsection
(b)(1)(A)(ii).
(g) The status of a person as a contractor or grantee of
the agency conducting the peer review shall not, in and of
itself, exclude such person from serving as a peer reviewer for
such agency because of the requirement of subsection
(b)(1)(A)(ii).
(h) Nothing in this section shall require more than one
peer review of a cost-benefit analysis or a risk assessment
during a rule making. A peer review required by this section
shall occur to the extent feasible before the notice of
proposed rule making.
Sec. 626. Deadlines for rule making
(a) All statutory deadlines that require an agency to
propose or promulgate any major rule during the 2-year period
beginning on the effective date of this section shall be
suspended until the earlier of--
(1) the date on which the requirements of this
subchapter are satisfied; or
(2) the date occurring 180 days after the date of the
applicable deadline.
(b) In any proceeding involving a deadline imposed by a
court of the United States that requires an agency to propose
or promulgate any major rule during the 2-year period beginning
on the effective date of this section, the United States shall
request, and the court may grant, an extension of such deadline
until the earlier of--
(1) the date on which the requirements of this
subchapter are satisfied; or
(2) the date occurring 180 days after the date of the
applicable deadline.
(c) In any case in which the failure to promulgate a major
rule by a deadline occurring during the 2-year period beginning
on the effective date of this section would create an
obligation to regulate through individual adjudications, the
deadline shall be suspended until the earlier of--
(1) the date on which the requirements of this
subchapter are satisfied; or
(2) the date occurring 180 days after the date of the
applicable deadline.
Sec. 627. Judicial review
(a) Compliance by an agency with the provisions of this
subchapter shall be subject to judicial review only--
(1) in connection with review of final agency action;
(2) in accordance with this section; and
(3) in accordance with the limitations on timing,
venue, and scope of review imposed by the statute
authorizing judicial review.
(b) Any determination of an agency whether a rule is a
major rule under section 621(7)(A) shall be set aside by a
reviewing court only upon a showing that the determination is
arbitrary or capricious.
(c) Any designation by the Director that a rule is a major
rule under section 621(7), or any failure to make such
designation, shall not be subject to judicial review.
(d) The cost-benefit analysis, cost-benefit determination
under section 623(d), and any risk assessment required under
this subchapter shall not be subject to judicial review
separate from review of the final rule to which such analysis
or assessment applies. The cost-benefit analysis, cost-benefit
determination under section 623(d), and any risk assessment
shall be part of the rule making record and shall be considered
by a court to the extent relevant, only in determining under
the statute granting the rule making authority whether the
final rule is arbitrary, capricious, an abuse of discretion, or
is unsupported by substantial evidence where that standard is
otherwise provided by law.
(e) If an agency fails to perform the cost-benefit
analysis, cost-benefit determination, or risk assessment, or to
provide for peer review, a court may, giving due regard to
prejudicial error, remand or invalidate the rule. The adequacy
of compliance with the specific requirements of this subchapter
shall not otherwise be grounds for remanding or invalidating a
rule under this subchapter. If the court allows the rule to
take effect, the court shall order the agency to promptly
perform such analysis, determination, or assessment or provide
for such peer review.
Sec. 628. Guidelines, interagency coordination, and research
(a)(1) Not later than 270 days after the date of enactment
of this section, the Director, in consultation with the Council
of Economic Advisors, the Director of the Office of Science and
Technology Policy, and relevant agency heads, shall issue
guidelines for cost-benefit analyses, risk assessments, and
peer reviews as required by this subchapter. The Director shall
oversee and periodically revise such guidelines as appropriate.
(2) As soon as practicable and no later than 18 months
after issuance of the guidelines required under paragraph (1),
each agency subject to section 624 shall adopt detailed
guidelines for risk assessments as required by this subchapter.
Such guidelines shall be consistent with the guidelines issued
under paragraph (1). Each agency shall periodically revise such
agency guidelines as appropriate.
(3) The guidelines under this subsection shall be developed
following notice and public comment. The development and
issuance of the guidelines shall not be subject to judicial
review, except in accordance with section 706(1).
(b) To promote the use of cost-benefit analysis and risk
assessment in a consistent manner and to identify agency
research and training needs, the Director, in consultation with
the Council of Economic Advisors and the Director of the Office
of Science and Technology Policy, shall--
(1) oversee periodic evaluations of Federal agency
cost-benefit analysis and risk assessment;
(2) provide advice and recommendations to the
President and Congress to improve agency use of cost-
benefit analysis and risk assessment;
(3) utilize appropriate interagency mechanisms to
improve the consistency and quality of cost-benefit
analysis and risk assessment among Federal agencies;
and
(4) utilize appropriate mechanisms between Federal
and State agencies to improve cooperation in the
development and application of cost-benefit analysis
and risk assessment.
(c)(1) The Director, in consultation with the head of each
agency, the Council of Economic Advisors, and the Director of
the Office of Science and Technology Policy, shall periodically
evaluate and develop a strategy to meet agency needs for
research and training in cost-benefit analysis and risk
assessment, including research on modeling, the development of
generic data, use of assumptions and the identification and
quantification of uncertainty and variability.
(2)(A) Not later than 180 days after the date of enactment
of this section, the Director, in consultation with the
Director of the Office of Science and Technology Policy, shall
enter a contract with an accredited scientific institution to
conduct research to--
(i) develop a common basis to assist risk
communication related to both carcinogens and
noncarcinogens; and
(ii) develop methods to appropriately incorporate
risk assessments into related cost-benefit analyses.
(B) Not later than 2 years after the date of enactment of
this section, the results of the research conducted under this
paragraph shall be submitted to the Director and Congress.
Sec. 629. Risk based priorities study
(a) Not later than 1 year after the date of enactment of
this section, the Director, in consultation with the Director
of the Office of Science and Technology Policy, shall enter
into a contract with an accredited scientific institution to
conduct a study that provides--
(1) a systematic comparison of the extent and
severity of significant risks to human health, safety,
or the environment (hereafter referred to as a
comparative risk analysis);
(2) a study of methodologies for using comparative
risk analysis to compare dissimilar risks to human
health, safety, or the environment, including
development of a common basis to assist comparative
risk analysis related to both carcinogens and
noncarcinogens; and
(3) recommendations on the use of comparative risk
analysis in setting priorities for the reduction of
risks to human health, safety, or the environment.
(b) The Director shall ensure that the study required under
subsection (a) is--
(1) conducted through an open process providing peer
review consistent with section 625 and opportunities
for public comment and participation; and
(2) not later than 3 years after the date of
enactment of this section, completed and submitted to
Congress and the President.
(c) Not later than 4 years after the date of enactment of
this section, each relevant agency shall, as appropriate, use
the results of the study required under subsection (a) to
inform the agency in the preparation of the agency's annual
budget and strategic plan and performance plan under section
306 of this title and sections 1115, 1116, 1117, 1118, and 1119
of title 31.
(d) Not later than 5 years after the date of enactment of
this section, and periodically thereafter, the President shall
submit a report to Congress recommending legislative changes to
assist in setting priorities to more effectively and
efficiently reduce risks to human health, safety, or the
environment.
Subchapter III--Executive Oversight
Sec. 631. Definitions
For purposes of this subchapter--
(1) the definitions under sections 551 and 621 shall
apply; and
(2) the term ``regulatory action'' means any one of
the following:
(A) Advance notice of proposed rule making.
(B) Notice of proposed rule making.
(C) Final rule making, including interim
final rule making.
Sec. 632. Presidential regulatory review
(a) This subchapter shall apply to all proposed and final
major rules and to any other rules designated by the President
for review.
(b) The President shall establish a process for the review
and coordination of Federal agency regulatory actions. Such
process shall be the responsibility of the Director.
(c) For the purpose of carrying out subsection (b), the
Director shall--
(1) develop and oversee uniform regulatory policies
and procedures, including those by which each agency
shall comply with the requirements of this chapter;
(2) develop policies and procedures for the review of
regulatory actions by the Director; and
(3) develop and oversee an annual government-wide
regulatory planning process that shall include review
of planned significant regulatory actions and
publication of--
(A) a summary of and schedule for
promulgation of planned agency major rules;
(B) agency specific schedules for review of
existing rules, including under section 610;
(C) a summary of regulatory review actions
undertaken in the prior year;
(D) a list of major rules promulgated in the
prior year for which an agency could not make
the determinations that the benefits of a rule
justify the costs under section 623(d);
(E) identification of significant agency
noncompliance with this chapter in the prior
year; and
(F) recommendations for improving compliance
with this chapter and increasing the efficiency
and effectiveness of the regulatory process.
(d)(1) The review established under subsection (b) shall be
conducted as expeditiously as practicable and shall be limited
to no more than 90 days.
(2) A review may be extended longer than the 90-day period
referred to under paragraph (1) by the Director or at the
request of the rule making agency to the Director. Notice of
such extension shall be published promptly in the Federal
Register.
Sec. 633. Public disclosure of information
(a) The Director, in carrying out the provisions of section
632, shall establish procedures to provide public and agency
access to information concerning review of regulatory actions
under this subchapter, including--
(1) disclosure to the public on an ongoing basis of
information regarding the status of regulatory actions
undergoing review;
(2) disclosure to the public, not later than
publication of a regulatory action, of ``
(A) all written correspondence relating to
the substance of a regulatory action, including
drafts of all proposals and associated
analyses, between the Administrator or
employees of the Administrator and the
regulatory agency;
(B) all written correspondence relating to
the substance of a regulatory action between
the Administrator or employees of the
Administrator and any person not employed by
the executive branch of the Federal Government;
and
(C) a list identifying the dates, names of
individuals involved, and subject matter
discussed in significant meetings and telephone
conversations relating to the substance of a
regulatory action between the Administrator or
employees of the Administrator and any person
not employed by the executive branch of the
Federal Government; and
(3) disclosure to the regulatory agency, on a timely
basis, of--
(A) all written correspondence relating to
the substance of a regulatory action between
the Administrator or employees of the
Administrator and any person not employed by
the executive branch of the Federal Government;
and
(B) a list identifying the dates, names of
individuals involved, and subject matter
discussed in significant meetings and telephone
conversations, relating to the substance of a
regulatory action between the Administrator or
employees of the Administrator and any person
not employed by the executive branch of the
Federal Government.
(b) Before the publication of any proposed or final rule,
the agency shall include in the rule making record--
(1) a document identifying in a complete, clear, and
simple manner, the substantive changes between the
draft submitted to the Administrator for review and the
rule subsequently published;
(2) a document identifying and describing those
substantive changes in the rule that were made as a
result of the regulatory review and a statement if the
Administrator suggested or recommended no changes; and
(3) all written correspondence relating to the
substance of a regulatory action between the
Administrator and the agency during the review of the
rule, including drafts of all proposals and associated
analyses.
(c) In any meeting relating to the substance of a
regulatory action under review between the Administrator or
employees of the Administrator and any person not employed by
the executive branch of the Federal Government, a
representative of the agency submitting the regulatory action
shall be invited.
Sec. 634. Judicial review
The exercise of the authority granted under this subchapter
by the President, the Director, or the Administrator shall not
be subject to judicial review in any manner.