Environmental Health News

What's Working

  • Garden Mosaics projects promote science education while connecting young and old people as they work together in local gardens.
  • Hope Meadows is a planned inter-generational community containing foster and adoptive parents, children, and senior citizens
  • In August 2002, the Los Angeles Unified School District (LAUSD) Board voted to ban soft drinks from all of the district’s schools

#435 - House Passes Costly Risk Assessment Bill, 29-Mar-1995

On March 3 the U.S. House of Representatives passed the "Risk
Assessment and Cost-Benefit Act of 1995," also known as "Division D" of
HR 9. (HR 9 is the Job Creation and Wage Enhancement Act of 1995;
Division D incorporates the text of a bill originally known as HR
1022).[1] Division D of HR 9 requires all government agencies to
conduct risk assessment and cost-benefit analyses for every major rule
or regulation. If the Senate approves this bill, it will effectively
end government regulation of health, safety and environment. This is
not an exaggeration.

HR 9 applies to any regulation that will cost society more than $25
million, regardless of what the benefits may be. If the combined costs
to all governments (federal, state, local, tribal) AND to the private
sector (including wage earners, consumers, and the general economy)
total $25 million or more, HR 9 kicks in. It is hard to imagine any
meaningful regulation that would not "cost" the economy $25 million,
even if it will "benefit" the economy far more than $25 million.
Banning a chemical, so that industry must substitute another, or
preventing the Forest Service from selling timber from public land at
give-away prices --such rules as these will "cost" the economy $25
million, even as they create benefits far exceeding $25 million. (After
all, any "benefit" can be viewed as a "cost" by someone. If we reduce
the rates of death and disease, undertakers and physicians will suffer
"costs." If we reduce crime, police may get laid off.) Therefore,
nearly all regulations worth passing will be covered by this law.
Furthermore, every Superfund cleanup costing more than $5 million is
explicitly covered by HR 9. Because the average Superfund cleanup has
cost $10 million so far, HR 9 will cover most cleanups.

For every decision covered by HR 9, here is what is required: (These
are quotations from HR 9; items inside square brackets [ ] are our

(1) When discussing human health risks, a significant risk assessment
document shall contain a discussion of both relevant laboratory and
relevant epidemiological data of sufficient quality which finds, or
fails to find, a correlation between health risks and a potential toxin
or activity. [In other words, all scientific literature will have to be
discussed.] Where conflicts among such data appear to exist, or where
animal data is used as a basis to assess human health, the significant
risk assessment document shall, to the extent feasible and appropriate,
include discussion of possible reconciliation of conflicting
information, and as relevant, differences in study designs, comparative
physiology, routes of exposure, bioavailability, pharmacokinetics, and
any other relevant factor, including the sufficiency of basic data for
review. [All subtle differences between all scientific studies will
have to be discussed.] The discussion of possible reconciliation should
indicate whether there is a biological basis to assume a resulting harm
in humans. Animal data shall be reviewed with regard to its relevancy
to humans.

(2) Where a significant risk assessment document involves selection of
any significant assumption, inference, or model, the document shall, to
the extent feasible--(A) present a representative list and explanation
of plausible and alternative assumptions, inferences, or models; (B)
explain the basis for any choices; (C) identify any policy or value
judgments; (D) fully describe any model used in the risk assessment and
make explicit the assumptions incorporated in the model; and (E)
indicate the extent to which any significant model has been validated
by, or conflicts with, empirical data. [Whew!]

Each significant risk characterization document shall meet each of the
following requirements:

(1) Estimates of risk: The risk characterization shall describe the
populations or natural resources which are the subject of the risk
characterization. If a numerical estimate of risk is provided, the
agency shall, to the extent feasible, provide--(A) the best estimate or
estimates for the specific populations or natural resources which are
the subject of the characterization (based on the information available
to the Federal agency); and (B) a statement of the reasonable range of
scientific uncertainties.

In addition to such best estimate or estimates, the risk
characterization document may present plausible upper-bound or
conservative estimates in conjunction with plausible lower bounds [sic]
estimates. Where appropriate, the risk characterization document may
present, in lieu of a single best estimate, multiple best estimates
based on assumptions, inferences, or models which are equally
plausible, given current scientific understanding. To the extent
practical and appropriate, the document shall provide descriptions of
the distribution and probability of risk estimates to reflect
differences in exposure variability or sensitivity in populations and
attendant uncertainties. Sensitive subpopulations or highly exposed
subpopulations include, where relevant and appropriate, children, the
elderly, pregnant women, and disabled persons.

(2) Exposure scenarios: The risk characterization document shall
explain the exposure scenarios used in any risk assessment, and, to the
extent feasible, provide a statement of the size of the corresponding
population at risk and the likelihood of such exposure scenarios.

(3) Comparisons: The document shall contain a statement that places the
nature and magnitude of risks to human health, safety, or the
environment in context. Such statement shall, to the extent feasible,
provide comparisons with estimates of greater, lesser, and
substantially equivalent risks that are familiar to and routinely
encountered by the general public as well as other risks, and, where
appropriate and meaningful, comparisons of those risks with other
similar risks regulated by the Federal agency resulting from comparable
activities and exposure pathways. Such comparisons should consider
relevant distinctions among risks, such as the voluntary or involuntary
nature of risks and the preventability or nonpreventability of risks.

(4) Substitution risks: Each significant risk assessment or risk
characterization document shall include a statement of any significant
substitution risks to human health, where information on such risks has
been provided to the agency. [End of quotations.]

That's the gist of HR 9. In other words, a full risk assessment will be
required for every significant rule, a full risk assessment like the
one EPA (U.S. Environmental Protection Agency) has been conducting
since 1991 for dioxin. EPA's dioxin risk assessment is in its 4th year,
is still a draft, and still hasn't reached the "policy" stage where the
"policy and value judgments" get debated. The dioxin risk assessment
has been EPA's best attempt to include everyone in the process, and all
available data. And the agency is already being criticized by paid
consultants for including dubious scientific studies and opinions, and
for reaching wrong conclusions. The real fight on dioxin still lies

But HR 9 requires much more than EPA is doing on dioxin now. After the
risk assessment is complete, then HR 9 requires a full cost-benefit
analysis. Furthermore, HR 9 requires that any decision likely to "cost"
the economy more than $100 million must be accompanied by a full peer
review panel of outside experts. (Or, the head of the Office of
Management Budget (OMB) can require peer review, even if "costs" don't
reach $100 million.) The peer review panel "shall not exclude peer
reviewers with substantial and relevant expertise merely because they
represent entities that may have a potential interest in the outcome,"
HR 9 says. In other words, flaks employed by regulated industries will
be welcome.

After the risk analysis is done, and the cost-benefit analysis is
tallied up, and the peer-panel of industry consultants has taken its
bite, then the head of the government agency must "certify" that all of
the analyses are "based on objective and unbiased scientific and
economic evaluations of all significant and relevant information and
risk assessments provided to the agency by interested parties relating
to the costs, risks, and risk reduction and other benefits addressed by
the rule." Further, the agency head must "certify" that "the
incremental risk reduction or other benefits of any strategy chosen
will be likely to justify, and be reasonably related to, the
incremental costs incurred by State, local, and tribal governments, the
federal Government, and other public and private entities." And the
agency head must "certify" that everything that has been "certified" is
"supported by substantial evidence of [sic] the rulemaking record." In
our opinion, this "certification" requirement will guarantees that only
scientific opinions shared throughout industry will be used in risk
assessments. If an issue is the least bit controversial or undecided
within the scientific community, no agency head is going to "certify"
it. Erring on the side of caution to protect public health will become
even rarer than it is today.

After all this, anyone who doesn't like the rule or regulation can
demand "judicial review" --in other words, take the agency to court,
saying that the HR 9 process has not been followed precisely or that
the record of evidence doesn't justify the final regulation.

Under HR 9, we can't imagine any regulations passing in less than a
decade; some could take even longer. If anyone ever tried to run a
business by the methods required in HR 9, they would soon be bankrupted
by delay and indecision. (No doubt, that's the real point of HR 9.)

The irony is that HR 9 says everything about how government must behave
and nothing about how the poisoners must behave. Polluters don't even
have to formally consider their options for reducing environmental
destruction --an alternative decision-making technique that would
improve things for everyone.

In truth, passage of HR 9 should come as no surprise. For 15 years,
agencies like EPA, and some mainstream environmentalists, have been
touting risk assessment as a valuable tool for decision-making. HR 9 is
the logical outcome. Environmentalists are now horrified by HR 9, but
they brought it on themselves by not promoting alternatives to risk
assessment from the start.

In our view, HR 9 should be called the Risk Assessors' Job Creation and
Wage Enhancement Act of 1995. It will guarantee employment for an army
of cynical consultants, soul-less lawyers, and "realistic" mainstream
environmentalists for whom risk assessment is "the only game in town."
But it will do nothing to reverse the accelerating deterioration of the

The good news is that HR 9 may finally drive home the truth about risk
assessment: that it focuses attention on the tiny details of just
exactly how we are going to permit the destruction of our health and
environment, meanwhile leaving the poisoners and murders free to strike
at will, without even requiring them to explain the choices they make.

In their personal lives, many people try to figure out how they can
leave the world a little better than they found it. Shouldn't the
nation spend its resources searching for least-damaging alternatives
rather than searching for the holy grail of 'acceptable risk'?

--Peter Montague


[1] The text of HR 9 is available on the internet via world wide web
(www) or via anonymous ftp. Via www, go to: http://thomas.loc.gov. Via
ftp, go to: ftp.loc.gov/pub/c[10]4 and get the file h[9].eh.FT.

Descriptor terms: legislation; congress; hr 9; hr 1022; contract with
america; house of representatives; risk assessment and cost-benefit act
of 1995; job creation and wage enhancement act of 1995; risk
assessment; dioxin; epa; omb;