Industrial Union Department, AFL-CIO v. American Petroleum Institute

PETITIONER:Industrial Union Department, AFL-CIO
RESPONDENT:American Petroleum Institute
LOCATION:U.S. Department of Labor

DOCKET NO.: 78-911
DECIDED BY: Burger Court (1975-1981)
LOWER COURT: United States Court of Appeals for the Fifth Circuit

CITATION: 448 US 607 (1980)
ARGUED: Oct 10, 1979
DECIDED: Jul 02, 1980

ADVOCATES:
Charles F. Lettow – Argued the cause for the respondents, Rubber Manufacturers Association, Inc., in both cases
Edward W. Warren – Argued the cause for the respondent, American Petroleum Institute, in both cases
George H. Cohen – Argued the cause for Industrial Union Dept
William H. Alsup – for petitioner in No. 78-1036
William Alsup – Argued the cause for the petitioner, Marshall, in 78-1036

Facts of the case

Acting under authority of the Occupational Safety and Health Act of 1970, the Secretary of Labor, after having demonstrated a link between exposure to benzene and leukemia, set a standard reducing the airborne concentrations of benzene to which workers could be exposed. The standard reduced the allowable amount from 10 parts per million (ppm) to one ppm. This case was decided together with Marshall v. American Petroleum Institute.

Question

Did the Secretary exceed his authority to set standards?

Warren E. Burger:

We will hear arguments next in Industrial Union against the American Petroleum Institute, and the related case.

Mr. Alsup, you may proceed whenever you’re ready.

William H. Alsup:

Mr. Chief Justice, and may it please the Court.

The issue in this case is whether the Secretary of Labor’s occupational standard for benzene satisfies the statutory requirements of the Occupational Safety and Health Act of 1970.

The United States Court of Appeals for the Fifth Circuit held that it did not satisfy those requirements, and the case is here on the petition of the Secretary of Labor and the Industrial Union Department of the AFL-CIO.

William H. Rehnquist:

Could you describe those statutory requirements?

William H. Alsup:

The specific statutory requirements at issue are several.

The Fifth Circuit held principally that the statutory requirement that was violated was a requirement that an occupational health and safety standard be, quote, “reasonably necessary or appropriate to achieving healthful and safe working places.

“The Court held that the standard which is not reasonably necessary, if it does not bear a reasonable relationship — the benefits of the standard bear no reasonable relationship to the clause of the standard, and that, in addition, it does not satisfy that clause if they are not quantifiable benefits.

William H. Rehnquist:

What’s the government’s position on the meaning of the statutory language?

William H. Alsup:

Well, with respect to that particular clause, Mr. Justice Rehnquist, we believe that that is an enabling clause that appears in many statutes that confer rule-making authority on agencies.

William H. Rehnquist:

What’s the government’s position as to the meaning of the words “if feasible”?

William H. Alsup:

Achievable.

William H. Rehnquist:

In what sense?

William H. Alsup:

Economically achievable, technologically achievable.

William H. Rehnquist:

Did Congress give any more guidance to the Secretary than that?

William H. Alsup:

Well, in the first sentence of Section 6(b)(5), which we believe is the heart of the Act, Congress gave a very definite guidance to the Secretary of Labor.

That sentence says that in setting health standards for toxic substances, the Secretary shall set the standard which most adequately assures, to the extent feasible, based on the best available evidence, that no employee shall suffer material impairment of health or functional capacity, even if he is supposed to the hazard dealt with for his entire working life.

William H. Rehnquist:

Well, what does the word “if feasible” mean in there?

William H. Alsup:

Well, we think that it means achievable, and the legislative history, we believe, supports our interpretation.

William H. Rehnquist:

Achievable in what sense?

William H. Alsup:

Technologically achievable means either that the technology is available to implement the engineering controls necessary to meet permissible exposure levels, or that as — although this is not involved in this case, as the Circuits have held — that the technology, even though it’s not available at this time, it’s reasonable to conclude that it is, as one of the Circuits said, looming on the horizon.

And there it is a technology enforcing statute in that sense.

William H. Rehnquist:

Well, doesn’t that almost take us back to the Schacter case in the Thirties, where Congress simply turns over to the agency the job of making legislation?

William H. Alsup:

Absolutely not.

Here there is a very clear directive that says that if Congress in fact meant technologically feasible, achievable, and economically achievable, it’s a clear directive, a mandatory directive to the Secretary of Labor to set that standard.

This is not a blank check for the Secretary of Labor to do anything he wants.

Congress has set a — weighted its own balance of health benefits versus health costs and has struck that balance and described it in the first sentence of Section 6(b)(5), and has directed the Secretary of Labor to follow that standard in every health standard that’s issued for toxic substances.

William H. Rehnquist:

If feasible.

William H. Alsup:

If feasible.

William H. Rehnquist:

But it doesn’t define the word “feasible”.

William H. Alsup:

Well, the statute does not, but it has used that word in numerous other statutes and beyond that the clear legislative history here shows that they meant by the word “feasible”, “to the greatest extent possible”.

We don’t think the legislative history is ambiguous on this point.

William H. Rehnquist:

But how about the legislation itself?

William H. Alsup:

Well, the legislation itself uses the word “feasible” and we believe that that means achievable.

It does not mean —

Lewis F. Powell, Jr.:

Mr. Alsup, would you discuss your understanding of what “economic feasibility” means?

Is that dollars only?

William H. Alsup:

Well, —

Lewis F. Powell, Jr.:

I so understood your brief.

William H. Alsup:

— the economic feasibility means, in our view that it is bearable by the affected industry.

Lewis F. Powell, Jr.:

By the entire industry or by each component of it?

William H. Alsup:

By the entire industry, not by each employer.

Lewis F. Powell, Jr.:

You’d put some out of business?

William H. Alsup:

Congress contemplated that there would be a substantial economic injury to some companies.

In fact, in Section 22 of the Act, they provided that the Small Business Administration is authorized to make small business loans to companies that, if they can demonstrate that they will be — and this is the wording of the statute — suffer a substantial economic injury as a result of compliance with the Act.

There is no question but that Congress recognized the achievement of the health and safety goals would be expensive.

Byron R. White:

Suppose you put every company in a particular industry out of business?

William H. Alsup:

That case has never arisen and it’s not likely to arise.

Byron R. White:

I know, but under your more or less absolutive view, what would the situation be if that did arise?

William H. Alsup:

Mr. Justice, we do not have an absolutist view in this case.

We believe that standards must be feasible, at a bearable cost.

Byron R. White:

Bearable to whom?

William H. Alsup:

Bearable to the industry.

Byron R. White:

Industry-wide.

William H. Alsup:

Industry-wide.

Byron R. White:

Well, the example I just put started out being industry-wide, but you lost — take the automobile industry, there are only four major companies in it, I think.

Would you let two go down the drain, or how would you draw the line?

William H. Alsup:

I should begin by saying that that question is not presented by this case.

Could I — because it’s important to understand that that question is not presented by this case.

Byron R. White:

I understand it’s not, but your brief leads one to wonder about it.

William H. Alsup:

Well, first let me say that in this case the Secretary found that the costs were well within the financial capability of the affected industry.

Byron R. White:

These were very large corporations and only about 10 or 12 of them, as I understand it.

William H. Alsup:

There are 20 industries all together.

Byron R. White:

Twenty.

William H. Alsup:

With hundreds of companies involved, not just 12 companies hundreds of companies in 20 different industries that are affected by this standard.

At a total cost of engineering controls of $266 million, recurring annual cost of $34 million, and first-year operating costs of $186 to $205 million.

Spread across hundreds of companies in 20 different industries.

Now, you raise a question.

We do say that all the Secretary has to do with respect to economics is to show that the industry can afford it.

The Third Circuit and, I believe, the D. C. Circuit have gone so far as to say that what economic feasibility means is that so long as there is no “massive dislocation” of the industry, the standard is economically feasible.

This Court does not have to decide whether or not it would take the word “feasibility” to that point.

Although the Secretary believes that in a proper case we might make that argument.

In this case this is so clearly within the capability of the affected industry that it is clearly a standard which is economically feasible.

William H. Rehnquist:

Is it within the capability of the affected industry even though particular enterprises or corporations within the industry might fail as a result of it?

William H. Alsup:

That’s correct.

Byron R. White:

And do I understand the government’s position still to be that within any conception of the meaning of feasibility that the Secretary need give no consideration whatever to the benefits to be derived, health benefits?

Your brief so states.

William H. Alsup:

Well, could I explain how the Secretary considers benefits and costs in deciding to issue a standard?

Byron R. White:

As you do that, bear in mind that your brief, on page 48, in black letters, says “Section 655(b)(5) prohibits basing a standard on a cost-benefit test.”

William H. Alsup:

Oh, that is correct.

Byron R. White:

That’s your position?

William H. Alsup:

That is correct that once the Secretary decides to issue a standard, he cannot trade lives against dollars.

He must either set a safe level or a level which is achievable at a bearable cost even though it’s not a —

Byron R. White:

Does that mean no consideration whatever can be given to the benefits?

William H. Alsup:

No, that is not correct.

Let me explain how that works.

First of all, there are thousands of carcinogens in industrial work places.

The Secretary and the National Institute for Occupational Safety and Health are trying to gather as much information as they can on the various extents of risk that each of these types of carcinogens and other toxic substances pose.

In fact, the Secretary has not regulated that many substances up to this point.

William H. Alsup:

But in making the decision which ones to regulate first, the Secretary does consider whether or not a substance is dangerous, it affects a lot of employees and there is a lot to be gained by attacking this carcinogen problem first.

So that’s the first way, in a rough sense, the balancing of benefits and cost come into play.

The second —

John Paul Stevens:

You now say that balancing is contemplated by the Act?

William H. Alsup:

In the priority setting phase, Your Honor.

As I say, the Secretary —

John Paul Stevens:

In determining which industry is to be regulated?

In selecting the industry.

Once you’ve selected the industry to which a standard is to be applied, is balancing then permitted?

William H. Alsup:

That’s not the way — the Secretary just doesn’t cast his gaze across industry and pick on one rather, it comes to his attention that there is a serious problem with particular industrial carcinogens.

Then the Secretary says, which industry is that thing being used in, and let’s take a look at it.

So that’s the way the priorities are set.

John Paul Stevens:

Once that industry is selected — I’m perhaps not understanding you — may the Secretary then consider the benefits in terms of being beneficial to the health of the individuals in the industry, as compared with the cost?

Your brief says not, as I remember it.

The state simply leaves no room for the conclusion that the Secretary must balance costs against benefits in setting an exposure level.

Now, perhaps by balance you could argue about what that means, but my question is whether the Secretary, once having chosen an industry, must consider whether or not the health benefits are rational, reasonable, —

William H. Alsup:

Yes.

John Paul Stevens:

— bear any relationship whatever to the expenditure of money.

William H. Alsup:

Yes.

The Secretary does consider that.

John Paul Stevens:

He does?

William H. Alsup:

But I have to make a very clear distinction here between a decision to issue the standard at all versus once the Secretary decides to issue the standard, where does he have to go?

Now, in the first question, the prior question of after all the evidence is in what does the Secretary do, does he decide that it’s worthwhile to go ahead and issue the standard, the Secretary does consider whether or not the standard required by 6(b)(5) has, in a rough sense, benefits and has, in a rough sense, costs.

The answer is clearly yes, and we have stated in our brief, certainly in our reply brief, that the Secretary does consider costs and benefits and balancing them for the purpose of whether or not to issue the standard at all.

But, Mr. Justice, my point is that once the Secretary decides to issue the standard, having decided that it’s worthwhile, he has to go to the standards required by the first sentence of Section 6(b)(5) and that is, either a safe level or the lowest feasible level that can be achieved.

John Paul Stevens:

To the point where no employee may suffer.

That’s the language of that sentence.

William H. Alsup:

To the point where it is most adequately assured, to the extent feasible, that no employee will suffer material harm.

William H. Rehnquist:

If he were to go to the extent where no employee would suffer, he would simply ban the use of benzene, would he not?

Because there’s no known use in which it’s safe.

William H. Alsup:

Well, if it were feasible to do that, that could be done.

In this case it’s not feasible to ban benzene.

And therefore —

William H. Rehnquist:

How does he determine that it’s not feasible?

William H. Alsup:

Mr. Justice, there was a very long rule-making hearing, a voluminous record, in which industry after industry came in and testified as to the ways in which benzene is used, and it was very clear from the preamble that the Secretary wrote and from the evidence that it would be impossible to say to industry you can’t use benzene at all.

I could add at this point that that’s been true in all of the carcinogen standards issued by the Secretary.

There’s never been any suggestion by the Secretary that he intends to ban a particular substance altogether.

Byron R. White:

Mr. Alsup, may I ask you two questions about the statutory language?

The first sentence, the toxic materials sentence says that “The Secretary shall set the standard which most adequately assures”, and so he’s got to pick the one which is most adequate.

Now, is the word “standard” as used in that sentence the same or something different than the word “standard” as defined in Section 3, subparagraph (8)?

Where the “reasonably necessary or appropriate” language is used in the definition of the word “standard”.

William H. Alsup:

The answer to that is that the word “standard” as used in Section 3(b) certainly includes a standard that’s used in Section 6(b)(5).

Byron R. White:

Does that mean that any standard within the meaning of this statute must be one that’s reasonably necessary or appropriate?

William H. Alsup:

That’s correct.

Byron R. White:

Then that’s a different position from the one you took in your brief.

You seem to argue in your brief that the reasonably necessary or appropriate language did not apply to toxic materials standards.

William H. Alsup:

Well, our position on the phrase “reasonably necessary or appropriate” is that Congress, in statute after statute, has said to agencies that they may issue regulations deemed by that agency to be necessary to carry out the purposes of the Act.

That is essentially what that definitional provision provides.

In case after case, this Court has said that that sort of enabling language “necessary or reasonable” or “reasonably necessary” — the case has never come up yet actually that said “reasonably necessary or appropriate” but “reasonably necessary” has, “appropriate” and “necessary”.

Now, in all of those cases the Court has said that what that means is that any regulation which the Secretary concludes advances the purposes of the Act falls within that language.

It is not a balancing test, it’s a rational basis test.

Byron R. White:

Well, putting to one side for a moment what the language means, I just want to get fixed in my mind your position on whether or not that language modifies the word “standard” as it appears in 6(b)(5) and I understand you to say yes.

But all it requires is that it fulfill the objective of the statute.

John Paul Stevens:

It doesn’t so much modify the word as that’s —

Potter Stewart:

Part of the definition.

Byron R. White:

— included in the definition of the word.

William H. Alsup:

That’s correct the Secretary is only authorized to issue standards that are reasonably necessary and appropriate.

Potter Stewart:

So that if this stand is not reasonably necessary or appropriate, then even if it’s — then it is not permitted by the statute even if it’s feasible to the new standard.

William H. Alsup:

Well, that would be theoretically correct.

Potter Stewart:

But you just differ on the meaning of “reasonably necessary or appropriate”.

William H. Alsup:

I don’t think it’s just a matter that we differ.

We think the Court’s precedent on the words like “reasonably necessary” and “appropriate” are quite clear that this is something that advances the purposes of the Act and therefore meets that definitional question.

Potter Stewart:

All right.

Let me ask you my second question.

You place a great deal of emphasis in your brief on the fact that this is toxic materials standards, and that the feasibility is the only test you have to meet basically in a toxic materials area.

But what if this were a regular work practice rule, what would be the guideline that the agency would follow there?

There’s nothing else in the statute that I can find that provides a different standard for work practice rules than the one it provides for toxic materials.

William H. Alsup:

Well, the first sentence of 6(b)(5) only covers toxic substances and harmful physical agents like radiation.

Potter Stewart:

I understand that.

William H. Alsup:

It does not — correct, it does not cover safe work practices.

Those, however, can be what we call 6(b) standards.

John Paul Stevens:

The real thrust of my question is If the introductory language of “reasonably necessary or appropriate” is not the criterion that guides the administrator, what is it?

For work practice rules or anything else.

Potter Stewart:

Like slippery floors or something like that.

William H. Alsup:

Well, for a slippery floor the Secretary’s only requirement — well, really twofold requirements.

One is the second and third sentences of Section 6(b)(5).

Byron R. White:

But you said they are just procedural.

William H. Alsup:

Those are procedural, that’s correct.

Byron R. White:

But we’re talking about the substance, what is it that avoids this being the Schechter case, as Mr. Justice Rehnquist pointed out?

William H. Alsup:

Well, as long as the standard is something that a court could find to rationally advance the purposes of the Act, and the Secretary is authorized to promulgate that rule.

Now, that’s in connection with safe work practice type regulations.

Potter Stewart:

Any regulation that may produce an appreciable safety benefit is authorized by this Act.

William H. Alsup:

That’s correct.

Potter Stewart:

Is coal mining covered by this Act?

William H. Alsup:

No, there’s a separate law that covers coal mining.

Potter Stewart:

Yes.

But you couldn’t apply that, the sentence you rely primarily on, to the coal industry without shutting it down, could you?

William H. Alsup:

Well, you know, there is somewhat similar language in the coal mining safety Act.

Potter Stewart:

Has it been so construed?

William H. Alsup:

Mr. Justice, we don’t contend that we would shut down industries.

William H. Alsup:

I’d like to root that out right now.

We do not contend that we would take a standard to construe “feasible” to the point where we would shut down an entire industry.

Potter Stewart:

Well, I understand the entire industry, but perhaps the coal industry might be different.

But you have agreed that the Secretary could adopt standards which shut down segments of an industry.

William H. Alsup:

Marginal employers.

Marginal employers who were surviving because they are able to survive because they don’t give their employees the kind of health protection they deserve.

That’s specifically what this statute was designed to prohibit.

Potter Stewart:

It depends on whether health protection ever can be absolute in terms of protecting against all of the hazards of this life, particularly in certain employment.

William H. Alsup:

But, Mr. Justice, we don’t contend that absolute protection can ever be achieved under the Act.

We simply say feasible protection can be achieved under the Act.

Byron R. White:

Could I ask You suggested that there has to be an appreciable health benefit before he should issue a standard is that it?

William H. Alsup:

Well, I don’t want — the word “appreciable” has gotten a bad lead in this case.

Byron R. White:

Well, what did you mean?

William H. Alsup:

What I meant by that was that it —

Byron R. White:

Well, you used the word.

William H. Alsup:

I’ll explain that.

It rationally advances the purposes of the Act.

Byron R. White:

So it doesn’t have to have an appreciable health benefit?

William H. Alsup:

It has to have a benefit.

It has to —

Byron R. White:

Does it have to have a health benefit?

William H. Alsup:

Yes.

Byron R. White:

What’s the evidence in this case that it might be a health benefit or that there was?

William H. Alsup:

Well, several points.

First of all, the Secretary concluded that there is at least general agreement among scientists that there is a dose-response relationship.

That means that less exposure to benzene is better than higher exposure to benzene, and that by reducing the standard from ten parts per million to one part per million there would be a benefit.

Even the Fifth Circuit said that they were convinced there would be some health benefit by going from ten parts per million to one part per million.

In addition, there are many cases and studies in the record which show that even at levels as low as two to 25 parts per million, people have died of benzene exposure to leukemia.

Take a look at Exhibit 154 in this connection, which is a study by Dow Chemical Company which showed that for three employees exposed at levels between two and nine parts per million, there was a statistically significant increase —

Potter Stewart:

Well, what about the level that is permitted in this case?

William H. Alsup:

The level that is permitted in this case is not a safe level.

It is, however, the lowest feasible level that could be achieved.

Potter Stewart:

So you do say that — well, I guess you would say the decision to issue a standard is always reviewable?

William H. Alsup:

We would say that in a case in which there is clear evidence that the — in the rule-making proceeding — that the standard will not promote the purposes of the Act in a broad rational basis sense, that a Court could say the decision to issue that standard was arbitrary and capricious.

Byron R. White:

And that would be based on the fact that there would be so little health benefits the costs weren’t worth it is that it?

William H. Alsup:

Not under the test that the Fifth Circuit used.

Byron R. White:

Well, what about your test?

I thought you said that that sort of consideration should go into the decision whether to issue a standard at all.

William H. Alsup:

I have to agree.

Under a rational basis test there could be a very limited amount of balancing, but not the kind of balancing that calls for quantifiable benefits or the rational, reasonable relationship test that the Fifth Circuit looked at.

Potter Stewart:

The balancing would go to which hazard to attack?

William H. Alsup:

No, not at all.

Potter Stewart:

Well, let’s say you had a regulation attacking some kind of hazard that everybody agreed was very minor compared to 20 other much more threatening hazards, that the Secretary hadn’t attacked at all.

Would that be a legitimate inquiry?

William H. Alsup:

Now, that’s a question of priority settings.

Let’s be sure we understand the hypothetical.

Let’s say that there are several very serious carcinogens that people are asking the Secretary to regulate.

And instead he postpones that momentarily to take on another carcinogen.

Potter Stewart:

Well, he just postpones it, we don’t know if it’s momentarily or not.

William H. Alsup:

And that is a decision on how to allocate the internal resources of the agency and I don’t think it’s reviewable.

Potter Stewart:

Is that reviewable?

William H. Alsup:

No, that decision is not reviewable.

But once the Secretary decides to issue a standard, it may be that that decision itself to issue a standard is reviewable under the Administrative Procedure Act, under the arbitrary and capriciousness standard.

Potter Stewart:

And in determining whether or not it was arbitrary and capricious, do you think there’s any room for inquiry into the comparative need for it?

William H. Alsup:

Not on the comparative need but on the —

Potter Stewart:

Let’s say he issues a standard attacking mice or rate feces in a factory, in factories around the country, and it’s shown that that’s not good and it’s harmful, but compared to all sorts of twenty other different harmful environmental conditions and carcinogens and poisons and what not, that it’s miniscule, it just isn’t that an appropriate subject for judicial inquiry in reviewing the standard, or the decision by the Secretary to attack one problem and not others.

William H. Alsup:

The decision is to attack which problem to attack first is not a reviewable decision.

However, once a decision and standard is issued, pursuant to decision to regulate a particular area, we think that the Court reviewing that could step back and say Was it arbitrary and capricious to issue this standard?

William H. Rehnquist:

But you said there can be no quantifiable balancing.

William H. Alsup:

In that review the court has never required, under the arbitrary and capriciousness test, that any agency come up with quantifiable benefits.

William H. Alsup:

All we’re asking for is the same sort of deference that the courts have traditionally given to agencies and decisions to issue regulations.

Now, we would have to meet — a separate question is whether or not we would meet the first sentence of 6(b)(5) granted.

But on the threshold question of whether to issue the standard at all, at most the arbitrary and capriciousness standard of review would apply.

And that does not call for quantifiable benefits.

William H. Rehnquist:

But don’t you concede that you would have to show that the worker would be better off under this standard than if you had not issued this kind of thing?

William H. Alsup:

Yes, and we do that.

We do that by saying to the court Your Honor, scientists and medical expert after medical expert came in and said benzene is a carcinogen, it kills employees no safe level is known and the prudent thing to do is to take it to the lowest feasible level.

And that’s exactly what happened here.

That is enough to satisfy the rational basis test for arbitrary and capriciousness standard of review.

Warren E. Burger:

Let me ask you one that takes us a little away from this case.

Twenty years ago, approximately, the Surgeon General of the United States said that if spinach, as sold in the supermarkets, was as dangerous as cigarette smoking, and the prevalance of cigarette smoke in the atmosphere, he, the Surgeon General, would take it off the market in 24 hours.

Now, since then, the Department of HEW has pursued that up to the most recently resigned Secretary of HEW.

It is pretty well established that there are carcinogens which come into the atmosphere from cigarette smoking, not only from the smoker but from those exposed to it.

Now, would the Secretary, under this statute, have authority to ban all smoking in a particular work place?

Not has he thought about it, but would he have the authority to do it?

William H. Alsup:

Well, I don’t know whether the Secretary has thought about that particular problem.

But if the Secretary were to conduct rule-making proceedings and if scientists and doctors, physicians were to come in and testify that smoking by other persons in a work place causes a health threat to the other employees working in that work place, and that cigarette smoking should be banned, this statute clearly would give the Secretary the authority to issue a rule which would limit or prohibit smoking altogether.

We think it would be feasible to do that.

Unless it could be shown, for some reason, that that is necessary, and I think that would be very unlikely necessary to continue the business.

Warren E. Burger:

Now, we’ve consumed a good deal of your time, the time allotted to you, we’ll extend your time five minutes and extend your time the same quantity.

If there are any other questions if not, you may —

William H. Alsup:

Well, I would like to take five minutes.

I think it is very important for the Court to focus —

Warren E. Burger:

You will be using your rebuttal time in this case.

William H. Alsup:

Well, that’s fine I won’t have any, then.

I think it is very important for the Court to understand the facts of this case because the facts of this case have somehow been forgotten and it’s been elevated to hyperbole and extreme hypotheticals that have not occurred in this case and are unlikely to occur in the future.

Here is what the Secretary found, and these conclusions are based on substantial evidence.

First, benzene causes cancer, it causes leukemia.

There are hundreds of deaths reported in the record where people are dying from leukemia where the doctor attributed it to benzene.

It causes aplastic anemia, which is fatal it’s a blood disease.

William H. Alsup:

And it causes chromosonal damage.

Respondents do not dispute these findings.

Now, then, the second inquiry came along because industry witnesses or some of them said, Well, maybe there is a safe threshold of exposure.

We’re already at ten parts per million, maybe that’s already safe.

Extensive evidence was taken on that point.

And a lot of negative studies were presented by industry which were found to have serious epidemiological flaws, and, in addition, scientist after scientist testified that no safe level for exposure to benzene could yet be determined on the basis of the best available evidence.

That is also a conclusion the Secretary adopted, and that’s based on substantial evidence.

John Paul Stevens:

There’s no empirical evidence, is there, Mr. Alsup?

William H. Alsup:

Well, there is.

There are — in the Dow Chemical Company study, for example, let me give you —

John Paul Stevens:

Is the Fifth Circuit opinion wrong in that respect?

There’s a sentence that I’m sure you’ve read several times, on page 18a of the Appendix to the Petition “The lack of substantial evidence of discernible benefits is highlighted when one considers that OSHA is unable to point to any empirical evidence documentating a leukemia risk at ten parts per million” —

William H. Alsup:

Well, in that connection, —

John Paul Stevens:

— even though the standard had been in effect since ’71.

William H. Alsup:

I think the Fifth Court is wrong on that.

John Paul Stevens:

Right.

You just disagree.

William H. Alsup:

We disagree with that, and let me direct your attention to Exhibit 154 and the Secretary’s discussion of that Dow Chemical study.

It was a study in which three employees at exposed levels of between two parts per million and nine parts per million over various durations died from leukemia, at a time when only.8 persons should have died from leukemia in that population.

Now, our reply brief says that was not statistically significant.

We were wrong.

It was statistically significant, and the report so states.

Now, I should say the Dow Chemical doctor said, Well, there are flaws in our study, we can’t really be sure that there was a causal nexus.

Nonetheless, there is a concrete example where, at least statistically significant evidence, such as the type that the Fifth Circuit seemed to want, showed that there was a risk at those low levels of exposure.

In addition, the Stallone study, which was actually one of the negative studies that was presented by industry, in which is also showed an excess risk of leukemia, something like 21 versus 20 that should have occurred, that one was not statistically significant, but, nonetheless, there is evidence that there is an excess risk.

Those levels were down around two parts per million.

It’s unclear exactly what the risk — the exposure levels were, but it was in the refining industry, where those companies are already at or almost below the one part per million standard now.

In addition, there was scientific testimony that there is, in general, a dose-response relationship — that means less exposure to carcinogens is better.

And we have to keep in mind that we’re dealing with a statute that requires that we protect employees or the Secretary protects employees, even if they work 40 or 45 years in that industry.

Potter Stewart:

Mr. Alsup, this is — you’re bringing in really a different argument than you made in your reply brief.

Potter Stewart:

You’re now contending that there is evidence in the record pertaining to the area between one and ten parts per million.

But I do not understand that the Secretary relied on any such evidence, but, rather, he assumed that we didn’t know the answer in that area, but there may be a danger, and the “may be” was enough.

William H. Alsup:

Well, that’s not quite —

Potter Stewart:

Isn’t that correct?

William H. Alsup:

Well, that’s almost correct.

What the Secretary said was that no safe level could be identified on the basis of the best available — nothing that we could say “this is a safe level”.

Potter Stewart:

Right.

William H. Alsup:

The Secretary did, however, say, and relied on the Dow Chemical study as evidence that is consistent with the possibility that —

Potter Stewart:

That there would be some risk in this area.

William H. Alsup:

— there is a risk at those low levels.

Potter Stewart:

But I don’t think he really made a finding that the evidence establishes a risk in this area, I think his — at least I certainly didn’t read him that way.

William H. Alsup:

Well, he did not make a finding —

Potter Stewart:

And I didn’t understand your brief to rely on that, and us looking into the background to see if there’s evidence.

William H. Alsup:

That’s correct.

We do not rely upon being able to go out and perform studies which show that at these low levels you are always going to be able to prove the risk.

The important thing about this case, Mr. Justice, is that that can’t be done for carcinogens, for most carcinogens, for some it can.

For benzene, the data is just not there.

And if this Court adopts a rule that says you have to show quantifiable benefits at these low levels, then the Secretary is going to be disabled from acting to protect workers against carcinogens in a whole host of areas.

Because —

Potter Stewart:

Well, there are other things he could do besides banning it or cutting it down, for example you could make a big point — or I guess that’s your colleague, making a big point about whether workers exposed for a full lifetime or just periodically would make a big difference.

Maybe you could just say they could only work for a year in these exposed areas, or something like that.

That would be something to do.

William H. Alsup:

No, that would be less feasible.

That might be less feasible in these industries where —

Potter Stewart:

Less feasible to give temporary assignments than to spend a half billion dollars?

William H. Alsup:

To tell an employee that he can only work in this industry for one year, I think that’s not acceptable.

Potter Stewart:

Well, in the particular exposed areas.

William H. Alsup:

Well, many employees are exposed.

They can’t —

Potter Stewart:

Thirty thousand.

William H. Alsup:

I admit, in some cases it’s possible to rotate employees.

That is not a satisfactory answer for all the employees exposed to benzene.

It’s a daily occurrence in their work.

Warren E. Burger:

Very well, Mr. Alsup.

Mr. Cohen.

George H. Cohen:

Mr. Chief Justice, and may it please the Court.

What I plan to do with the time allocated to me is to highlight to the Court our belief that Section 6(b)(5) of the Act is in fact the focus and contains the charters and indeed the command to the Secretary of Labor as to how he should set and how he must set standards governing occupational exposure to toxic substances.

However, in order to frame the argument, I think that it’s important to keep in mind that we bring to this Court a Secretary of Labor’s determination, based on findings of fact, supported by substantial evidence which was not upset by the Fifth Circuit.

That not only determined that exposure to benzene causes leukemia or cancer, that not only that hundreds of workers had already died of cancer, and that indeed 30,000 or more workers will be exposed to this hazard every single day of their working life and that, based on experts scientific opinion testimony, the determination was made that there is no level, there is no level that can be considered safe for employees who are in fact exposed to benzene.

Now, the Secretary of Labor didn’t rest on that, the Secretary endeavored to ascertain what that would mean in terms of the risk, the magnitude of the risk on employees in this work place.

The Secretary just didn’t say, “Let’s forget about it”, the Secretary endeavored to do that and took an enormous amount of testimony, and ended up concluding that there simply was no way to do it.

There was no way to make an educated estimate or to quantify the benefits.

William H. Rehnquist:

Why didn’t he then simply ban the use of benzene?

George H. Cohen:

Well, Mr. Justice Rehnquist, there is a statutory procedure here.

The guiding light of 6(b)(5) says you do what will most adequately protect.

But there is one constraint, there is one restraint, this is not an absolute statute, you must do what is feasible.

And the Secretary went on to find that one part per million was in fact feasible, and it’s our understanding that finding has not been challenged, was not upset by the Court of Appeals, —

William H. Rehnquist:

How does the Secretary know what is feasible under this statutory language?

George H. Cohen:

Well, I know you were exploring that with counsel for the Solicitor.

The answer to that, of course, is not a simple one, but there is an evidentiary record made.

Engineering controls, work practices — in other words, a whole series of items have evolved during the proceedings, in which companies come in, engineering experts come in, plant superintendents come in, and they say, Now, what can we do to eliminate?

This is not a mystical practices.

We’re talking here about gust and emission from the source of a machine going into the air.

And testimony comes in, how can we eliminate it?

And to what extent can we eliminate it?

And, for example, in the Appendix, at page 141 in this case, there’s discussion about the need to replace burners, the need to replace pumps.

There are systems, as the product goes through the plant where these emissions are detectable, and where possible, feasible, available kinds of things can be done, and customarily what’s grown up, Mr. Justice Rehnquist, is a series of things that are done.

You look to see hood ventilating machinery equipment, you look at what kind of work practices can be brought to bear you look at a whole slew of possible ways to deal with the hazard, and that ends up in a determination on the record by the Secretary as to what level can be achieved.

For example, in many of these standards, employers have — or certain employers in an industry have reached a certain level and they bring people who testify, and under direct and cross-examination it is explained and determined on what basis they were able to reach that level and those kinds of findings in turn become the predicate for the Secretary’s technological feasibility determination.

And I want to emphasize, Your Honor, as we understand it, that question is not before this Court.

George H. Cohen:

Industry does not challenge its capacity, through engineering controls and work practices, to reach the one ppm level.

William H. Rehnquist:

Well, when you say “industry” I understand you to mean an entire industry in the sense of the coke processors, or the oil industry.

I also understand that the Secretary considers the statute to authorize him to impose controls which would make it impossible for a particular company within that business to continue to do business, using those controls.

George H. Cohen:

Well, I think we have to take a look back.

I think we have to see what was going on in Congress’s mind when they passed this statute.

And what was going on in Congress’s mind was this parade of experiences had demonstrated that there were all these known toxic substances cut there, that indeed with the new technology and processes coming to the fore, it was expected that every twenty seconds another substance, a potentially toxic substance was going to be introduced into the working environment.

And, as you know, Your Honor, there is no protesting requirement under this statute or any other.

So the Congress of the United States knew and understood this, and they knew another thing they knew that the state of the scientific knowledge at this point in time was such that since most of these particular diseases had what we refer to as a long latency period — in other words, 20 or more years from the time that you initially begin to inhale these particular substances until the disease actually manifests itself.

That they knew that there was not going to be immediate, imminent scientific certainty in this area.

And in the face of that, and because the Congress decided that the overriding concern here had to be to maximize the protection to employee work, even in the face of the scientific uncertainty, they, in 6(b)(5), started out, not only with the mandates of the Secretary — and I want to discuss that mandate — but with the use of a very carefully chosen phrase, we believe, “best available evidence.”

In other words, what this Congress contemplated that the Secretary, on the basis of a record submitted before him, would take the best available evidence, everything that the Secretary knew for sure, and also having to deal with the reality that things would not be known.

And, for example, in response to Mr. Justice Stevens’ question, one of the things that was not known here was what would happen to employees who were currently exposed at 10 ppm.

And I believe the Fifth Circuit Court of Appeals opinion at Pet-Ap 18a was a correct statement, but it misses the mark.

It was correct in this sense there is no direct empirical evidence as to what would happen to people at 10 ppm, because the vast majority of people today have been exposed in excess of that number, and there have been no studies that have demonstrated what would happen to you.

There was no direct empirical evidence.

Worse than that, there is no reason to believe there will be such direct empirical evidence for many years to come, precisely because of the problem with the latency period.

In 1971 the statute was passed, but for anyone to make an intelligent and rational judgment as to what the impact will be, they have to take into consideration we have a latency period here of approximately 20 years.

So that any study, any study that would begin to say there is no problem at 10 ppm unless there was a latency period of a minimum of 20 years, certainly could not be a valid basis for going ahead and regulating.

But —

William H. Rehnquist:

Mr. Cohen, all of what you say makes sense, except for the words “if feasible”.

George H. Cohen:

All right.

William H. Rehnquist:

Which sound to me as if Congress is simply completely passing the buck to the Secretary.

George H. Cohen:

Completely the opposite, Your Honor.

Let me spell that out.

We have to start out first of all —

Potter Stewart:

If I may, just to be sure I understood your last one, and I don’t want to interrupt your answer to my brother Rehnquist but it’s your submission, therefore, that even if the answer to this question, what happens to employees exposed to 10 ppm of benzene, and even if the answer to that question turns out to be nothing happens to them at all, nothing nonetheless, this is a perfectly and completely valid regulation.

George H. Cohen:

All right.

Now, you’ve asked me if that’s part of the balance and I’ll respond just as candidly as I can.

Yes.

But is there reason to think that, I’d have to say no.

George H. Cohen:

We have 30,000 —

Potter Stewart:

But your point is we don’t know, and it’s even.

George H. Cohen:

That’s right, we don’t know.

We do know this much —

Potter Stewart:

And so if we don’t know, it’s possible — if we don’t know, then it’s certainly possible that the answer is nothing.

George H. Cohen:

That’s correct, Mr. Justice Stewart.

But 30,000 workers are exposed, we know that, and we know that, notwithstanding industry’s effort to identify 10 ppm as a safe level, the Secretary, after carefully reviewing and analyzing all of those studies, concluded that those studies simply do not stand for that proposition, and indeed, Mr. Justice, indeed that for the Secretary to have acted on the basis of those studies would have been a reckless and irresponsible act, because the studies didn’t stand for the answer that industry was submitting them for.

So what we have said in our brief, and we think this is the posture that this case comes to this Court, that we are in a situation where there is a potentially great adverse consequence, albeit of unknown dimensions.

And I think —

Potter Stewart:

Potentially zero consequences.

George H. Cohen:

That’s right.

I can’t say it is not zero.

I certainly would suggest when you read the literature and you see what has happened to date, you would not want to come to that conclusion.

Potter Stewart:

Well, if you don’t know, you don’t know.

And then you —

George H. Cohen:

If you don’t know, you use your best available evidence, and that’s what the Congress had in mind.

Because, if you don’t know and if you conclude, in these circumstances, that you are not going to regulate, then, notwithstanding what the Fifth Circuit said as a disclaimer, that this is not a statute in which we have to wait for a body count we submit that is exactly what has got to happen.

And the reason why it’s got to happen here is the end product of what I was developing in my answer to Mr. Justice Rehnquist the direct empirical evidence at 10 does not exist and will not exist for years to come.

Potter Stewart:

Well, then this would be true of any environmental ingredient.

George H. Cohen:

No, it would not.

Potter Stewart:

Why not?

George H. Cohen:

Because, as this record makes clear, —

Potter Stewart:

Because anything used to excess can be harmful, including drinking water.

George H. Cohen:

I don’t mean to suggest that.

What I thought I was responding to was your question as to whether or not with respect to all toxic substances, is it impossible to create a dose-response curve?

The answer to that is no, it is not.

Indeed, Dr. Krable of the National Cancer Institute, on the record in this case, bemoaned the fact that he tried, he tried to come up with a dose-response curve for benzene, and he points to that fact, as I did with respect to vinyl chloride.

I and other scientists.

And then he resorted to the language of the —

Warren E. Burger:

Well, I understand your answer.

Warren E. Burger:

Your time is running out, perhaps you’d better answer my brother Rehnquist’s question.

Oh, you forgot it.

Harry A. Blackmun:

Isn’t this a fairly common situation in which medical research finds itself?

George H. Cohen:

Well, Mr. Justice Blackmun, I certainly think under this statute it is conceivable that this will be something that will happen at least on occasion, but, conversely, — and I want to emphasize this — the experience under several of the other very serious carcinogens that have been regulated to date by the Secretary, there were in fact adequate dose-response curves produced.

That was true, for example, in vinyl chloride, as I just mentioned and that was true to a major extent in the so-called steel industry coke ovens’ emission standard, which is here in a companion case pending before the Court.

Harry A. Blackmun:

Well, I’m trying to support you really, because —

George H. Cohen:

Good!

[Laughter.]

Warren E. Burger:

— medicine has gone through this.

George H. Cohen:

Yes, they have.

Warren E. Burger:

They went through it with cortisone.

George H. Cohen:

Yes.

Potter Stewart:

And now twenty years, twenty-five years after the discovery of what they thought was the answer to many diseases, namely cortisone, they now discover it isn’t.

George H. Cohen:

Let me give you another point in that regard.

When the Secretary of Labor regulated employee exposure to asbestos, he did it on the grounds that it was a serious toxic substance, but it was at that point in time not identified as a carcinogen.

Regrettably, three years later but well after the standard has been in place, it’s been identified as a carcinogen.

So the Secretary of Labor is standing there, Mr. Justice Stewart, in these situations —

Potter Stewart:

Yes, but Justice Rehnquist has an outstanding question.

George H. Cohen:

Oh, I’m sorry.

Warren E. Burger:

If you can remember it?

George H. Cohen:

I wish I could say I remember it.

Warren E. Burger:

Well, then, Mr. Justice Rehnquist can try again.

William H. Rehnquist:

I don’t remember, either.

[Laughter]

George H. Cohen:

I do know that I was trying to articulate to you a legislative history of 6(b)(5) on you.

I want you to understand that 6(b)(5) just didn’t throw those words “feasible” at us without a very carefully conceived formulation.

Senator Javits, the ranking minority member, introduced an amendment to add after the words “most adequately” the words “and feasibly”.

And when he did so he made it clear that he was concerned that there might otherwise be interpretations that we have an absolute safety and health standard.

They then reported the bill out and a major debate took place on the Senate Floor, and that debate focused on the question of feasibility.

And Senator Dominick, who was the leading spokesman for the minority side of the aisle on this issue, time and again was concerned that the words “most adequately” and “feasibly” sure could literally be read to mean the most adequate way of doing it is shutting down the business, shutting down industry and banning the occupation.

George H. Cohen:

That was on the one side.

He did not want perfection, he wanted possibilities.

John Paul Stevens:

Of course at that time the bill read “any impairment” —

George H. Cohen:

Excuse me, sir?

John Paul Stevens:

At that time the bill read “any impairment of health” as opposed to material impairment.

George H. Cohen:

Yes, but I think we gave that argument the shrift that it is entitled to.

Those words, “Material” —

John Paul Stevens:

I agree, that argument doesn’t answer this case, but I think it may answer the concern in the legislative history at the time.

George H. Cohen:

No, I think the “material impairment” language went directly to the physical condition of human beings —

John Paul Stevens:

Right.

George H. Cohen:

— and that language, although the legislative history is almost silent, is a marvelous little statement by Senator Saxbe in his classic fashion saying, “Well, with just the word ‘impairment’, and we’ve got a provision which says even if that individual was regularly exposed throughout his whole working life, what if a bunch of mosquitoes keep coming back every month and biting this fellow, are we going to be in a position that some bureaucrat is going to say ‘that’s an impairment within the meaning of Section 6(b)(5)’.

“I believe that the word “material” was inserted, although I don’t have any definitive citation for you, to eliminate that kind of a possibility.

Not, I want to emphasize, not because Congress wanted to inject a cost-benefit analysis concept.

This is a legislative history that there was not one single mention, not one reference, with all the discussion and all the focus on what they meant by “technological feasibility” or “feasibility”, never was there an allusion to the concept of a cost-benefit analysis.

William H. Rehnquist:

But in a critical point like this, where we’re talking about the extent feasible, why should we have to look to vague legislative history that is subject to ambiguous construction when Congress could have said either this or that or the other?

George H. Cohen:

My only disagreement, Mr. Justice Rehnquist, would be your description of this is vague.

The focus of the battle that Senator Dominick generated was the need to eliminate impossibilities.

And he kept emphasizing repeatedly, “I don’t want to create a Utopia which is impossible of achievement, that won’t accomplish anything I do want to accomplish what can be achieved through the state of the art”, at 480, 481 and 482 of the legislative history that was the focus.

And it was a very clear focus and it was the one constraint, the one constraint that the Congress was willing to put on what otherwise, we believe, is a singular dedication and devotion in that language of 6(b)(5) to the broadest possible — the maximum possible employee protection.

Warren E. Burger:

Your time is up now.

George H. Cohen:

My time is up, and I thank you, Mr. Chief Justice.

Warren E. Burger:

Thank you.

Mr. Warren.

Edward W. Warren:

Mr. Chief Justice, may it please the Court.

I have been —

Warren E. Burger:

If it would be more convenient, Mr. Warren, you may elevate the lectern.

Edward W. Warren:

How does one do that?

Byron R. White:

The state of the art.

[Laughter]

Warren E. Burger:

It’s remained static for forty-some years now.

Warren E. Burger:

[Laughter]

Edward W. Warren:

Mr. Chief Justice, may it please the Court.

I have been allotted 35 minutes to argue the main issue presented by the decision below my colleague, Mr. Lettow, will argue for ten minutes exclusively on the dermal exposure issue.

But this case is far simpler, I believe, than my opponents have made it appear.

The Fifth Circuit held that the — that OSHA must shoulder the affirmative burden of showing, by substantial evidence, that its standard is reasonably necessary and otherwise complies with the Act.

Here OSHA failed to show that the standard would do any discernible good, and the Court of Appeals reversed.

The decision and the result reached by the Fifth Circuit should be affirmed.

Congress wanted OSHA to issue standards that are reasonably necessary to protect workers.

My opponent, Mr. Alsup, has suggested that that provision is nothing more than a broad enabling statute — broad enabling provision.

But, in fact, there is in this very statute enabling language in another provision which says — this is Section 8(g)(2) which says that OSHA may issue such rules and regulations as he may deem necessary.

I don’t believe, when Congress has so clearly used enabling type language, one can construe a reference, such as in 3(8) as an enabling provision.

Quite the contrary.

The term “reasonably necessary” has consistently and uniformly been construed by the Court, First Circuit, in the Bean case and the Fifth Circuit in the Aqua Slide case, to be fully consistent with what this Court held in the Benzene case.

Now, the purpose that Congress wanted OSHA to follow, namely, to set reasonably necessary standards, is carried forward in Section 6(b)(5).

That provision intends for the Secretary to set feasible standards which, the legislative history shows, were to encompass reasonable and practical steps, to control material health hazards.

Such standards are also to be based not on views evidenced in arguments, and there is a great deal of legislative history, or some legislative history about this but instead on substantial evidence and the best-available evidence.

The Fifth Circuit — and in each of these instances, the determinations on these issues are required by the statute to be supported by substantial evidence.

The Fifth Circuit, in light of these provisions, held essentially two things it held that OSHA must show that its standard will produce discernible benefits and, secondly, that there is some reasonable relationship between the benefits expected and the cost of the standard.

My opponents have both mentioned that there are, as Mr. Alsup said, thousands of carcinogens in the work place, I believe he said, and Mr. Cohen noted the Congress envisioned every twenty seconds a new chemical going into the work place.

Now, it’s our position that the statute expected OSHA to be reasonable under the circumstances, and that if the over-all aims of the statute, which are to protect workers, are going to be achieved, they can only be achieved if there is a judgment made that each individual standard is going to accomplish some discernible good, and is going to benefit the worker, because obviously unless that sort of determination, that sort of reasonableness determination is made with every standard in the first instance, what you’re going to find is, ten years from now, industry will have spent fortunes to control what could turn out to be de minimis, or hazards which really are not terribly significant.

Warren E. Burger:

When you say “discernible” do you mean quantifiable, measurable?

Edward W. Warren:

Well, let’s — I think the key to that question is what does best-available evidence mean?

But let’s first try to answer the question, were there discernible benefits here?

We’ve talked a lot back and forth about the question of whether there is any evidence under a hundred parts per million — and I say a hundred parts per million because there’s really no evidence below a hundred parts per million — as Mr. Justice Stevens noted, the Fifth Circuit said there was no empirical evidence below ten.

My opponents have brought up the Dow study.

I felt that with their reply brief we had seen the end of last-minute ex post facto rationalizations of counsel, but we have yet a new study brought on the scene.

And I just call your attention to page 15 of the Joint Appendix, which is EPA’s interpretation of that study.

It says that the Ott study, that’s the Dow study, indicated no excess risk of leukemia with relatively well-documented exposures.

I cross-examined the witnesses for Dow in the proceeding, and I will tell you that there is evidence that some of the exposures in that instance were up in several hundred parts per million.

They varied all over the lot.

Edward W. Warren:

There were different groups of workers exposed to different amounts.

I don’t believe I’ve really answered your question, though, because, coming back to the question of discernible benefits, we feel that here there are no demonstrated benefits whatsoever.

There is no empirical evidence of any harm under a hundred parts per million.

Now, the question —

Potter Stewart:

There was an industry practice of reducing it to 10 ppm, wasn’t there?

Wasn’t that how it went?

Edward W. Warren:

Well, it’s not an industry practice, it’s a consensus standard, and the consensus standard has the effect of law.

It is law.

It was adopted without rule-making pursuant to section 6(a) of the statute.

Potter Stewart:

The 10 ppm.

Edward W. Warren:

Yes.

And it’s been in effect —

Potter Stewart:

You’re not in any way attacking that, are you?

Edward W. Warren:

No, we’re not attacking that.

In fact, we think that standard is providing fully adequate protection to the workers.

Coming to the question of whether those benefits ought to be quantifiable or not, our position on that issue is that it’s controlled by what the best-available evidence is.

Now, in this instance, you’ll recall that the Environmental Protection Agency did quantify the benefit, was able to make estimates of what the standard would accomplish.

The same was true of Dr. Wilson, EPA’s own contractor in this case, who testified that it could conduct a cost-effectiveness type analysis which would encompass some sort of estimate of this type.

The Fifth Circuit noted that in its decision and was fully aware of the capability in this regard.

So I think in answer to your question, Mr. Chief Justice, the best-available evidence language is controlling with respect to the degree to which quantification can occur but I think the evidence is overwhelming in this case that such quantification can indeed be done, because it was done by EPA’s sister agency, the Environmental Protection Agency.

Now, the important thing, I think, here is that if you carefully analyze what the Fifth Circuit said, it boils down to really three things.

It says, first of all, that the agency must carefully consider and address any evidence on benefits submitted by the parties.

Secondly, it should attempt to estimate, as well as possible, what the standard is intended to accomplish.

And third, it ought to explain and justify the standard which it ultimately promulgates in terms of what it is expected to accomplish.

What is this going to accomplish for the worker?

Now, I submit that no one can seriously contend that that was done in this case.

I don’t think any of those things was done in this case.

It’s probably the best evidence, perhaps, that this wasn’t done in this case is that the Secretary has felt called upon to submit a very lengthy addendum to its reply brief, which sets forth what I think it wishes it had said but didn’t say about benefits in this case.

To me this tests the limits of what the ex post facto rationalization of counsel doctrine must mean.

This takes it awfully, awfully far.

Edward W. Warren:

Now, Mr. Justice Rehnquist asked a question earlier about how feasibility was determined in this case.

I think that bears a short discussion, because here’s what happened.

The Secretary said, We have a health hazard here, and the Secretary said to the contractor, not should we set a standard at one or five or ten, and what’s feasible, or even one-half what he said is, We’re talking about a one part per million standard, can you go out and show us that industry can come up with the hardware to accomplish this, and that it won’t drive the industry out of business, or it won’t require the industry to be shut down?

That’s all that happened in this case.

There was no consideration of alternatives.

The contractor was given no discretion to look at any of the questions which we’re talking about here today.

It is for that very reason, of course, that we presented evidence and discussed this question at length in the hearings.

The kind of evidence we presented — and I might add that the Council on Wage and Price Stability was weighing in there on exactly the same basis — was that this statute is sufficiently flexible to permit the kinds of things that the Fifth Circuit said.

We’re not talking about elaborate cost-benefit analyses, what we’re talking about is reasoned, rational decision-making in a situation where Congress wanted ultimately the most protection for workers to be accomplished.

But that can’t be accomplished by taking a tunnel vision approach as was taken in this case to one hazard, and a hazard, indeed, where the evidence strongly suggests we don’t have nearly as great a problem as we’ve had with others, and will probably have with others in the future.

William H. Rehnquist:

But a form of that protection could certainly be accomplished by simply banning carcinogens.

Edward W. Warren:

I think that of course adequate protection could be accomplished by banning carcinogens, but the problem —

Potter Stewart:

Everybody agrees that wouldn’t be feasible.

Edward W. Warren:

Yes, that wouldn’t be feasible, it wouldn’t — it just is an affront to common sense, and I don’t —

William H. Rehnquist:

Well, Congress, by putting in the language “if feasible” apparently ruled out that.

Edward W. Warren:

Congress definitely ruled that out.

I don’t think that —

William H. Rehnquist:

But what did it put in its place?

Edward W. Warren:

Well, it puts the word “feasible” in.

Now we have to ask what was “feasible” intended to accomplish?

Let me talk about the legislative history for a minute, because I think there are two key points on the legislative history which tell us that Congress wanted that term “feasible” to mean reasonable and practical.

Things that are really very consistent with Section 3(8), the “reasonably necessary” language.

First, Senator Javits was the author of the original Administration bill on occupational health and safety.

It was submitted in 1969.

That bill provided for standards to be set by an independent board and the standards would be set based upon a report regarding the standard’s feasibility, which the statute says in so many words would include reasonableness, practicality and technical feasibility.

Now, Senator Javits was the principal proponent of the feasibility amendment.

He added it as an amendment in the Senate Committee.

It is reasonable — it is unreasonable in fact to assume anything else — that when Senator Javits put that “feasibility” word in there, that is precisely what he meant.

He did not mean this to mean some sort of arbitrary business shutdown criterion.

William H. Rehnquist:

Well, you say that’s precisely what he meant.

William H. Rehnquist:

What is precisely what he meant?

Edward W. Warren:

Well, I think what he meant was that the standard ultimately — it’s the ultimate objective we’re talking about, the standard should be reasonably necessary.

I think that’s consistent with saying that it should be — protect the worker to the extent feasible.

Now, the ultimate objective is that the standard be reasonable, the standard be practical.

What the Fifth Circuit is doing is telling us how that objective can be achieved.

How that goal of reasonably necessary standards, how that goal of reasonable and practical standard can be achieved.

It’s saying, first of all, that the agency must show some discernible benefit and, secondly, that there must be some reasonable relationship between the costs and the benefit.

And it’s giving us a means of doing that.

It seems to me that this whole approach is strongly supported also by the evidentiary requirement, including the substantial evidence test.

Congress put the substantial evidence test in for a very clear and important purpose, I think.

Congress was concerned that the Secretary of Labor would issue arbitrary, unreasonable, unexplained, unsupportable standards.

Some of the original bills had the arbitrary and capricious review standard in them.

That was felt by Members in the House to be insufficient protection.

It was changed in the Senate to provide additional protection through the substantial evidence test.

I don’t think there’s any question that the — the feeling was that in return for letting the Secretary of Labor establish the standards, that there needed to be guarantees so that the Secretary of Labor would not go overboard.

That’s what I believe the substantial evidence test was for in this instance.

I make the same point to the second and third sentences of Section 6(b)(5), which, incidentally Mr. Alsup, contrary to his opening brief, now appears to concede, apply to both health and safety standards.

Those provisions were put into the statute in response to the concern stated in the House by the House minority members who had, I think, an immense power in this legislative debate.

The House originally passed the Daniels bill.

The Daniels bill was thought by the minority members to be too oriented towards the labor point of view.

They ultimately passed through the House the Staggers-Sikes substitute, which was the new Administration bill.

In any event, the House minority members criticized the Daniels bill as allowing standards to be set on the basis of views and argument.

In the Senate that was remedied by including the second and third sentences of Section 6(b)(5).

Those provisions require that standards be based on research demonstration and experiments.

It requires that in addition to the health and safety protection, that the agency consider the latest scientific data in the field and the experience of other health and safety agencies.

We think the experience of other health and safety agencies is very significant in this case, too.

We called, throughout the proceedings, to the Secretary’s attention the fact that other agencies, and I include EPA as a prime example, were making the very kind of estimates of benefits that we’re talking about in this case.

That was never addressed in their decision.

That point is still, I think, unaddressed by the agency.

Why EPA can engage in the kind of exercise that the Fifth Circuit contemplated and this agency can’t is absolutely beyond any understanding, as far as I can see.

Potter Stewart:

Mr. Warren, can I ask you, if you believe the Secretary’s authority with respect to — I’m not talking about his duty but his authority — with respect to standards concerning toxic materials is any different from his authority with respect to work practices, such as cleaning floors and the like?

Edward W. Warren:

Well, I think there are differences which reflect only one thing.

The reasonably necessary clause applies to both.

The first sentence of section 6(b)(5) applies only to toxic substances.

The second and third sentences of section 6(b)(5) apply to both health and safety standards.

The third sentence of 6(b)(5) mentions feasibility, so that I think safety standards comprehend feasibility.

But I do think there’s a difference, and I think that difference is why the debate arose and why there is a separate provision, and that is, section 2(b)(6) recognizes that there are differences in toxic substances because there are the problems that Mr. Cohen was referring to, the problems of determination of cause and effect relationships and determinations of latency period.

I don’t think that Congress wanted any different, you know, level of stringency of protection.

But I do think that Congress wanted the agency to recognize, and I think the language of the first sentence of Section 6(b)(5) does this, however inartfully, that there were going to be somewhat different problems presented.

John Paul Stevens:

Well, the first sentence imposes the duty on him to take — put on the toughest feasible standards.

But wouldn’t he have the authority as opposed to the duty to take the toughest feasible standard in anything he regulates?

Now, I’m asking you if his authority is any broader, rather than — I’m not asking …

Edward W. Warren:

I think that the two are really the same.

If you ask me, at the level of stringency, I think they come out the same, frankly.

Potter Stewart:

So that he would have authority in a different, as Justice Stewart suggested, cleaning floors or something, if a certain cleaning practice wasn’t followed there might be a risk of people slipping and killing themselves.

He’d have authority to change that practice.

Edward W. Warren:

And I think in coal mining, although it’s not covered by the statute, is an example, too.

There are going to be industries where it’s very dangerous work in those industries, if you’re working on tall buildings or construction projects, obviously it’s a terribly, terribly dangerous thing to do.

And I think the Secretary should be equally concerned about those hazards.

I think Congress was equally concerned about those hazards, really, because the legislative history reveals a discussion of, I think it was, 14,000 deaths a year and some 600,000 injuries.

And most of those obviously are related to health and safety problems, related to construction accidents and so forth and so on.

So I think that, in terms of the level of stringency, we should be coming out about the same place.

I think it’s significant, though, once again, to point out that the Secretary’s opening brief seemed to pin almost its entire case upon some distinction between toxic substances and safety standards, and now I think they concede, if I understand Mr. Alsup correctly, that safety standards are covered by the second and third sentences of section 5(b)(5), and it seems to me the reference to feasibility there must mean that we’re talking about roughly comparable levels of stringency.

I don’t want to leave the substantial evidence test without a little more discussion, because I think that is so critical to the case.

We must remember we don’t have any evidence here of actual problems below 100 parts per million.

Now, it’s important to recognize that industry in this case has looked, and looked very hard, to discover whether there are any problems.

There have been numerous epidemiological studies covering tens of thousands of workers.

To be sure, the Secretary has faulted the methodology of some of those studies, but every epidemiological study has the same problems, including the ones they are relying on.

The studies we are talking about have been relied on by, for instance, the National Academy of Sciences, who felt that the Thorpe study which was done by the petroleum industry showed there just couldn’t be any significant problem there, regardless of any minor methodological defects.

You just couldn’t have a big problem, there just isn’t a big problem that we’re talking about here.

Thurgood Marshall:

Is this much the same argument that was made in the asbestos case?

Edward W. Warren:

No, Your Honor, this is a very different case than the asbestos case.

The asbestos case was a challenge by my friend, Mr. Cohen, here to the asbestos standard, it was not a challenge by employers.

Employers didn’t participate except as amici in that case at all.

They made no arguments, such as I’m making here today nobody raised the statutory provisions I’m talking about, nobody raised the issues of whether or not there were benefits to justify a standard.

Thurgood Marshall:

But they did find that it was bad?

Edward W. Warren:

They did find it was bad.

Thurgood Marshall:

Three years later.

Edward W. Warren:

They found it was bad.

Thurgood Marshall:

Three years later.

Edward W. Warren:

They found it was bad at the time.

Thurgood Marshall:

Is there a chance that might happen here?

Edward W. Warren:

Pardon?

Thurgood Marshall:

Is there a chance that might happen here?

Edward W. Warren:

Is there a chance that it might happen here?

Thurgood Marshall:

Yes, sir.

Edward W. Warren:

There’s always a chance that anything can happen, and I’m not saying that this won’t happen.

Thurgood Marshall:

Well, you know I didn’t ask that question, don’t you?

Edward W. Warren:

Yes.

[Laughter]

Thurgood Marshall:

Why don’t you answer the one I asked?

Edward W. Warren:

The answer is I can’t say that it will not happen.

But I can give you some pretty good reasons why I don’t believe that it is going to happen.

The reason I don’t believe it’s going to happen is because benzene has been in use for over a hundred years.

This is a substance we know about as much about as any substance we’re going to know about.

There have been workers working with it for a hundred years.

There have been lots of studies.

There is more medical literature on this substance than just about any one I know anything about.

And I don’t think we’re going to find anything like that.

I might add, too, that the asbestos situation, Mr. Cohen is saying that it was not shown to be a carcinogen at the time that standard was passed.

Edward W. Warren:

That’s really sort of an exaggeration, I think.

It was shown to be a very, very serious respiratory problem-causing mesothelioma and asbestosis.

Everybody has known this has been a significant problem, and I don’t think it’s —

Thurgood Marshall:

And you found no harm from benzene?

Edward W. Warren:

There’s no harm from benzene recorded under 100 parts per million.

That’s true.

And there’s been a lot of effort to find harm.

What we’re talking about here is a theoretical construct, really we’re talking about a substance where some harm has been demonstrated at high levels and where we’re theorizing that there must be some harm at lower levels, and from that theory, and nothing else, the Secretary is constructing an argument, which is unsupported by anything, that there are so-called depreciable benefits.

Thurgood Marshall:

But the theory is supported by the leukemia death, is it not?

In that people are exposed to leukemia — exposed to benzene, died of leukemia at a higher proportion than the normal population even though you couldn’t say as a fact that the leukemia was brought on by benzene.

Edward W. Warren:

Well, you can say this, that at the levels such as Dr. Aksoy looked at, 2, 3, 400 parts per million, Dr. Vigliani, you can see a relationship between the instance of leukemia and the benzene exposure.

On the other hand, when we look below that level, we don’t see anything.

There certainly is a theoretical case.

There were witnesses who testified to that theoretical case in the hearing but it proves too much, because it proves that something that can be proved for any of those thousands of carcinogens that Mr. Alsup was talking about, or many of those chemical substances which are introduced into the work place every 20 minutes.

What it says is we don’t have any way of distinguishing all of those chemicals.

But what the Fifth Circuit is trying to do is to grapple with that question without going too far.

What the Fifth Circuit is trying to do is to take what Congress did, the reasonably necessary provision and the feasibility provision and the substantial evidence provision and the best available evidence provision, and it’s saying this much they must do.

Substantial evidence requires this much. Reasonableness, reasonable necessity requires this much.

The agency cannot carry out its over-all statutory objective unless it does this much.

It didn’t do those things here.

Potter Stewart:

Mr. Warren, there’s another issue in this case, the dermal issue or the protective clothing issue, that I think your brothers didn’t talk about.

Edward W. Warren:

Mr. Lettow, yes.

Potter Stewart:

He’s going to do it?

Edward W. Warren:

Yes.

Potter Stewart:

All right.

Edward W. Warren:

In conclusion, I think it’s significant that workers here are not going unprotected.

We have an existing 10 parts per million standard.

I think that’s providing fully adequate protection.

What the Secretary’s argument boils down to as far as I can tell is a plea that this Court give blind deference to its so-called expertise.

As I understand their argument, what it amounts to is that the absence of any evidence supporting their position is really a factor that works in their favor.

Edward W. Warren:

What they’re saying is because we don’t have any evidence, we’re free to do whatever we want.

Now, maybe I’ve misinterpreted them, but that’s the way I read their position.

And if that’s the position, it seems to me that neither the substantial evidence test nor judicial review of standards such as this —

Potter Stewart:

That’s not a fair statement of their position.

Their position is that by extrapolating here you can reasonably conclude that there may be an appreciable risk of harm here.

And the agency so found.

There may be a harm that’s their position, and that’s enough.

That’s not like saying we don’t know — I mean that there is no harm.

Edward W. Warren:

Well, I appreciate what you’re saying, and I think that is their position.

What their position is is that we don’t know, and so therefore we can extrapolate — without trying to extrapolate we can simply say there are appreciable benefits.

It’s like a bolt out of the blue.

Potter Stewart:

Well, I thought your position was that will in effect admit that there may be an appreciable harm but that’s not enough.

Edward W. Warren:

No, I —

Potter Stewart:

Aren’t you willing to admit that there’s a possibility of an appreciable harm?

Edward W. Warren:

If you say a possibility of appreciable harm, yes.

Potter Stewart:

That may be a possibility.

Edward W. Warren:

That’s right.

Potter Stewart:

So we’ve got to start from the assumption, I think, that there is a possibility that somebody may die of cancer if we set aside these regulations.

Edward W. Warren:

We accept that.

Potter Stewart:

Because of exposure to this substance at less than 10 parts per million.

Edward W. Warren:

We accept that there is some possibility of appreciable benefit.

What we’re saying —

Potter Stewart:

And if you win this case, somebody may die as a result of it.

We have to start with that assumption.

Edward W. Warren:

I wouldn’t put it quite like that, but I would think —

Potter Stewart:

Well, we’ve got to be realistic.

[Laughter]

Edward W. Warren:

But I think that — but seriously, I think we would say that there is some possibility of the appreciable harm you’re talking about.

What we’re talking about, though, is did the Secretary do the best job it could, given the best available evidence to look at this problem?

We say the answer to that question is no.

Edward W. Warren:

EPA has shown, and I don’t think there can be a better demonstration, that the agency could have done a better job than it did in this instance, to show what — appreciable harm, what does that mean?

It’s unexplained.

It cries out for meaning.

And it seems to me the very fact that the agency said it suggests that they must recognize that they can’t operate with utter and complete freedom.

Byron R. White:

Mr. Warren, what troubles me about your position, frankly, is what happens if you win and it goes back and they say, Well, we’re going to accept — I forget the name of the doctor’s finding we think that two deaths in six years are predictable, and will occur.

And a half billion dollars is worth saving those two lives.

Then what happens?

Edward W. Warren:

Well, that’s an important question, I think, which — the first point is that once the proceedings begin to focus in on what I think is the real question What’s this going to accomplish for the worker?

I think we’re going to find the differences between sides narrowing.

I think we’re going to find people beginning to agree about what’s useful protection and what’s less useful protection.

I don’t think that if the hypothetical you pose were to occur that the Secretary would issue this standard.

I really don’t believe that.

I don’t believe that, in light of the other standards which have been issued, which I think can easily be demonstrated and indeed have been demonstrated to produce much more benefit than this, I just don’t believe this would occur.

There may come a time when the Secretary does something as flatly unreasonable.

I would say unreasonable is bad, and I think in that event we can argue about it, but we’ll be arguing about it at a time where the assumptions on both sides are clearly stated, where the evidence supporting the assumptions on both sides are clearly stated.

Where we really have substantial evidence which can be reviewed on both sides of the question.

That’s not what we have here.

So it seems to me that’s a question that theoretically arises, but I don’t think it will happen in the real world.

Those questions are going to be resolved rationally and reasonably once people start looking at what really matters.

And what really matters is protecting these workers that Mr. Cohen and I both would like to protect.

Thank you.