Text-only Preview

Richard N. Langlois
People today are therefore far less divided than one imagines; they
argue incessantly to know into which hands power will be placed;

but they content themselves calmly with the duties and rights of
that power.

—de Tocquevill&
T0 the practical mind, particularly one trained in economics, it
might sound a trifle paradoxical to suggest that the truly important
issues in a public policy debate have nothing to do with the relative
efficacy of the policy options. Paradoxical or not, this is very often
the case. And there is no better illustration than the ongoing — and
heated — debate over alternative approaches to pollution control.
Issues of workability and economic efficiency are certainly promi-
nent in this policy discussion. But, as with many political questions,
the more important issues lie at a deeper, or at any rate a less visi-
ble, level.
In political terms, this debate has lately come to be portrayed as a
Reaganites, who favor applying the test of cost-benefit analysis to
all administrative regulations put forth by the government and who
support “economic” approaches to pollution control. On the other
side are environmental, “consumer,” and other groups, who op-
pose the use of cost-benefit analysis and who are suspicious of any
Cato Journal, vol. 2, No. I (Spring 1982). Copyright © Cato Institute. All rights
The author is Adjunct Assistant Professor of Economics at New York University,
New York 10003. He would particularly like to thank Gerald O’Driscoll for helpful
advice and comments.
‘Alexis de Tocqueville, in J.P. Mayer, ed., De Ia Dimocratie en Amirique: Les
Grands TheThes
(Paris: Gallimard, 1968), p. 315 (translation mine).

“economic” approaches. The dialectic between the two points of
view came to some national attention this past summer when the
Supreme Court barred one federal agency from employing cost-
benefit analysis in formulating its regulations.
One consequence of this controversy is that the ethical and
‘‘economic” regulatory techniques have come increasingly under
fire from environmentalists and others. Most critics have taken the
moral high-road, charging that utilitarianism —
the doctrine
underlying cost-benefit analysis — is ethically flawed and morally
inferior. In so arguing, most critics have also insisted upon berating
utilitarianism for its inattention to basic human rights and for its
My intention in this paper is not to defend utilitarianism or cost-
benefit analysis. It seems to me that the critique of utilitarianism of-
fered by environmentalists (and others) is very often on target.
Rather, my point is that a critique of utilitarian moral philosophy is
not by itself a critique of the use of cost-benefit analysis in
regulatory decision-making. What is universally overlooked, I con-
tend, is that the latter necessarily involves a joint critique, a
utilitarianism) and the institutional structure of regulation to which
that approach is applied.
One implication is that
perhaps surprisingly — the ethical case
for utilitarian behavior by an agent (or agency) acting for a political
collective can be strong even when the ethical case for utilitarian
behavior by an individual is weak. This does not mean that cen-
tralized administration using cost-benefit analysis is desirable.
There are good reasons for wishing to rid social governance of the
technocratic. But an ethics of basic rights and absolute claims re-
quires an institutional structure appropriate to such rights and
There do exist rights-based regulatory alternatives, some based
on historic conceptions of rights in common law and some based on
government-granted “pollution rights.” One might thus expect
that, whatever pragmatic reservations they might have, opponents
of cost-benefit analysis would be somewhat sympathetic to such
rights-based schemes on philosophical grounds. In fact, quite the
opposite is the case. Environmentalists, for example, bristle at the
2See, e.g,, Philip Shabecoff, “Safety Agency to Forgo ‘Cost-Benefit Analysis,’ “New
York Times, 13 July 1981.

very mention of rights-based pollution regulation, and insist on
maintaining centralized administrative control. The reason for this,
I will be forced to conclude, is that, in the end, the environmentalist
is concerned with basic human rights to no greater extent than is
the utilitarian economist. And recognizing this is the key to under-
standing what the debate over pollution policy is really all about.
It is fairly easy to demonstrate, I believe, that, in one sense at
least, this policy debate is not ultimately a battle between groups
with radically different philosophies. The argument between
Reaganites and environmentalists, I will suggest, is ultimately a tus-
sle between formally identical political philosophies — both of
which are “conservative” in a well-defined sense — that disagree
only about the specific constellation of personal virtues each wishes
regulation to promote.
The Critique of Cost-Benefit Analysis
• One of the most persistent beliefs about cost-benefit analysis and
related disciplines, writes Laurence Tribe, “is a conviction of their
transparency to considerations of value and their neutrality with
respect to fundamental world-wide views and to more-or-less
ultimate ends.”
This, says Tribe, is much mistaken. Such tech-
niques necessarily imply a world-view of their own — the world-
view of economic utilitarianism.
Modern policy-analytic techniques, lie argues, are wedded to an
conception of rationality. Also referred to as
“means/ends” rationality, this conception defines the rationality of
a policy or action in terms of its efficacy

indeed, its optimality

in achieving a stated goal.
In a typical cost-benefit analysis, the social questions are framed
in terms of a “social decision” — whether to institute a regulation,
how much pollution to allow, whether to lease certain offshore oil
tracts — understood in strict analogy with an individual decision.
The relevant structure of means is established by the analyst, in the
service of the (often hypothetical) “decision-maker,” and the objec-
tive is to choose the best social policy in light of “social’’ costs and
benefits, of the somehow aggregated preferences of the many in-
dividuals who would be affected.
3Laurence H. Tribe, “Policy Science: Analysis or Ideology?” Philosophy and Public
Affairs (1972), p. 75.
41n simple terms, one chooses the option with the highest net “social benefit’’ over
‘social cost,” More formally, though, one is normally concerned with picking a

There are two closely related considerations that Tribe and like-
minded critics adduce against this approach: reductionism and col-
lapsing process into result. Cost-benefit techniques operate by
disassembling the effects of a ‘‘social decision’’ into its component
parts, ascertaining the benefit or cost of each part, and then totaling
up the results. This procedure obscures and does violence to the
very complex and not fully measurable web of social valuation that
actually underlies the problem. In particular, the cost-benefit
technique is ad hoc and ahistoric; it is concerned only with the goal
of maximization, relegating all else — including human rights — to
the derivative status of means toward that end.
Interestingly, Tribe illustrates his complaint by reference to the
literature of law and economics. In the dominant form of the
economic approach to law, which Tribe traces to Ronald Coase and
Guido Calabresi,
and which we would now associate with the
name of Richard Posner, the assignment of rights for purposes of
tort law is conceived, as both a descriptive and a normative matter,
in terms of the “efficiency” of the allocation.
In the classic example analyzed by Coase, a train passing through
farmland emits sparks that set afire and destroy the wheat growing
on a farmer’s land near the tracks. The efficiency approach to law
asks: Does the farmer have a right to grow wheat unmolested (and
therefore a claim against the railroad) or does the railroad have a
right to emit sparks? The answer, according to the efficiency view,
should depend — and in fact has depended — upon which assign-
ment of rights maximizes net wealth, that is, upon which assign-
ment minimizes the sum of wheat-destruction and fire-avoidance
costs.°Indeed, when there are no transactions costs (like the costs
“Pareto-improving” choice, i.e., a choice that makes at least one person better off
without making anyone worse off. In circumstances in which income redistribution
is possible, this criterion reduces to the excess-social-benefits-over-social-cost
criterion, Indeed, under the so-called Kaldor-Hicks criterion, one need show only
that the best choice is a potential Pareto improvement — that the winners could in
principle (or in the absence of transactions costs) compensate the losers, even though
the losers need never actually be compensated. See generally E.J. Mishan, Cost-
Benefit Analysis, 2d ed. (New York: Prager Publishers, 1976).
5Tribe, “Technology Assessment and the Fourth Discontinuity: The Limits of In-
strumental Rationality,” Southern California Law Review, 46(1973): 627-631, Readers
of Robert Nozick (whom Tribe cites) will recognize an affinity to the arguments in
Anarchy, State, and Utopia (New York: Basic Books, 1974), esp. chap. 7.
6Tribe, “Technology Assessment,” notes 37 and 38.
‘See generally Posner, Economic Analysis of Law (Boston: Little, Brown & Co., 1972).
8”The Problem of Social Cost,” Journal ofLaw and Economics 3 (October 1960): 1.
9As I will suggest below, maximizing wealth is not the same as maximizing utility.
Posner makes this distinction and defends the wealth-maximization approach in

of organizing dozens of farmers to bargain with the railroad), the ini-
tial assignment of rights is irrelevant, since, through bargaining,
the right would ultimately be acquired by the party whose exercise
of it maximizes net wealth. When there are transactions costs, it is
the job of the court to undertake a wealth-maximization calculation
and to assign the right to the party who would ultimately have
possessed it had there been no transactions costs.
To Tribe, this attempt to analyze the assignment of rights from
the common-denominator of wealth obscures the complex socio-
cultural role that rights play in a person’s structure of preferences;
“for to be ‘assigned’ a right on efficiency grounds fails to satisfy the
particular needs that can be met only by a shared social and legal
understanding that the right belongs to the individual because the
capacity and opportunity it embodies is organically and historically
a part of the person that he is and not for any purely contingent and
essentially managerial
Reading Tribe and other critics, one is left with a strong sense that
utilitarianism and cost-benefit analysis are flawed — and are to be
rejected — because of their callousness towards the individual, his
rights, and the processes by which those rights are exercised. ‘‘The
notion of human rights,” as Steven Kelman puts it in his recent
“ethical critique” of cost-benefit analysis, “involves the idea that
people may make certain claims to be allowed to act in certain ways
or to be treated in certain ways, even if the sum of benefits achieved
thereby does not outweigh the sum of costs.” A right is not some-
thing that can be assigned on “efficiency’’ grounds; a right is
precisely an individual’s ‘trump’” against the claims of efficiency,
his protection against social “utility monsters” like the one that
recently devoured the Poletown section of Detroit.
The problem
“Utilitarianism, Economics, and Legal Theory, “Journal of Legal Studies VIII (January
1979), p. 103.
10Tribe, “Technology Assessment,” p. 629, emphasis original. This quotation may
perhaps paint the argument as rather vague and emotional. But other philosophers
have argued a similar point on logical grounds. See, for example, Charles Fried, Right
and Wrong (Cambridge: Harvard University Press, 1978), esp. pp. 102-104.
“Cost-Benefit Analysis: An
Ethical Critique,”
January/February 1981, p. 36.
‘2 “Trump” is a term favored by Ronald Dworkin. See his Taking Rights Seriously
(Cambridge: Harvard University Press, 1977).
13For a coherent accouut, see William Safire, “Poletown Wrecker’s Ball,” New York
30 April 1981, p. A31. I should note that it seems unlikely that Detroit’s abuse
of eminent domain powers was a real utility monster in the sense that the sum of
benefits actually outweighed costs. To the contrary, it strikes me as likely that the
project was, by most standards, a big utility loser. This underscores a point I will

with cost-benefit analysis, we are encouraged to believe, is that, in
reducing social questions to the common metric of a homogenized
utility, it treats human beings — and their historically rich and
idiosyncratic circumstances — with insufficient respect.
I’m inclined to agree. To a large extent, this is indeed the problem
with cost-benefit and related approaches. But I’m also inclined to
think that this line of criticism is neither central to, nor maintained
environment-minded critics.
Utilitarianism, Moral Theory, and Political Theory
Early on in his critique, Kelman marvels at the complacency of
economists in using cost-benefit analysis uncritically at a time
when that technique’s philosophical underpinnings —
ianism —
represent “a minority position among contemporary
moral philosophers.”1
Controversiality, or even unfashionability,
is never a very good argument against a philosophic doctrine. But it
is certainly true that the recent trend in political philosophy (let
alone moral philosophy) has been away from utilitarianism and
toward theories of basic human rights. In chiding economists in
this way, Kelman would seem to be allying himself with those he
describes as the “non-utilitarian philosophers.”
He is not entirely
unjustified in allowing us to draw this inference; but, as I will even-
tually suggest, his reasons for objecting to utilitarianism are
ultimately rather different from those of the more prominent of
these philosophers.
The Two Forms of Utilitarianism
Let me begin with a not-so-minor quibble about Kelman’s attack
on utilitarianism.
Utilitarian doctrine comes in two basic flavors - The primary ver-
sion, and the
one Kelman implicitly means when he says
‘‘utilitarianism,’’ is ‘‘act’’ (or sometimes ‘‘extreme’’j utilitarianism.
As a moral theory, act-utilitarianism holds, roughly speaking, that a
person’s action is morally correct when the benefits to society of
taking the action outweigh the costs to society. Given such a defini-
tion, it is quite easy to construct, as Kelman does, hypothetical
situations in which the balance of costs and benefits makes it
make below: It is not utility calculations so much as the notion of an overriding
“social interest,” however constructed, that causes “monsters.”
Kelman, p. 4.
‘ lbid., p.36.

morally correct (say) to lie, to break a sworn promise, or to frame an
innocent man — and actually immoral by definition not to do so.
Rule-utilitarianism, advanced in large measures in reaction to act-
utilitarianism’s counter-intuitive quality, is quite a different matter.
Here the cost-benefit test is applied not to an individual’s actions
directly but to a rule or system of rules. That system of rules is best
which maximizes social utility; and the morality of an individual’s
conduct is measured by its conformance with the maximizing set of
rules — not by its own costs and benefits measured in isolation.’
Rule-utilitarianism is an institutional or, as they tend to say
nowadays, a “systemic” notion.
This immediately points up a small flaw in Kelman’s analysis.
Are we not entitled to view pollution, health, and safety regulation
as problems in the selection of systems of rules? To the extent that
this is so, the application of cost-benefit analysis to regulatory deci-
sions may very often be understood in rule-utilitarian — not act-
utilitarian — terms. It may remain, as I am indeed inclined in part
to agree, that such decisions should not in the end be made on
utilitarian grounds; but this follows not at all from Kelman’s
arguments against ‘utilitarianism.”’
is a
teleological’’ form of moral theory: It postulates a conception of
the good (in this case social utility) and deduces the right — the set
of moral rules — using the logical procedures of maximization
The good is prior to the right, and moral rules are but in-
‘ Mociern discussions of rule-utilitarianism generally trace from John Rawls’s
seminal article, “Two Concepts of Rules,” The Philosophical Review LXIV (1955),
reprinted in (among otherplaces) Michael D. Bayles, ed., Contemporary Utilitarianism
(Garden city: Doubleday, 1968), p. 59. (Sir Roy 1-larrod’s article “Utilitarianism
Revised,” Mind 45 (1936): 137, seems to have attracted far less attention.) The con-
cept probably goes back at least as far as I-fume, though, and there is much discus-
sion over whether John Stuart Mill was a rule-utilitarian. The terms “act” and “rule”
themselves were not coined until 1959 by RB. Brandt, Ethical Theory (Raglewood
Cliffs, N.J.: Prentice-Hall).
‘ Kelman notes that a sophisticated act-utilitarian will include in his calculation the
effect his action might have on his own and other people’s future propensity to
follow utilitariaa precepts, arid he is careful in his own examples to construct situa-
tions in which the act in question has only negligible effects on the institutions of
truth-telling, promise-keeping, law enforcement, etc. But this makes act-
utilitarianism no less exclusively the focus of his arguments; rule-utilitarianism re-
quires that one obey the institutional rule even when, as it were, the existential costs
and benefits suggest otherwise. Of course, this is not to say that cost-benefit analysis
is ever actually used in a rule-utilitarian way. But that is a different argument from
the one Kelman makes, and it is not an indictment of utilitarianism in priaciple~
t5See, for example, John Harsanyi, “Morality and the Theory of Rational Behavior,”
Social Research 44(1977): 623; and ‘Rule Utilitarianism, Rights, Obligations, and the

struments for achieving the good. There is also a quite different
kind of moral theory: non-consequentialist or “deontological”
theory. And it is here we must look to find the center of gravity of
modern “non-utilitarian” thought.’
Deontological Moral Theories
A deontological theory turns the tables and gives priority to the
right over the good. Instead of attempting to derive the ap-
propriateness of a rule solely from a consideration of consequences,
it seeks to deduce rules from more primitive postulates about rules
themselves. This approach is no less “rational” than utilitarianism.
But it may arguably be “rational” in a different sense; and if there is
a distinction between instrumental and non-instrumental rationali-
ty, perhaps here is where that distinction is to be made.
Almost without exception, recent non-utilitarian theorists have
chosen to associate themselves with the teachings of Kant and, in
particular, with the notion that a theory of rights should derive
from a consideration of the person and an analysis of the treatment
appropriate to him. Philosophers have differed in their interpreta-
tions of Kant on this score, but all agree that moral behavior in-
volves treating others with suitable respect. Some emphasize a
respect for the historical uniqueness and particularity of the in-
dividual, while others stress not respect for its own sake but the
equality with which an institutional structure displays such
respect. °
To Charles Fried, for example, systems of rights must embody “a
respect for persons as the ultimate moral particulars
21 Robert
Nozick holds that such a system must “reflect the underlying Kantian
principle that individuals are ends and not merely means; they may
not be sacrificed or used for the achieving of other ends without their
Theory of Rational Behavior,” Theory and Decision 12 (June 1980).
‘°Myown view is that rule-utilitarianism of a very sophisticated sort is not without
merit — and, indeed, not entirely inconsistent with deontological theories. But this is
not the place to elaborate this view. (I’m thinking here especially of the work of F.A.
Hayek. See his “Notes on the Evolution of Systems of Rules of Conduct,” in Studies
in Philosophy, Politics, and Economics
(Chicago: The University of Chicago Press,
1967); and generally Law, Legislation, and Liberty, vol. 1: Rules and Order (Chicago:
The University of Chicago Press, 1973).) In examining deontological theories in this
section, I mean primarily to contrast this important school of “non-utilitarian
philosophers” with the ideas of Kelman, Tribe, and other pro.environment writers
who attack utilitarianism.
20H.L.A. Hart, “Between Utility and Rights,” in Alan Ryan, ed., The Idea ofFreedom,
Essays in Honour of Sir Isaiah Berlin
(Oxford: Oxford University Press, 1979), p. 77.
2tPried, p.20.

And Ronald Dworkin speaks of “the vague but power-
ful idea of human dignity,” associated with Kant, which “supposes
that there are ways of treating a man that are inconsistent with
recognizing him as a full member of the human community and
holds that such treatment is profoundly unjust.’
Like utilitarian approaches, deontological theories are both moral
theories and political-institutional theories. This is so not merely
because their authors invariably jump back and forth between the
two realms; rather, it is because the moral and the institutional can-
not easily be separated. This is not therefore to say that a moral
theory is immediately a political or institutional one; indeed, many
of the difficulties that arise in trying to find moral foundations for
social policy attend a careless transition from the level of personal
morality to the level of social institutions.
The connection between the two levels is illuminated by casting
moral prescriptions in the language of rights. A deontological moral
theory yields a system of rules or, as Fried argues, of “moral
abso1utes”~that can be understood as moral rights. The prob-
lem, as we’ll see, then becomes one of transforming into political
or legal rights whichever moral rights emerge from one’s moral
Ethics and Institutions
Those who criticize the use of cost-benefit analysis in regulatory
decision-making are frequently quite anxious to talk in terms of
rights and absolute claims and to deride utilitarianism for ignoring
such claims. “We do not do cost-benefit analyses of freedom of
speech or trial by jury,” says Kelman. “The Bill of Rights was not
BARGed. - - - [T]he Emancipation Proclamation was not subjected
to an inflationary impact statement.”
What this seems to imply is
that administrative decisions by regulatory agencies are logically
equivalent in form to the Bill of Rights or the Emancipation Pro-
clamation, and that non-utilitarian regulation somehow involves
22Nozick, p. 30.
23Dworkin, pp. 198-199. Not content to stop here, though, Dworkin quickly adjoins
this to the notion of political equality, which “supposes that the weaker members of
a political community are entitled to the same concern and respect of their govern-
ment as the more powerful members have secured for themselves
Fried, p. l.
25Kelman, pp. 35-36. The remark about the Emancipation Proclamation was
evidently made by the United Steelworkers Union in a comment on an 051-IA rule
“to reduce worker exposure to carcinogenic coke-oven emissions.” The acronym
RARG refers to a risk-analysis procedure, I believe,

inserting absolute claims directly into the rule-making process as
now constituted. But to believe this, as you might guess, is in fact to
make a number of illegitimate leaps.
Positive and Negative Rights
Let’s back up. On the level of moral theory, once again, it is possi-
ble to cast the imperatives of deontology in terms of rights. For ex-
ample, a Kantian respect for the person dictates that one has an
obligation not to harm someone else intentionally; another way to
say this is that a person has a moral right not to be harmed inten-
tiohally by another. This right against harm is a negative right: it dic-
tates what you may not do to me rather than what you must do for
me. There are also positive rights, which do detail what you (or
some specified persons or institutions) must do for me: a child, for
example, may be viewed as having a positive right to proper care
and feeding. This is a much-used distinction, and it is not, I think,
The important point is that the two types of rights are not sym-
metric. A negative moral right can carry over immediately into the
institutional realm without much ado. If I have a moral right not to
be harmed,
this can quite easily become a legal right not to be
harmed. But if I have some kind of positive moral right, I cannot
transform into the institutional realm without specifying the institu-
tional mechanism that is to provide me with the content of my
positive legal right. (Whom do I take to court if I don’t get fed?)
Positive legal rights always make a claim on the resources of
others. As Fried puts it, such rights ‘‘are inevitably asserted to
scarce goods, and consequently scarcity implies a limit to the claim.
Negative rights, however, the rights not to be interfered with in for-
bidden ways, do not appear to have such natural, such inevitable
Moreover, some positive rights have more natural
limits than others. For example, it is quite clear — to within fairly
narrow limits — what it means to give someone a jury trial. But it is
far less clear what it means to give someone “a clean environment”
or ‘‘safety.’’
For the most part, the Emancipation Proclamation and the Bill of
Rights guarantee negative rights. And there is no problem in deduc-
ing negative political rights of this sort directly from a deontological
moral theory. (In fact, these particular rights accord extremely well
course, enforcing that right may involve a positive right: The right to law-
enforcement services: but that makes the right against harm itself no less negative.
27Fried, p. 110.