Dworkin, R

Justice in the Distribution of Health Care

pp. 203-222

Clayton, M & Williams, A, (2000) The Ideal of Equality, Basingstoke; New York: Macmillan Press Ltd

Staff and students of University of Warwick are reminded that copyright subsists in this extract and the work from which it
was taken. This Digital Copy has been made under the terms of a CLA licence which allows you to:

• access and download a copy;

• print out a copy;

Please note that this material is for use ONLY by students registered on the course of study as stated in the section
below. All other staff and students are only entitled to browse the material and should not download and/or print out
a copy.

This Digital Copy and any digital or printed copy supplied to or made by you under the terms of this Licence are for use in
connection with this Course of Study. You may retain such copies after the end of the course, but strictly for your own
personal use.

All copies (including electronic copies) shall include this Copyright Notice and shall be destroyed and/or deleted if and
when required by University of Warwick.

Except as provided for by copyright law, no further copying, storage or distribution (including by e-mail) is permitted
without the consent of the copyright holder.

The author (which term includes artists and other visual creators) has moral rights in the work and neither staff nor students
may cause, or permit, the distortion, mutilation or other modification of the work, or any other derogatory treatment of it,
which would be prejudicial to the honour or reputation of the author.

Course of Study: PO107 - Introduction to Politics

Title: The Ideal of Equality
Name of Author: Clayton, M & Williams, A
Name of Publisher: Macmillan Press Ltd
9
Justice in the Distribution of
Health Care
Ronald Dworkin *

You might think it very odd that, living as I do half the year in
Britain and half in the United States, I've come to Canada to talk
about justice in the distribution of medical care. In both those coun­
tries, the Canadian structure for the distribution of health care is
taken by many people to be a model of success. Your Dean men­
tioned my former colleague at University College, Oxford, who is
now my President. Under his administration, and under the super­
vision of his wife, the United States is now, as you know, engaged
in a massive re-examination of health care. Almost every day an
article appears in the American press about the Canadian plan, which
is widely proposed as a model for Americans to follow. Neverthe­
less, strains on your system of health care are beginning to become
evident. There's talk of rationing, and more people go south of the
border to seek medical care. Doctors in negotiation with the pro­
vincial authorities claim with greater stridency that they are seriously
underpaid. The system, nevertheless, is producing more Canadian
doctors than economists think wise. You, too, will face the prob­
lems I'll discuss this evening. You too must worry about justice in
the distribution of health care when it comes and I'm sure it will
come to rationing health care explicitly.
Some people, particularly in America now, say there is really no
need to ration health care. They agree that medical expense already

• This lecture was delivered on 17 March 1993 at the Faculty of Law, McGill
University, as the Inaugural Lecture of the McGiIl Lectures in Jurisprudence
and Public Policy.

203
204 Ronald Dworkin

constitutes an alarming proportion of the American economy, and

that, even though America spends that much in the aggregate, forty
million Americans are wholly uninsured or without any adequate
health care, which is intolerable. But they deny that correcting these
deficiencies will reqUire some form of rationing: they say that there
is so much waste and inefficiency in the health care system that, if
these were eliminated, we could save enough money to insure that
everyone had all the medical treatment he needed. We know what
they mean. The administrative inefficiency of United States medi­
cal insurance companies and carriers is legendary. American doctors'
salaries are large and, according to many people, inflated. The av­
erage medical salary two years ago in the United States was over
$160000 the average. Nevertheless, a series of recent studies sug­
gests that even if administrative efficiency were greatly improved,
and even if doctors' salaries were capped at some reasonable level,
rationing of health care would still be inevitable, because by far
the biggest cause of the explosion in health care costs (not only in
the United States, but, I believe, in Canada as well) is a massive
supply of new technology. It isn't that we're paying all that much
more for what we formerly bought cheaper; it is that we now have
so much more to buy.
Many politicians and some doctors say that much of the new,
expensive technology is 'unnecessary' or 'wasteful' or 'inappropri­
ate'. But if you look to see what they mean, you find they have in
mind techniques that are (as it's often put) 'Iow yield', which is
not the same as 'no yield'. They pOint, for example, to massive
mammography screening of women under the age of fifty, or to
the heavy use, in some medical facilities, of magnetic resonance
imagining. A society that spends a great deal of money on routine
screening or expensive diagnostic equipment may not save many
more lives than a society that does not. But it will presumably save
some more lives, and that means that we cannot appeal just to
efficiency as an abstract value to justify saving the cost. We cannot
recommend eliminating 'inappropriate' medical care without de­
ciding what medical care is appropriate, and why, and that, in turn,
depends upon how we answer the question: 'how much medicine
should a SOCiety provide?'
That question can usefully be divided into two more specific ones.
The first is the question of a community's aggregate health care
budget. Money spent on health care (l include not just acute care
but also preventative medicine, care of the chronically sick or disabled,
 Justice in the Distribution of Health Care 205

and so forth) is money that might be spent on education, or on

economic infrastructure that will produce more jobs. How much of
the overall budget should be devoted to health care instead of other
plainly valuable projects, like these? The second question, though
it's really part of the first, is the question of distribution. Once it's
established what a society should spend overall on health care, then
it must also be decided who should have that care, and on what
basis it should be allocated. Of course, nations struggling with health
care costs must resolve many issues beyond these twin questions
of justice. There are economic questions - what are reliable predic­
tors of how much a particular health care plan or structure will
cost? There are administrative issues: what is the most efficient
organization for administering any particular plan? There are medical
questions: what is the likely impact of a particular program on
morbidity and mortality? Above all, there is the political problem:
what plans will a particular democracy in fact be willing to accept
and pay for? I don't mean to denigrate the importance of these
various problems, or to deny their evident connections with the
problems of justice. But I shall concentrate on the latter. To repeat:
In all justice, how much should a decent society spend on medical
care, broadly described'! In all justice, how should that society dis­
tribute what it does spend - who should get what? Behind these
two questions lies a more explicitly philosophical one. What is the
right standard to use in answering these questions? What should
we take as our ideal of justice in medical care?

II

I begin by describing an ancient and attractive ideal that many

people instinctively accept, which I shall call the ideal of insula­
tion. It has three features. The first argues that health care is, as
Rene Descartes put it, chief among all goods: that the most important
thing is life and health and everything else is of minor importance
beside it. The second component of the insulation ideal is equality,
The ideal supposes that even in a society which is otherwise very
inegalitarian - indeed even in a society in which equality is de­
spised as a general political goal medical care should nevertheless
be distributed in an egalitarian way so that no one is denied care
he needs simply because of an inability to pay. The third compo­
nent (it really flows from the other two) is the old principle of
rescue, which holds that it is intolerable when people die, though
206 Ronald Dworkin

their lives could have been saved, because the necessary resources
were withheld on grounds of economy.
This ideal of insulation has exerted great power throughout his­
tory. Hospitals have always been paradigm examples of appropriate
charities, and religion has, from ancient days, always been associ­
ated with them. Contemporary political philosophers I have Michael
Walzer in mind, for example - say that the provision of medicine
constitutes a separate sphere of justice, and that in that sphere
decency, community, solidarity, and equality must reign. The power
of the insulation ideal is so great that people often think that though
the administrative, medical, economic and political problems I de­
scribed are intellectually daunting, the questions of justice are not:
that it is clear what the ideal of justice demands in health care,
and that our only problem is that we are unwilling to live up to
that ideal. That is, I believe, a serious mistake. The crisis in health
care includes a crisis in our conception of what a just health care
system would be - what answers we should give to the questions
of justice I set out. We face that intellectual crisis because it has
become clear that the insulation ideal, for all its ancient popular­
ity, is now irrelevant. Consider the first question I posed: the problem
of the aggregate expense a decent society will commit to health
care, as against competing needs and values. What advice does the
ideal of insulation give? It says a society should spend all it can on
health care until the next dollar it spends would buy no gain in
health or life expectancy at all. Of course no society ever did orga­
nize its affairs in that way, any more than any sane individual
organizes a plan of life with the goal of making that life as long
and as healthy as possible. In past centuries, however, there was
not so Significant a gap between the rhetoric of the insulation ideal
and what it was medically possible for a community to do. It was
possible to give lip service to the ideal, and charge social failure to
live up to it to collective moral shortfall. But now - when technology
continues to produce more and more ways to spend great sums on
medical care - it is self-evidently preposterous that a community
should treat health as lexicographically prior to all other values.
Any community that really tried to do so would secure for its citizens
marginally longer lives, perhaps, but these would be lives barely
worth living. Once, however, this suggestion of the ancient ideal is
rejected as incredible, the ideal has nothing more to say. It has, as
it were, no second best or fall-back level of advice. It simply falls
silent.
 Justice itl the Distribution of Health Care 207

In fact, as a result, philosophers, theorists and medical specialists

who nominally subscribe to the ideal of insulation all despair of
attacking the first question. After some discussion, they announce
that the size of the overall medical budget will be 'decided in poli­
tics', which is an academic way of saying that abstract considerations
of justice have nothing much to contribute to this part of the health­
care discussion. I believe that that is a mistake; if I am right, then
the dominance of the insulation ideal has been a hindrance, and not
just not a positive contribution, to achieving justice in health care.
Now look at the second question, the question of distribution.
When the theorists finish saying that politiCS will set the overall
health-care budget, they quickly add that justice will require that
that budget, whatever it is, be spent in a fair way. But how does
the insulation ideal help us to define a fair distribution? It tells us
something negative and undoubtedly important: that how some­
one is medically treated should not depend, in our SOCiety, simply
on ability to pay. It tells us that if rationing is necessary, the prin­
ciple of rationing should not be, as it now largely is in the United
States, the pocket book. But we need more positive advice. What
should the principle of rationing be, if it is not to be money? Once
again, the ancient ideal has very little to say. The egalitarian im­
pulse of the ideal seems to recommend that medical care be distributed
according to some principle of efficacy and need. And so people
committed to the ideal speak about rationing according to cost­
effectiveness or according to some principle that requires money to
be spent where it will do the most good.
As many of you know, the state of Oregon established a commis­
sion some time ago to try to give structure to that idea, to try to
describe what rationing health care in accordance with effectiveness
would mean. The difficulty, of course, as that commission discovered,
is that the concept of doing the 'most good' (or, in more academic
terms, of maximizing welfare, or utility, or well-being, or happi­
ness, or capability) is systematically and multiply ambiguous. These
various terms, when properly used, do not name psychological con­
cepts. Or medical or, in my view, economic ones. They name contested
ethical concepts: the proposal that health care money should be
spent to do the most good means that it should be distributed in
whatever way will make the lives of citizens better lives to have
lived, and that goal cannot be restated, without controversy, as the
goal of making lives more pleasant, or economically more produc­
tive, or socially more beneficial. Whenever you attempt to describe
208 ROllald Dworkin

in more detail what making the lives of citizens better actually

means, you enter the kind of controversy that it was the promise
and hope of the insulated ideal to avoid, and it would be sheer
disaster to try to reduce that ideal to something mechanical enough
to be measured by a computer. The Oregon commission discovered
this. It developed mechanical measures of the cost-effectiveness of
various sets of treatment matched to various kinds of disease, typed
these metrics into a computer, typed in a great deal of further in­
formation, and watched the computer produce a ranking of
cost-effectiveness that ranked capping a tooth higher in social pri­
ority than appendectomy. It's perfectly true, the computer said, that
you will die if you have appendicitis and don't have your appen­
dix removed. But it costs four or five thousand dollars to do that
and dentists can cap a great many teeth and prevent a massive
amount of toothache if you spend that five thousand dollars on
dentistry instead. Well, of course, as soon as that result appeared,
the commission saw that its algorithms were hopeless, re-rigged its
operational definitions, and produced something at least less im­
plausible than that. But the story indicates the character of the
problem I have in mind.
So the old ideal of insulation fails to answer our second ques­
tion as well as our first. Its proposal, that health care should be
distributed according to need, or so as to do the most good, or so
as to improve overall welfare, is fatally ambiguous, and becomes
evidently unattractive when the ambiguity is resolved by defining
success in terms of some utilitarian reading. We have not, after all,
settled the question of what justice in health care means, and that
philosophical problem stands beside the economic, medical and
administrative problems we know we face, and it may be at least
equally daunting.

III

This evening I shall try to construct, at least in very broad outline,

an alternate approach to justice in health care, which is based not
on the insulation of health care as a separate sphere of justice or
activity, but, on the contrary, on the integration of health care into
competition with other goods. I shall describe an approach that, I
believe, is more instructive about the two great issues of justice I
named. I can state the central idea in advance: we should aim to
make collective, social decisions about the quantity and distribution
 Justice in tile Distribution of" Health Care 209

of health care so as to match, as closely as possible, the decisions

that people in the community would make for themselves, one by
one, in the appropriate circumstances, if they were looking from
youth down the course of their lives and trying to decide what risks
were worth running in return for not running other kinds of risks.
At some point (as those of you who have read any political philos­
ophy written after the Middle Ages know) an imaginary story gets
told. My story has the virtue of being less imaginary than some
others, but it will nevertheless require you to exercise your imagi­
nation. Suppose that your community were to develop and change
in the following three ways.
First, per impussible, suppose it developed into a society in which
the economic system provided 'fair equality' in the distribution of
resources. I mean that government recognized its inevitable respon­
sibility to choose amongst economic and tax structures, and chose
a structure that treated all members of the society with equal con­
cern. I have my own idea about what that means in practice, and
I've tried to spell this out in a series of articles,! I said (this is a
very crude summary) that an economic structure treats all mem­
bers of the community with equal concern when it divides resources
equally, measured by the opportunity costs of each person owning
a particular resource, and then leaves each member free in prin­
ciple to spend those resources designing a life that each believes
appropriate. That conception of equality will not make people equal
in the amount of money or goods each has at any particular time;
still less will it mean that everyone will lead the same kind of life.
Some people will have invested and some people will have con­
sumed. Some will have spent early and some will have saved for
late. The result will nevertheless be egalitarian, because the choices
people will have made will answer to their own conceptions of
what life is right for them.
These are my views about what a just economic system would be
like, but I offer it only by way of illustration. You may you prob­
ably do - have a different conception of what economic structures
genuinely treat all people with equal concern; if so, your view of
how that community would have changed, in order to meet my
first condition, will be different from mine. That does not matter
for the present exercise: I merely ask you 10 assume that it has
changed, in whatever way you think justice and equal concern require.
Second, imagine that your community is also different in that all
the information that might be called. roughly, state-of-the-art
210 Ronald Dworkin

knowledge about the value and cost and side-effects of particular

medical procedures - everything, in other words, that very good
doctors know - is known generally by the public at large as well.
Third, imagine that no one in your community including in­
surance companies - has any information available about the
antecedent probability of any particular person contracting any par­
ticular disease or infirmity that he or she does not evidently already
have. No one would be in a position to say, of himself or anyone
else, that that person is more or less likely to contract sickle-cell
anemia, or diabetes, or to be the victim of violence in the street,
than any other person. So no information exists about how likely
it is that young blacks, as distinguished from people generally, will
die in violent fights, for example. 2
The changes I am asking you 10 imagine in your community are
heroic. But they are not, r think, beyond the reach of imagining,
and 1 am not inviting you to imagine other changes. Indeed, I am
asking you not to: I want you to assume that your preferences and
ethical convictions, and those of other members of your commu­
nity, have remained constant in spite of these changes. Very well.
Suppose that your community is indeed changed in those three
ways, and then also suppose that health care is simply left to indi­
vidual market decisions in as free a market as we can imagine.
Medical treatment is not provided by the government for anyone,
as it is for everyone here and for some people in America. Nor are
there any government subsidies for health care in particular, the
premiums people pay for health care insurance are not, as they are
now in the United States, tax-deductible. If people choose to pur­
chase such insurance, they do so as they buy anything else: out of
post-tax funds.
What kind of health care arrangements would develop in such a
community? How much of its aggregate resources would end up
devoted to health care? How would medical treatment in fact be
distributed among its members? Well, of course, it is hard to say;
indeed it is impossible to say with any precision, though I shall
offer you some speculations in a moment. But I'm anxious to make
two claims in advance of any such speculation, to show you why
the question of what such a society would do is important. The
first is that whatever the society I've just described spends as its
total health-care budget, which means simply the aggregate of what
individuals spend, would be the just and appropriate expense for
that SOCiety. The second is that however health care is distributed
 Justice in the Distribution of Health Care 211

in that society would be a just distribution of health care for that

society. 1 must qualify those two dramatic claims to some degree,
but the qualifications I need are not major, and I'll relegate them
to a note. 3 So I shall proceed on the flat assumptions just stated:
nothing that society does, by way of health care arrangements, is
open to objection on grounds of justice, though of course what it
does might be questionable or objectionable in many other ways.
If so, then it is indeed important to consider what health care
arrangements our society would make if it were changed in the
ways r described, because, as I shall argue, what they would do
through independent decisions can serve as a guide to what we
should do, in whatever way we can, to improve justice in our own
circumstances. So speculation seems worthwhile. It seems likely that
even though the members of the imagined community - our com­
munity transformed - would perhaps begin by making individual
insurance decisions, they would soon develop, through these indi­
vidual decisions, collective institutions and arrangements; it also
seems likely that progressively more and more people would join
those collective arrangements. They would develop very large co­
operative insurance plans, or very large health maintenance organ­
izations which provide stipulated categories of medical care for a
stipulated advance contract price, or both, for example. As such
plans became larger, and more efficient, it would become progres­
sively more and more expensive, relatively, for people to make wholly
personalized, individual medical arrangements for themselves, and
progressively fewer and fewer people would do so. (Remember that
in this society wealth is much more equally distributed than it is
in our society now, and though some people are relatively rich
there, they are mainly people who have decided to concentrate on
saving.) So the number of people who could and would turn their
back on the economies of scale and administration of the collec­
tive provisions will be few, and, as the process continues, fewer
still. The result of the process might very well be something func­
tionally very close to the single, comprehensive health care provi­
sion scheme that you have reached here in Canada. Large insurance
cooperatives or health maintenance organizations might negotiate
a basic scheme of provision that would be much the same for every­
one. If so, however, the community would probably also develop a
secondary insurance market: people would be free to negotiate
specialized insurance in addition to that basic insurance package.
What form that secondary market would take, and how large a
212 Ronald Dworkin

market it would be, would, of course, depend on factors we cannot

sensibly predict. But even in a much more egalitarian society, some
people would be able and willing to make provision for queue­
jumping, or elective cosmetic surgery, or other benefits that the
basic provision made available through general collective schemes
would not provide. (In a more egalitarian society, the cost of some
of these special benefits might well be lower than it is now - since
doctors' salaries, for example, would presumably be lower, special­
ized services might be available at lower cost.)
We need not dwell on the character of that secondary market: it
is more important to consider the basic, standardized coverage pack­
ages which j'm assuming that the large cooperative institutions would
provide. What would be the character of those packages? Well, of
course, that would depend upon the mix of preferences and con­
victions. But we can speculate with some confidence about what
would not be covered in such a plan. Some private insurance deci­
sions would be plainly irrational in the imagined community: they
would be what the economists might call dominant mistakes, by
which I mean they would be mistakes, even in retrospect, no mat­
ter what happened in the future, including the worst. I'll give you
one or two examples: they are extreme, but of course they would
be, given the claim I've just made about them.
Almost no one would purchase insurance that would provide life­
sustaining equipment once he had fallen into a persistent vegetative
state, for example. That would be a dominant mistake: the sub­
stantial sum spent year-by-year in insurance premiums to provide
that coverage would be at the expense of training or experience or
culture or investment or jobs that would have enhanced real life.
Even someone who lived only a few months after purchasing the
insurance before he fell into a vegetative state would have made,
in retrospect, a mistake, giving up resources that could have made
his short remaining conscious life better to buy a longer uncon­
sciousness. My second suggestion might seem more controversial. I
suggest that almost no one would purchase insurance providing
for expensive medical intervention, even of a life-saving character,
after he entered the late stages of Alzheimer's disease or other forms
of irreversible dementia. Almost everyone would regard that deci­
sion, too, as a dominant mistake, because the money spent on
premiums for such insurance would have been better spent, no matter
what happens, making life before dementia - life in earnest - more
worth while.
 Justice in the Distribution of Health Care 213

Now I come to a further suggestion, more controversial still. In

most developed countries, a major fraction of medical expense in
the United States it approaches forty percent of the health care
budget is spent on people in the last six months of their lives.
Of course, doctors don't always know whether a particular patient
will die within a few months no matter how much is spent on his
care. But in many cases, sadly, they can say, with considerable con­
fidence, that he will. I believe that if people reflected on the value
of buying insurance that would keep them alive, by heroic medical
intervention, four or five more months, in the condition in which
most such patients undergoing that intervention live, compared with
the value the premiums necessary to purchase that insurance could
add to their earlier lives if spent in other ways, they would decide
that buying that kind of insurance was not a wise investment. That
is not to say, of course, that most people would not want those
additional months, no matter in what state or condition they spent
them. Many people want to remain alive as long as possible, pro­
vided they remain conscious and alert, and provided the pain is
not too great. My point is rather that they would not want those
additional months at the cost of the sacrifices in their earlier, vig­
orous life that would be necessary if they had to make that choice.
They would think the money better spent, earlier, on job-training
or education or investment or on something else that would ben­
efit their lives as a whole more than just taking on a few months
of very limited life at the end. I cannot quite make the claim here
that I made about persistent vegetative state or advanced-stage
Alzheimer's disease: that purchasing insurance for costly procedures
extending life a few months would be a dominant mistake. We can
imagine circumstances someone falls fatally ill the day after buy­
ing a policy providing for such care in which, in retrospect, the
decision to buy it turned out to be a good one. But most people
would agree, I think, that in the circumstances we are imagining
in which, remember, no one knows he is more likely than anyone
else to contract a disease not already evident that decision would
be an antecedent mistake.
How much further can we go down this road? How much more
insurance can we be reasonably confident people would not buy in
the circumstances we are imagining? I'm not sure, and anyway have
no time to explore other examples now. But I do want to raise, at
least, one further issue which, as I suggested to you earlier, is already
of crucial importance and will become even more critical in the
214 Ronald Dworkin

next decades. How far would people in the imagined community

make provision for access to the ultra-expensive medical equipment
now in use, or which is being developed, or is still over the horizon?
I came here from attending a meeting at the Harvard Medical
School in which new advances in technology were being described.
You ain't see nothin' yet. I've already mentioned advances in diag­
nostic radiology: expensive magnetic resonance imaging, for example.
Much of the talk at this meeting was about molecular biology: about,
for example, promising research into treating cancer by creating
monoclonal antibodies specific for each patient, from the patient's
own genetic material, at stupendous cost, and new, very expensive,
blood tests that marginally very marginally improve the accu­
racy of a diagnosis of heart disease. Each of these examples illustrates,
though in different ways, how technology might come to be re­
garded as 'Iow yield' relative to its large cost. Both would undoubtedly
save some lives. But at a cost, in development and production, that
might seem very high when we consider how a community might
use the funds in other ways that would enhance the economy and
provide more jobs and a higher standard of living for more people.
Would people in the imagined society, ultimately deciding for
themselves how to allocate their resources, provide for expensive
and/or speculative technology? People informed and reflective might
make distinctions along the following lines. They might pay to provide
life-saving techniques for diseases that tend to occur relatively early
in life, particularly when these techniques have a high probability
of success. But they might not spend to insure for technology that
is very speculative, even though it will save some lives, or for tech­
nology whose main results benefit people in relatively old age. Paying
all our lives to secure the latter kind of technology, if we need it,
might seem a poor decision when it means that we run a higher
risk than we need to run of unemployment or an otherwise less
satisfactory life. I won't pursue these speculations further. I hope
I've given you some idea of the kind of choices that people in the
conditions we're imagining would have to make, and of how they
might be tempted to make them.

IV
Do you resist my claim that whatever such a society spent, through
collective institutions governed by individual decisions of this
character, would be just, and that the distribution of health care
 Justice in the Distribution of Health Care 215

such a society achieved would also be just? You will not, if you
accept a conception of social justice that assigns individuals re­
sponsibility for making the ethical choices for their own lives against
a background of competent information and a fair initial distribu­
tion of resources. If you accept that vision of a just society, then
you will accept my claim though, as I said, you may well have a
different conception of what a fair initial distribution of resources
would be like, and how unjustified inequalities should be remedied,
than I do - in which case your understanding of the conditions I
described will be correspondingly different from mine.
So I will assume that you do agree with my main claim: that
whatever our imagined society achieves, by way of health care
arrangements, cannot be faulted on grounds of justice. I suggested,
earlier, that we might therefore make practical, political use, for
our own communities, of at least our less speculative conclusions
about what people in the imaginary community would provide for
themselves. There is a natural way in which we might be tempted
to do this. Almost all government-sponsored or supervised health
schemes now in existence, and almost all of those that have been
proposed as vehicles of reform in the United States, define a basic
health-care package of benefits that must be made available, at re­
sponsible cost, to everyone, and supplied without charge to those
who cannot pay that responsible cost themselves. We might use
our speculations about the imaginary society to help us to define
what should be in that basic package, and what that responsible
cost should be.
In one way, at least, the imaginary story might be helpful for
countries, like the United States, who have not settled on a par­
ticular structure for health care reform. As I said, many people in
America believe we should follow your example in constructing a
single-payer arrangement in which government, not private insur­
ance firms or health care proViders, decides what medicine to offer
and at what price. But others think the United States should adopt
what is called a scheme of 'managed competition', in which private
insurers compete to offer a basic package stipulated by government,
and government supervises their performances and premium struc­
tures. As of this evening, at least, most commentators predict that
a managed competition scheme will be adopted, primarily, they
say, because it is better suited to the political culture of the United
States than a single-payer scheme would be. But our imaginary story
might be helpful in guiding the choice between the two forms of
216 Ranald Dworkin

scheme, in the following way. The decision might turn, among other
things including suitability to the political culture, on the degree
of confidence we have in our speculations about what people would
choose in the imaginary community. If we were reasonably confi­
dent that we knew roughly what such people would buy - what
the dominant collective arrangements they would reach would provide
then that would argue for trying to set in place a single-payer
system like yours or like the National Health Service in Britain.
Government can more effectively guarantee people what it is per­
suaded justice demands that they have if it is free to provide it
itself, in some such way. To the degree we are uncertain about what
people in the imaginary world would decide, however, that argues
for a scheme of managed competition with enough flexibility to
allow different people to choose different packages all meeting a
common stipulated standard. The choices that actual people make
among such schemes would provide a self-regulating mechanism
that would bring us closer to the just distribution of the imagined
world.
But of course whether the United States ultimately chooses a single­
payer scheme like yours or, as seems more likely, a scheme that
includes private competition, is more likely to depend on consider­
ations other than justice. Nor, I think, is justice decisive of that
issue one way or the other. Both types of scheme include the idea
of a basic package (or set of such packages) of insurance made avail­
able to all, and the main issues of justice consolidate in the question
of what should be in that basic package or set of packages. That is
the question, as I suggested, that is most directly responsive to the
exercise I've been imagining. I offered you reasons for thinking that
certain kinds of insurance or health organization contractual pro­
vision would be rare in the community we imagined, and that, I
now submit, is a good reason why that kind of provision should
not be part of the basic package that will be the heart of any re­
form in the United States and any readjustments here in Canada.
Since those are expensive provisions, this is an important result.
But it is a negative one, and the exercise must be conducted on
the other side as well. I have little doubt that people in the imag­
ined community would insist on provision for standard prenatal
care, for example, and on the kinds of primary medical care, in­
cluding relatively inexpensive routine examinations and inoculations
that poor people in the United States so conspicuously lack. It fol­
lows, from the argument I have been making, that these are essential
 Justice in the Distribution of Health Care 217

elements in the basic package that any responsible health care re­
form would establish.

v
I have been exploring ways in which practical health care adminis­
tration and reform could be guided by the exercise I hold out: trying
to imagine what health care people in the imaginary circumstances
I described would provide, out of their own pockets, for themselves.
It is past time, however, for me to consider the drawbacks and
pitfalls of my overall argument. One danger is evident: my sugges­
tions about how people would behave in the imaginary society are
speculative, and even though some of these speculations seem very
plausible, we cannot test them by asking how everyone actually
behaves in communities as they are now constituted. Resources are
unjustly distributed among us: Canada is not as bad in this respect
as is the United States, but even Canada is very far from ideal
justice in economic distribution. We obviously don't have a society
in which people enjoy state-of-the-art information about medicine.
On the contrary, people's medical ignorance is often cited as one
reason why medical expenses continue to rise. And, of course, our
insurance companies do know that risks are higher among certain
groups within the community than others, and the curse of expe­
riential rather than community rating for premiums has dogged
attempts to make commercial medical insurance fair.
But it doesn't follow that our speculation about what people would
want under very different, and fairer, circumstances must remain
just speculation. The choices Americans of average income make
about their employee insurance package in wage negotiations, for
example, can offer some guidance. And research and publicity can
provide better guidance. Not only government but private organi­
zations - large medical schools, for example - could help design a
few sample paradigm insurance protocols representing different in­
surance strategies. Some of these would provide for catastrophic
care or transplant surgery in circumstances in which others denied
it, for example. The protocols could be accompanied by the medi­
cal information of the kind that is crucially missing from public
awareness now: by some realistic expert opinion of the expected
consequences for mortality and morbidity from a public commit­
ment to each protocol, together with, for each, some estimate of
its total cost and consequent macro-economic effect. If information
218 Ronald Dworkin

of that sort were put into the public domain, and challenged and
debated there, the resulting discussion would be at least minimally
informative about how much people value what kind of care, and
might be very informative. When we think of the kind of opinions
that pollsters examine now, and that feature on television discus­
sion shows and radio phone-ins, we might welcome a shift to the
kind of discussion I'm now imagining.
A second difficulty is potentially much graver, however, at least
conceptually. I've imagined a utopian (in some respects) society
and I've then suggested that we set out to copy one feature of that
utopian society: the provision it would make for medical care. An
economist will remind me that, when the first best is impossible,
the second best is not always achieved by mimicking the first best
partially. That may, indeed, make matters worse than the status quo,
and it is not difficult to see this possibility as a threat to my argu­
ment. Suppose, for example, that we decide that if our community
were just, and different in the other ways I imagined, the standard
medical package nearly everyone would purchase through collec­
tive insurance arrangements would include a particular set of benefits.
If we decide, therefore, that that set of benefits should make up
the basic package that must be made available to everyone in our
own community, some relatively low-income people may end up
paying a higher share of their actual income for medical care (for
themselves and, through taxes, for others) than they would have
chosen to pay in a just society. Or, to put the matter the other way
around, they may have less left over for other expenses than they
would have chosen in those circumstances. That may not seem,
particularly to them, an improvement in justice.
I do not want to minimize the problem this hypothetical example
illustrates. But the possibility that the test of justice I propose might
produce unjust results is not, in itself, a sufficient argument against
accepting that test; someone who objects must show a strong like­
lihood that the result would in (act be worse, from the pOint of
view of justice, than using some other defensible standard for de­
signing the basic package of protection. This is not a question of
who has the burden of proof. If it is true that if our economic
structure were just, everyone would be able to and nearly everyone
would purchase a particular medical provision, that supplies a very
strong even if not decisive argument that our structure would be
closer to a just one if we made sure that everyone had that provision
now. We should act on that strong argument unless we have some
 Justice in the Distribution of Health Care 219

positive reason, not just the bare possibility, that it is mistaken.

It is true, however, as the example I just gave demonstrates, that
the new model of health care provision and distribution I am pro­
posing will work more dependably as the community's tax system
grows more just. If relatively low-paid workers pay much more than
their fair share of taxes, because the tax structure is insufficiently
progressive to be fair, then any governmental program that relies
on the redistribution of tax proceeds to improve justice for those
with scarcely any income at all will be compromised for that rea­
son. It will involve an unjustified transfer to the worst-off group
from the almost-worst-off-group. That reflection provides a strong
reason why tax reform must be at the centre of any general cam­
paign to improve social justice. It would be ironic and disappointing,
however, if the point were stood on its head, and if those who
resisted redistribution to the very poor were able to pOint to im­
perfections in the tax structure as justification for doing nothing,
and retaining their own privileges under the status quo.
We must next consider a very different kind of issue, which I
must not evade, though my views on that issue, I fear, will disap­
point many of you. Suppose that everything I've been describing
as possible came to pass. Suppose that, after the right kinds of col­
lective consultation, after meetings and discussions and polls and
electronic politics and all the rest, we settled on a particular basic
program of medical care that we collectively thought government
should, in one way or another, make available to everyone. That
basic package, as my earlier argument suggests, will not include
some treatment that rich people are now in a position to buy for
themselves. I said earlier, for example, that the test r proposed would
very likely rule out ultra-expensive marginal diagnostics or extra­
ordinarily costly treatments that have some but very little prospects
for success. Some people in Canada and America now have the
money to buy health care that would be excluded from the basic
package. They have the money to buy a liver transplant when the
odds are very small but nevertheless real that the procedure would
save their lives. In England, people are standardly denied even re­
nal dialysis on the National Health Service when they are sixty-five.
So people of that age die if they cannot afford to pay for dialysis
themselves.
If we adopted the kind of scheme that I'm describing, in our
admittedly imperfect society, and took no steps to forbid people
buying more expensive care than the basic package provides, some
220 Ronald Dworkin

people would have better medical care - some people would live
longer and healthier lives - only because they had more money. In
most cases, since the basic economic structure would continue to
be unjust, because they unjustly had more money. Should we therefore
take steps to prohibit or constrain the private market in medicine?
Should that be part of any respectable campaign to improve justice
in health care? Of course, we couldn't actually abolish the private
market in health care altogether: we would end by producing back
street dialysis. But should we do what we can, aiming to prevent
anyone from buying better medical care than the basic package
provides, so far as that is possible?
The insulation model of medical justice I began by describing, if
taken seriously, would insist that we should, and I believe many
people here this evening would agree. Solidarity is compromised,
they think, when some people can live while others die only be­
cause the former have more money. That seems to me the wrong
answer, however. The spirit of the argument I have been making
suggests that no one can complain on grounds of justice that he has
less of something that someone else does, so long as he has all he
would have if society were overall just. And, of course, in the cir­
cumstances we are now considering, people whose basic provision
does not include liver transplants, and cannot afford to buy such
an operation for themselves, are by hypothesis not denied what
they would have if economic justice were perfect.
Some of you will hate that argument, as I said: you will think it
intolerable. May I remind you, however, that the hypothetical in­
equality in medical care I'm now considering is, in one important
respect, relatively benign compared to other inequalities in our SOCiety.
If health care were rationed in the way we are contemplating, then
everyone would have at least the medical care he would have in a
just society, and that would not be true in most other departments
of resource allocation. In education, employment, culture, recreation,
travel, experience and a host of other goods and opportunities that
for most of us make up the value of being alive, the poor would
continue to have much less than they would if we had reformed
not just health care but our economic and social life more gener­
ally. If we somehow manage to succeed in providing the poor with
the medical care that justice requires, it would be perverse, given
that a rich man can spend on more comfortable housing or better
education for his children, not to alIm¥ him to spend on more
expensive health care. We would do better to put an excise tax on
 Justice in the Distribution of Health Care 221

special health care, and use the proceeds of that excise tax to im­
prove public education, or the economic infrastructure, or to reduce
public debt that blights employment prospects, or in some other
ways that would make the community distinctly more egalitarian.

VI

I will offer you no final summary of my somewhat discursive re­

marks; I shall try, instead, to broaden the argument in closing it. I
began by criticizing the insulation model, as I called it, and you
may think I've been undermining that criticism in the last part of
the lecture. I've been arguing how we might make our communi­
ties better in just one respect, and that goal seems to assume, with
the insulation model, that health care is special, 'chief' among goods.
But my special interest in medical care is largely practicaL Medi­
cine is now a problem for people so high up in the economic scale
- well up into the middle, fat part of the economic diamond where
the votes are. People generally, not just the poor, agree that govern­
ment should take a larger role in structuring, controlling and financing
the provision of health care. We can seize on this opportunity to
make the distribution of health care more just as well as more efficient.
But if America does make new progress in that direction, as Canada
already has, then the lesson might be of more general political
importance. For one thing, it might teach us that the bad press the
ideal of equality has had for some time is unjustified. There is a
rap against equality: that accepting equality as an ideal, even one
among others, means levelling down and requiring everyone to live
the same kind of life. But the conception of equality I've been
relying on has quite the opposite character: it is dynamic and sen­
sitive to people's differing convictions about how to live.
I end with this further observation: the question of health-care
reform in America, including politically acceptable and fair health­
care rationing, is ideologically leveraged. If we fmd, after all the
fuss, that politically we can't do much to make the distribution of
medical care more just, in spite of the apparent present opportuni­
ties to do so, then a pessimistic conclusion may be irresistible: we
may abandon hope for any more widespread or general democratic
concern for social justice. But if we do now make substantial and
recognizable political progress in this one urgent matter, we may
learn more, from the experience, about what justice itself is like,
and we might find it to our taste, so that we can steadily, bit by
222 Ronald Dworkin

bit, incrementally, fight the same battle in other areas. So the war
against injustice in medicine that you have been fighting so well
here, and that we are about to take on in America. is indeed a
crucial one. Health might not be more important than anything
else but the fight for justice in health might well be.

Notes
The central article, for purposes of this lecture, is 'What is Equality?
Part I1', which appeared in Philosophy & Public Affairs, Fall 1981.
2 I am ignoring an important issue that I will have to consider in a sub­
sequent full presentation of this material. Is it right, in the hypothetical
exercise I am constructing, to exclude information relating risk of dis­
ease to voluntarily chosen behaviour? Should insurance companies be
in a position to charge cigarette smokers or mountain climbers higher
premiums, for example'? If so, then what counts as voluntary behaviour?
Should sexual behaviour of a particular kind be treated as voluntary for
this purpose? Should insurance companies be able to charge active male
homosexuals higher premiums because they are more likely to contract
Aids?
3 Some paternalistic interference with individual decisions about health
care insurance, particularly those people make early in their lives, might
be necessary out of fairness to people who might make imprudent in­
surance decisions when young. And some constraints and requirements
might be necessary in the interests of justice toward later generations.