GROUPE

 D’ANALYSE  ET  DE  THÉORIE  ÉCONOMIQUE    LYON  -­‐  ST  ÉTIENNE      

Documents  de  travail  |  Working  Papers  

 

 
 

 
WP  1   333  
 
 
   
Welfare  economics  
 
 
 
Antoinette  Baujard    
 
   
November  2013  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 

Electronic copy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2357412
http://ssrn.com/abstract=2357412
  
 
 
 
 
 
 
 
 
 
 
 
 
 
 
 
GATE  Groupe  d’Analyse  et  de  Théorie  Économique  Lyon-­‐St  Étienne  
 
93,  chemin  des  Mouilles    69130  Ecully  –  France  
Tel.  +33  (0)4  72  86  60  60    
Fax  +33  (0)4  72  86  60  90  
 
6,  rue  Basse  des  Rives  42023  Saint-­‐Etienne  cedex  02  –  France    
Tel.    +33  (0)4  77  42  19  60  
Fax.  +33  (0)4  77  42  19  50  
 
Messagerie  électronique  /  Email  :    [email protected]  
Téléchargement  /  Download  :  http://www.gate.cnrs.fr    –  Publications  /  Working  Papers  
 
 

Electronic copy
Electronic copy available
available at:
at:https://ssrn.com/abstract=2357412
http://ssrn.com/abstract=2357412
 Welfare economics1
Antoinette Baujard2

November 8th, 2013

Abstract. This paper presents the Paretian Watershed

and the fundamental theorems of welfare economics. It
distinguishes the British approach (à la Kaldor-Hicks)
from the American approach (à la Bergson-Samuelson)
to new welfare economics. It develops the more recent
domains of happiness economics, the comparative
approach by Amartya Sen, and the theory of fair
allocation by Marc Fleurbaey.

JEL Code: B2, D6

Key-words: Welfare economics, Kaldor-Hicks, Social
Welfare Function, Pareto, comparative approach,
happiness economics, fair allocation

Welfare economics is the economic study of the definition and

the measure of the social welfare; it offers the theoretical
framework used in public economics to help collective decision
making, to design public policies, and to make social evaluations.
Questions usually tackled by welfare economics are the following:
What is social welfare? Is there a reliable and satisfying way to
measure it? If social welfare is based on individual preferences,
can we derive a social preference from the preferences of
individuals? Are competitive equilibrium outcomes optimal in the
sense that they lead to the highest social welfare? Can any optimal
outcome be achieved by a modified market mechanism? Can we
really formulate recommendations for public policies on the basis
of such welfare analyses?

1
To be published in Gilbert Faccarello and Heinz Kurz Eds. Handbook of the
History of Economic Analysis, Vol.3: Developments in Major Fields of
Economics. Edward Elgar Publishing Limited, Cheltenham (U.K.)
2
GATE L-SE, Université de Lyon, Lyon, F-69007 France; CNRS, GATE Lyon
Saint-Etienne, Ecully, F-69130, France; and Université Jean Monnet, Saint-
Etienne, F-42000, France. Contact: [email protected], (33)
+4 77 42 13 61.
1

Electronic copy available at: https://ssrn.com/abstract=2357412

 In spite of the uncontroversial importance of all these issues,
some have been overshadowed while others have drawn enormous
attention. From then on, the death of welfare economics has been
often foretold (Hicks 1939a: 697; Chipman and Moore 1978: 548;
Mishan 1981; Hausman and MacPherson 1996: 96). Setting out the
history of welfare economics implies firstly to recall its evolution,
secondly to discuss the reasons why it has almost missed its project
—among others the role of interpersonal comparisons of utility, the
subjective interpretation of utility, and the rejection of value
judgment out of economics. Thirdly, we shall claim there are strong
reasons to hope: welfare economics is back (Sen 1999a; Fleurbaey
and Mongin 2005), yet at the cost of accepting the normative nature
of (welfare) economics.
The pre-history of welfare economics is as old as political
economics: classical and neo-classical economists were studying
the efficiency and equity of productive systems, more specifically
wondering how to value commodities or labor, and to assess the
best allocation of goods and of tasks for the society (Myint 1965).
Utilitarianism which, since Bentham, aimed at providing tools to
measure and improve individual and collective well-being, may be
considered as the genuine root of welfare economics. From then on,
the evolution of welfare economics marks up different periods and
types of contributions. Following here Philippe Mongin
(2002c,2006b), its history may be divided in at least four
successive stages. First stage, the creation of the first tool of
welfare economics goes back to Marshall —or even before in the
works of Jules Arsène Dupuit—: the introduction of the notion of
consumer surplus is meant to provide a method to measure relative
change in consumers’ utility. They may derive some policy
recommendations from the surplus analyses. But it was more
clearly born with Arthur Cecil Pigou’s book published in 1920, The
Economics of Welfare3 in which he has among others developed the
famous distinction between private and social marginal cost or
productivity4, the role of the size and the distribution of the national

3
Notice the first version of this book, Wealth and welfare, was published back in
1912.
4
A private marginal cost is the marginal cost borne by the individual who
decided the change, which is induced by the infinitesimal growth in the use of
one input. The social marginal cost is that which is borne by the whole
population. Notice externalities emerge in case of a difference between both
measures.
2

Electronic copy available at: https://ssrn.com/abstract=2357412

dividend in measuring economic welfare5, and his defense of the
transfer principle6. The definition of welfare was not really unified
at this stage: it could be the ‘national dividend’ or a mix between
the amount of the dividend and the distribution of income, and even
something else. Second stage, the new welfare economics
established a clear separation between the optimality conditions
based on the Paretian condition and their applications to the
market. The definition of welfare was uniformly based on strictly
ordinal and subjective individual utilities. The best-known
applications are the fundamental theorems of welfare economics.
The question of income distribution, including when applying the
principle of compensation, was then mostly left aside. Third stage,
after the arrovian negative result tolled the bell knell in the fifties,
social choice theory, public economics and the theories of
inequality and poverty have been kept separate for decades. The
only noteworthy element of continuity and unity is that most
contributions were then welfarist, that is to say that the only
relevant information for social welfare or public decision was
individual utilities. Fourth stage, some post-welfarist economic
theories of justice or fairness have been recently developed. Some
economists suggest redirecting their research for example to
analyze rights, or to integrate information such as talents and
handicaps, opportunities and capabilities among others.
This chapter is organized as followed. The Paretian watershed
exposed in the next section marks the evolution from the first to the
second stage of welfare economics and the formulation of the two
fundamental theorems of welfare economics. The problems raised
with both approaches of the new welfare economics described in
section 2 provide some clues to understand the disintegration of the
third stage. Recent and promising avenues for researches are
developed in section 3. Section 4 concludes.

5
There are some connections between the national dividend and our gross
national product, standardly used nowadays to assess and compare social states.
National dividend basically distinguishes from national income by its specific
focus on the actual overall consumption rather than on raw production.
6
According to the Pigou-Dalton transfer principle, a distribution of income is less
unequal when the rich become less rich and the poor become less poor, when the
national dividend remains equal. This can be obtained by progressive transfers
from the rich persons to the poor persons. On certain conditions, economic
welfare may increase when inequality decreases.
3

Electronic copy available at: https://ssrn.com/abstract=2357412

1 – The Paretian watershed
1.1 – The old and the New Welfare Economics
At the turn of the century, Vilfredo Pareto introduced the
concept of ophelimity in economics and, on the basis of scientific
criteria, encouraged to narrow the amount of information we could
derive from it: it should be an ordinal concept, and interpersonal
comparisons of ophelimity ought to be ruled out. Would the term
‘ophelimity’ not be retained afterwards, this watershed has yet been
confirmed by the publication in 1932 of Lionel Robbins’ famous
book, An essay on the nature and significance of Economic
Science, in which he disputed the meaning of interpersonal
comparisons of utility and the material definition of economics. As
far as a subjective account of utility holds, there exist no way,
whichever by introspection or by observation, to compare the
intensity of satisfactions of two different persons (Cooter and
Rappoport 1984). If assertions implying the meaning of cardinal
utility or of comparisons, such as the rule of decreasing marginal
utility7 are formulated, they necessarily derive from a value
judgment. Notwithstanding, if economics claims to remain a
science, hence to be value neutral, such comparisons should be
absolutely avoided.
Paul Anthony Samuelson (1947: 249) draws the consequences
of the ban of interpersonal comparisons of utility, as well as of the
restrictions of utility to the scientific theory of demand, by
distinguishing the old from the new welfare economics: “While in a
real sense there is only one all-inclusive welfare economics, which
reaches its most complete formulation in the writings of Bergson, it
is possible to distinguish between the New Welfare Economics [...]
which makes no assumptions concerning interpersonal
comparability of utility, and the Old Welfare Economics which
starts out with such assumptions."

1.2 – The Pareto Criterion

The only uncontroversial normative criterion at the collective
level for the New Welfare Economics relies on individual utilities,
as far as comparing utilities among individuals is not required nor
even allowed. According to Pareto (1906: 261), “the members of a

7
Remind that a consequence of the law of diminishing marginal utility may be
that giving one more Euro to a poor person than to a rich person is collectively
better.
4

Electronic copy available at: https://ssrn.com/abstract=2357412

collectivity enjoy maximum ophelimity in a certain position when
it is impossible to find a way of moving from that position very
slightly in such a manner that the ophelimity enjoyed by each of the
individuals of that collectivity increases or decreases. That is to
say, any small displacement in departing from that position
necessarily has the effect of increasing the ophelimity which
certain individuals enjoy, and decreasing that which others enjoy,
of being agreeable to some, and disagreeable to others.” A social
state is hence said Pareto optimal if it is not possible to improve
the situation of certain individuals without making the situation of
at least one other individual worse off.
Let us consider how to use this criterion. Compare different
social states for a given population, where everyone has
monotonous preferences over the commodities x and y. In state S1,
the allocation of resources among individuals is fully equal for each
commodity. Now, if individuals’ tastes are heterogeneous —some
prefer to have more x while others want relatively more y—, they
will find opportunities for exchange between x and y. In the social
state S2, the situation of individuals who saw an interest in the
exchange has improved while the situation of others did not
deteriorate from S1 to S2. S2 is better than S1 according to the
Pareto criterion as the situation of some has improved without
damaging the others’. Nobody has a vested interest to go back from
S2 to S1, and at most, some are indifferent. The fundamental
theorems of welfare economics characterize this optimum, and
specify the conditions of its existence (See subsection 1.3). The
choice among different Pareto-optimal equilibria, notably on the
basis of explicit value judgments, is the task devoted to the
Bergson-Samuelson version of welfare economics (See subsection
2.2).
Imagine now that, in state S3, resources entirely belong to a
single rich individual, while the others are totally deprived. The
Pareto criterion does not help to compare S1 and S3. May the
unequal distribution of S3 be repellant, the rich individual’s
satisfaction would drop from S1 to S3, which implies that at least
one person would suffer a downturn. As the Pareto criterion does
not apply, no ranking between S1 and S3, or S3 and S2 may be
derived. Hence a strict Paretian welfare economics is mute as to
whether or not public policies should go towards state S1 or S2
while in S3. More generally, it cannot disentangle situations in
which trade-offs among the satisfactions of different individuals are
required. However, most policies are likely to hurt some
individuals or groups of individuals in order to improve the
situation of another significant group of people. They imply trade-

Electronic copy available at: https://ssrn.com/abstract=2357412

offs at the end, hence they rely on some kinds of interpersonal
comparisons of utility. That is why, after Robbins’ attack against
the normative aspects of economics and especially against the use
of interpersonal comparisons, welfare economics was likely to
become silent for any policy recommendations and could have lost
it raison d’être. The British version of the Paretian welfare
economics provides some tricks to generate recommendations
without, so they claim, involving any value judgments (See
subsection 2.1). The new approaches to welfare economics such as
the capability approach (See subsection 3.2) and the equity theory
(See subsection 3.3) succeed in considering these situations and
formulate explicit normative criteria to justify the trade-offs.

1.3 – The fundamental theorems of welfare economics

The social optimum is well described through the fundamental
theorems of welfare economics, which formalize some ideas
already present in Pareto’s works (especially for the first theorem)
and in Walras’ (especially for the latter).
Oskar Lange, Abba Lerner, and Harold Hotelling have provided
the first order conditions for economic efficiency, and the primary
proofs of the first theorem. The problem of maximizing overall
welfare amounts to maximizing the utility of each individual under
the constraints of others’ utility, possible allocation and
transformation functions, which is up to three conditions. First,
individual utilities are maximized if the marginal rates of
substitution for two given commodity between two different
individuals are equal. Second, the aggregate output and the optimal
allocation of goods among individuals are obtained by equalizing
the marginal rates of substitution with the marginal rates of
transformation between the two given commodities. Finally, the
marginal rates of transformation of the different firms among any
two commodities must be equal to guarantee the efficiency of
production for the various technologies. Notice these results, now
rigorously established, resume some economic laws previously
discovered by the precursors of the marginalists, such as Hermann
Heinrich Gossen in 1854, and by the marginalists themselves, such
as William Stanley Jevons in 1871.
Kenneth Arrow (1951b), Gerard Debreu (1951), and then the
two together (Arrow and Debreu 1954) have generalized the proofs
and these results. In formal terms, they have overcome the use of
calculus, though intuitive to use for demonstrations related to
optimization problems, by now using set theory. They have shown,
with very few conditions, that the optimum more fundamentally
derives from the price system. That is how they have formulated
6

Electronic copy available at: https://ssrn.com/abstract=2357412

what is now called the two fundamental theorems of welfare
economics.
The first theorem of welfare economics states that competitive
equilibria are Pareto-optimal, if individual preferences are
monotonic and if there are complete markets.
The second fundamental theorem of welfare economics states
that one can achieve any Pareto-optimal allocation in a
competitive equilibrium when the social planner undertakes an
appropriate redistribution of endowments. Among several Pareto
optima, some are probably more satisfactory than others. The
theorem points out that the preferred social optimum can be
achieved by a competitive equilibrium if accompanied by proper
redistribution policy which shall establish the new ‘initial’
allocations. An important consequence of this theorem is that it is
not necessary to alter the competitive system to obtain Pareto
optimality. A trade-off between efficiency vs. equity is not any
more required; however, the issue of the redistribution is pregnant.

2 – The new welfare economics

Two types of approaches of the new welfare economics have
been developed in the 1930ies and the 1940ies, which we may call
the British approach on the one hand and the American approach
on the other hand.

2.1 – The British approach to the New Welfare Economics

As far as the only uncontroversial normative criterion is the
Pareto criterion, welfare economics establishes a clear test: a
situation is economically efficient if it could not be better for the
individuals without decreasing some people’s satisfaction, which
implies unanimity to justify any change. If it were nonetheless
confined to such unanimous improvements, its object would be far
too restrictive. The British approach, particularly represented by the
works of Nicholas Kaldor (1939), John Hicks (1941) and Tibor
Scitovsky (1941), essentially coming from the London School of
Economics, developped a new concept of Pareto improvements in
order to reach a decision and bypass the problem of comparisons.
They propose a ‘Pareto efficiency criterion’ which considers the
possibility of hypothetical compensations, and then applies the test
of unanimity. Because the compensations are just hypothetical,
they claim their consideration does not imply any value judgments.
Imagine a single individual i looses x by a new public policy,
while all others gain. The strict version of Pareto criterion cannot
conclude that this policy should be implemented. Imagine now that
7

Electronic copy available at: https://ssrn.com/abstract=2357412

others gain of an amount that is greater than x. Would the winners
compensate Mrs. i by transferring her the amount x, they would
still gain from the new policy, while Mrs. i would now be at most
indifferent. The change would be a Pareto-improvement, i.e. would
be unanimously better, if such compensation were made. In all
cases, this change passes the test of hypothetical compensations
and is considered to be “Pareto-efficient”, then could be
recommended. Economists are however not entitled to decide
whether or not these transfers should eventually be made; such
responsibility should be left to politicians on a second and distinct
stage. This division of tasks between the economist as a scientist
and the policymaker, as a politician, allows to comply with
Robbins’ contentions, yet to formulate public policy
recommendations. From then on, this general framework
rehabilitated surplus analyses and paved the way to the widespread
use of cost-benefit analysis.
Extremely serious and skeptic critics have been raised against
this approach by the leading experts in the field (Arrow 1963, Sen
1979d, Boadway and Bruce 1984 among others). Firstly, the
internal consistency of the model is challenged. Among others, this
welfare criteria “could not escape the possibility of giving rise to an
inconsistent sequence of policy recommendations, unless either the
distribution of income and wealth or the forms and degree of
dissimilarity of consumers’ preferences were assumed to be
suitably restricted” (Chipman and Moore 1978: 578). Secondly, the
normative aspects of this approach are strongly contested: even
though it pretends to avoid interpersonal comparisons of utilities, it
operates exactly on the basis of their existence (Cooter and
Rappoport 1984, Blackorby and Donaldson 1990). Yet it does
prevent any discussion of the value judgments involved in such
analysis. Thirdly, beyond the problem of aggregation, these tests
are more generally blamed because they are ‘welfarist’. A social
welfare evaluation is called welfarist when it relies on subjective
individual utilities only (Sen 1979a, 1979b). Amartya K. Sen and
many others have shown the logical, pragmatic and normative
limits of such account of individual welfare in the context of
designing or assessing public policies. In short, Chipman and
Moore concluded in 1978: “judged in relation to its basic objective
of enabling economists to make welfare prescriptions without
having to make value judgments and, in particular, interpersonal
comparisons of utility, the New Welfare Economics must be
considered a failure." In spite of such an acknowledgement, the
success of this approach in occupying a leading position in most

Electronic copy available at: https://ssrn.com/abstract=2357412

contemporary works of public economics, industrial economics or
international economics remains today unchallenged.

2.2 – The American approach to the New Welfare Economics

What we shall call here ‘the American approach’ is associated
with the position of Abram Bergson, from the MIT, and Paul
Samuelson, from Harvard University, i.e. both coming from
Cambridge (Mass.) in the United States. Bergson formalized the
concept of social welfare in 1938 (Burk 1938). He defines it as a
function of all the elements relevant for welfare: all products,
consumer’s goods, the amount of work of each type, non-labor
factors, characteristics of the environment, etc. Through the
application of the Pareto criterion, the function may emphasize the
“fundamental value of individual preference." The social welfare
function, as eventually formulated by Samuelson (1947), is defined
as a function of the individual utility functions that each individual
derive from the social state. The shape of these functions captures
some value judgments that are explicitly formulated.
How can we legitimately decide which would be the right
social welfare function? And what does a "social preference" even
means? The question was notably asked by the logician Olaf
Helmer to Kenneth Arrow when both were working at the Rand
Corporation in 1949. Consistently, this function should rely on the
individuals’ views, yet without resorting to interpersonal
comparisons of utility. Arrow (1963) provides a first answer in
1950. He shows that, under certain conditions, it is impossible to
aggregate the preferences of at least three rational individuals in a
single collective preference, which would itself be rational (i.e.
represented by a complete and transitive relation over social states).
These conditions are the following: we must not exclude any
combination of individual preferences (no restriction domain); we
do not wish to resort to dictatorial decision (non-dictatorship); the
collective decision should not contradict the unanimous preferences
(Pareto Principle). Arrow also imposes a independence to non-
relevant alternatives conditions, which he interpreted as a ban on
interpersonal comparisons of utility8. This impossibility is at the

8
This interpretation is though debatable. It has been shown that the independence
condition also rules out further relevant ordinal information on individual
preferences since it focuses on binary comparisons. This nuance has been taken
seriously by the theorists of equity, so that they go beyond the arrovian
impossibility, as we shall see below, yet still avoiding interpersonal comparisons
of utility.
9

Electronic copy available at: https://ssrn.com/abstract=2357412

very least annoying: we cannot derive a collective judgment on the
basis of individual preferences unless it is dictatorial. It is hence
questionable whether the notion of collective welfare would at all
make sense. For this reason, the New Welfare Economics seemed
bound to a failure again. Fortunately, this prediction turns out not
to materialize.

3 – A promising future for welfare economics

Different challenges indeed need to be taken up to restore a
future for Welfare economics. It should be possible to make
recommendations of public policies; either interpersonal
comparisons of utility are impossible and not required, or their
meaning and their status should be clearly defended; a framework
to explicit which value judgments are at stake is needed; it is
necessary to go beyond the arrovian impossibility to legitimate the
use of social welfare function. While the economics of happiness
(subsection 3.1) has provided some positive evidences of the
necessity to challenge the notion of welfare, the comparative
approach (subsection 3.2) and the theory of equity (subsection 3.3)
lead us to expect a promising future for a normative welfare
economics.

3.1 – Economics of Happiness

Back in the 1950ies, Richard Easterlin has examined whether
income promoted happiness in the population on the basis of
opinion surveys. In his famous article published in 1974, he has
observed that, in a given country, people with higher incomes are
more likely to claim to be happy. However in international
comparisons, at least for countries with income high enough to
meet basic needs, the expressed level of happiness does not vary
much with the national per capita income. Finally, although the per
capita income has increased steadily in the United States between
1946 and 1970, expressed happiness recorded no upward trend in
the long run, and even decreased between 1960 and 1970. Facing
the Easterlin paradox, the standard public policies, which are based
exclusively on economic growth, seem to be missing their target. If
growth and wealth is not all what counts, the least would be to
primarily identify the factors for happiness. ‘Economics of
happiness’ is essentially a positive, interdisciplinary, and empirical
literature. It describes what is, but does not study what ought to be.
Happiness studies are interdisciplinary in the sense that they belong
to economics, cognitive sciences, humanities and social sciences.
Notice it constitutes an alternative to the standard economic model.
First, it moves away from the revealed preference model and from
10

Electronic copy available at: https://ssrn.com/abstract=2357412

the usual assumptions of rationality. Then the overall satisfaction of
individuals is at stake, rather than just the satisfaction they derive
from the consumption of market goods. It consists in conducting
econometric studies of happiness, emotion, subjective well-being,
quality of life, life satisfaction —insofar as those terms are, in this
specific context, interchangeable— to identify their factors.
Measurement of happiness often relies on self-assessment scales,
based on responses to questionnaires in which participants express
how happy they feel.
Since the Easterlin paradox, many studies have tried to explain
why at the aggregate level, growth of national income did not
necessarily enhance well-being. Among others, results of
economics of happiness reveal that poverty reduce more happiness
than wealth increases it; an increase of income for a poor person is
more likely to increase her happiness that an increase in income for
a rich person. Happiness can be enhanced by reducing inequalities,
improving working conditions, the reduction of working time and
in some cases, neutralizing the negative effects of unemployment
and some school reforms. Besides, we learn that the influence of
purely economic factors in the happiness of people is generally
overestimated in our representations as compared with factors.
However, unemployment and labor relations can have considerable
influence in the lives of people. Unemployment kills happiness,
even after individuals got their jobs back. Some think happiness
may constitute a yardstick, and that it is possible to transcribe it in
money measures, which allow cost-benefit analysis to be
completed.
Gathering information on the factors to enhance or to avoid
decreasing of happiness, as well as on the measure of happiness,
may most likely be of great help for policy-makers. It appears to be
a particularly innovative and important contribution to
understanding the determinants of happiness, for making ex post
evaluation of certain public policies, and to complete the data
needed by policy makers who should not be satisfied with
economic data. Nevertheless, the analyses of surveys have given
rise to many criticisms, at the methodological and the normative
level. Some highlight difficulties to interpret the replies, challenge
their reliability, and doubt cross-country comparisons are
meaningful. More generally, the very status of subjective data is
discussed. Would individuals be with what they have and what they
do, they may be happy out of adaptation. This becomes highly
problematic if adaptation is nothing but resignation. Beyond the
methodological criticisms, some question its ability to formulate
policy recommendations. At a pinch this research could justify to

11

Electronic copy available at: https://ssrn.com/abstract=2357412

administer tranquilizers to everyone, as it comes in Layard 2005.
Though very few economists would seriously defend this view, this
counter-intuitive example invites to beware of any possible
manipulations of happiness indicators. Furthermore, economics of
happiness describes what could be the target of a benevolent policy
maker —as did classical utilitarianism–, but as a pure positive
science, regardless of any justifications neither discussions of the
relevance of the happiness criterion. Happiness may be important
for individuals, yet this does imply governments are responsible for
enhancing it (See the justice cut by Dworkin).
Lucie Davoine (2009: 905) concludes: “happiness is a
necessary but not sufficient: even though the economics of
happiness can prevent a form of paternalism and ethnocentrism, the
surrounding methodological doubts and the objections in principle
induce not to consider happiness as the barometer of public action."

3.2 – The comparative approach

Throughout his critical analysis of the welfarist approach (See
in particular Sen 1979a,b), Amartya Sen suggested assessing social
situations by considering quality of life rather than just utility or
wealth. He developed the bases for the ‘comparative approach’ in
general and the illustration of what it could be, the capability
approach, notably in his Hennipman lecture published in 1985 and
his first Tanner lectures published in 1980 (See also Sen 1987,
1992a, 1993b). It constitutes an intermediate response to the debate
on ‘equality of what? ’ which opposes welfarist approaches to
resourcist theories. It provides rankings of situations based on
explicit criteria of justice, and considers objective descriptions of
life situations as relevant information to capture quality of life, i.e.
an index of individual welfare as an ‘individual basis for justice’.
On the one hand, utility, says Sen among many other critics, is too
sensitive to adaptation, and on the other hand, resources do not pay
attention to the particular individual ability to transform
commodities into well-being. Quality of life may hence be better
captured with functionings, which Sen (1985: 6-7) defines as “what
the person succeeds in doing with the commodities and
characteristics at his or her command. [...] It is an achievement of a
person: what he or she manages to do or to be. [...] The alternative
combinations of possible functionings a person can achieve and
from which he or she can choose one collection” is called
‘capability’ (Sen 1985: 7). At any moment, according to his
situation, his tastes, his life plans, a person may choose some
particular functioning among the capability set. The wider this set
is, the more the individual is free to choose between different
12

Electronic copy available at: https://ssrn.com/abstract=2357412

lifestyles. The use of capabilities as an informational basis to assess
quality of life therefore focuses not only on the role of commodities
in generating well-being, takes into account individual’s specific
ability to transform commodities into well-being, but also values
for itself the freedom to choose their lifestyle.
We should add two technical remarks and one further
discussion. Firstly, as far as this information is objective in the
sense that they are observable and measurable on a common scale,
interpersonal comparisons are meaningful (Baujard 2011) so that
the latter are justified and well accepted in this context. Secondly,
the assessment of capabilities is based on some valuation of lists of
different functionings, themselves being a vector of achieved
doings and beings. Such multi-dimensionality is likely to cause
moral dilemmas in certain situations, hence to generate substantial
incompleteness. For instance, what if I have more health but less
education or social relations? A possible answer, specific to Sen, is
to accept the rankings of social states may be incomplete. He does
not indeed consider incompleteness as a relevant problem in the
context of normative issues (See notably Sen 2009). Another
approach is to gather each functionings into an index by weighting
them according to their importance (Robeyns 2005a). Thirdly, the
crux of the debate opposing these two different philosophical
capability approaches lies in the question of operationalization (on
this opposition, see Lessman 2006, Robeyns 2005b, Baujard
2007b). Following Aristotle, Martha Nussbaum believes that there
is a single notion of the human good, virtues and flourishing life
(Nussbaum 1988, 1993). This leads to propose a concrete and
comprehensive list of functionings, so that the approach belongs to
fundamental universalism. Operational applications are therefore
implementable for scientists (Alkire , Robeyns 2006). The fact that
values and weighting are determined by scientists rather than by the
individuals themselves explains why this approach is often
criticized for its paternalism. In contrast, Sen’s position meets
certain relativism, in order to give to public deliberation the main
role in a democracy. Therefore, he refuses to provide a clear list of
functionings which could measure well-being for everyone on a
common scale. It is therefore difficult to implement a mere
application of Sen’s capability approach since it fundamentally
relies on the public debate. The latter is the only legitimate place to
decide which moral values should be at stake, hence to retain
specific lists of functionings and, eventually, to measure
capabilities.
The extremely extensive literature on the capability approach,
at least since the 2000s, is but a multi-dimensional analysis of

13

Electronic copy available at: https://ssrn.com/abstract=2357412

living conditions, for which the UNDP human development index
(HDI) is only one very rough illustration. As the approach was
generalizing, it has yet lost its specificity, which was to pay special
attention to the value of freedom, understood as the possibility for
everyone to live the life one has reason to value.

3.3 – Equity theory

The theory of fairness or equity theory, including fair
allocations theory and even applications to public economics,
borrowed the axiomatic methods from social choice theory and the
theory of bargaining to study the implications of equity criteria in
the framework of Arrow-Debreu general equilibrium model.
Different fairness criteria can be contemplated for division
rules. The idea of ‘no-envy’ was independently introduced by Jan
Tinbergen (1953), Duncan Foley (1967) and developped by Serge-
Christophe Kolm (1971), Allan Feldman and Alan Kirman (1974).
An allocation is ‘envy-free’ if no individual would like better
anybody else’s basket. A fundamental result of equity theory is
such that the competitive equilibrium with equal endowments, that
is to say equal budgets, satisfies both the criteria of no-envy and
Pareto. Refinements of such analyses were first conducted in the
context of distribution of a consumption economy without
production, then to study equal opportunities, incentives and
optimal taxation, division of a single divisible good with single-
peaked or monotonic preferences, the allocation of several
commodities, the properties of a production economy...
The no-envy criterion, however, may conflict with the criterion
of efficiency. This was proved by Elisha Pazner and David
Schmeidler in 1974: no allocation respects Pareto efficiency and
fairness (as no-envy) in the context of production with unequal
skills —in other words with production handicaps. This
impossibility result can be interpreted as the incompatibility
between a principle of reward and a principle of compensation. The
non-envy test indeed requires that the allocation of individuals with
identical preferences must be on the same indifference curve.
According to the principle of reward, individuals with similar
talents should not envy each other, since it should not be any
different treatment for different preferences. And, according to the
principle of compensation, individuals who have identical
preferences should have the same benefit, eliminating the
inequalities due to talents.
The same authors proposed in 1978 another test of fairness
based on egalitarian equivalent allocations. An allocation is
egalitarian-equivalent when each one is indifferent between the
14

Electronic copy available at: https://ssrn.com/abstract=2357412

basket of goods in the allocation and the basket she would have in
an egalitarian economy. In this perspective, Marc Fleurbaey and
François Maniquet (2005)—among other similar contributions—
considered the introduction of skills heterogeneity, and studied the
consistency between compensation of skills inequalities and the
condition of equal access to resources for all preferences. To
deepen the subject of responsibility and unequal handicaps, see
Fleurbaey (2008) and, for a comprehensive presentation of the
economic theory of fairness, see Fleurbaey and Maniquet (2011).
The theory of equity took up the different challenges welfare
economics was facing. First, it’s worth noticing it eventually
overcame the arrovian impossibility. Second, it did reject
interpersonal comparisons of utility. Unlike standard economics
which relies on the model of subjective revealed preferences,
welfare is here described as an index of resources; and unlike the
comparative approach, they still keep some account of individual
ordinal preferences, which avoids the risk of paternalism. Third, the
theory of fairness accepts the challenge of value judgments
transparency in making clear the criteria of justice.

4 – Conclusion
Public policies are expected to increase social welfare. Welfare
economics aims at providing the framework to accomplish such
goal, developing a wide range of techniques to adapt different
situations. But looking careful, we have seen this wonderful
textbook world may be gloomier than it seems at first sight. Is
welfare economics bound to death because of its difficulty to
handle value judgments? Recommendations are indeed always
linked with some normative involvement, even through the
undebated Pareto criterion. Beyond, welfare economics suffers
from the fact that a necessary discussion on the very definition of
welfare had been shirked for too long (Baujard 2011). What is
indeed welfare? How can we justify this or that meaning of
welfare for public policy? Pareto or Pigou acknowledge that
overall welfare is much more, or even different, than economic
welfare9. Yet Pareto developed a pure theory of ophelimity which
was afterwards taken as such by economists. Pigou eventually

9
We here refer to the distinction of utility and ophelimity for Pareto, and the
distinction between total welfare and economic welfare by Pigou. Pigou (1920 :
20-33) for instance argues the difference is meaningful as soon as you pay
attention to time and interactions.
15

Electronic copy available at: https://ssrn.com/abstract=2357412

focused on economic welfare, and especially on the national
dividend; from then on GDP appeared as an acceptable
approximation of welfare for decades. These assumptions are today
more and more debated.
Among others, it seems now generally accepted that GDP is a
questionable goal, and the definition of welfare becomes a topic of
discussion. The theory of fairness and social welfare, which took
over all challenges faced by welfare economics, now provides a
unified approach of social choice theory, the theory of fair
allocations and public economics. Sen’s capability approach
rehabilitates the role of public debate and reintroduces democracy
in economics. All is there to expect a promising future for (welfare)
economics.
References
[1] ALKIRE, S. Valuing Freedoms. Sen’s capability approach
and poverty reduction. Queen Elizabeth house series in
development studies. Oxford university Press, Oxford,
2002.
[2] ARROW, K. J. A difficulty in the concept of social welfare.
The Journal of Political Economy 58, 4 (Aug. 1950),
328–346.
[3] ARROW, K. J. An extension of the basic theorems of
classical welfare economics. In Proceedings of the Second
Berkeley Symposium on Mathematical Statistics and
Probability, J. Neyman, Ed., vol. 1 of The international
library of critical writings in economics. University of
California Press, Berkeley and Los Angeles, 1951,
pp. 507–532. reprinted as Cowles Commission Paper,
New Series, N.54 ; also reprinted in Baumol and Wilson,
Vol.2, chap.5.
[4] ARROW, K. J. Social Choice and Individual Values, 2 ed.
Yale University Press, New Haven and London, 1963.
Corrected and completed from the first ed.: John Wiley &
Sons, New York, 1951.
[5] ARROW, K. J., AND DEBREU, G. Existence of an
equilibrium for a competitive economy. Econometrica 22,
3 (1954), 265–290.
[6] BAUJARD, A. Commensurable freedoms in the capability
approach. A discussion of Pattanaik and Xu’s ‘Minimal
relativism, dominance, and standard of living
comparisons based on functionings’. WP CREM 2007-03,
series “Public Economics and Social Choice" (2007).

16

Electronic copy available at: https://ssrn.com/abstract=2357412

[7] BAUJARD, A. L’économie du bien-être est morte. Vive
l’économie du bien-être ! Working Papers CREM 2011-
02, ‘Public Economics and Social Choice’ Series, june
2011.
[8] BLACKORBY, C., AND DONALDSON, D. A review article:
The case against using the sum of compensating
variations in cost-benefit analysis. Canadian Journal of
Economics 23, 3 (1990), 471–494.
[9] BOADWAY, R., AND BRUCE, N. Welfare Economics.
Blackwell, Oxford, 1984.
[10] BURK, A. A reformulation of certain aspects of welfare
economics. The Quarterly Journal of Economics 52
(1938), 310–334.
[11] CHIPMAN, J. S., AND MOORE, J. C. The new welfare
economics, 1939-1974. International Economic Review
19, 3 (october 1978), 547–584.
[12] COOTER, R., AND RAPPOPORT, P. Were the ordinalists
wrong about welfare economics. Journal of Economic
Literature 22, 2 (june 1984), 507–530. In M. Blaug ed.,
A. C. Pigou, Edward Elgar Publishing Limited, chap. 12,
pp. 150–173, 1992.
[13] DAVOINE, L. L’économie du bonheur. Revue économique
60, 4 (2009), 905–926.
[14] DE SCITOVSKY, T. A note on welfare propositions in
economics. Review of Economic Studies 9 (1941), 92–95.
[15] DEBREU, G. The coefficient of resource utilization.
Econometrica 19, 3 (1951), 273–292.
[16] EASTERLIN, R. A. Does economic growth improve the
human lot? In Nations and Households in Economic
Growth: Essays in Honor of Moses Abramovitz, P. A.
David and M. W. Reder, Eds. Academic Press, Inc., New
York, 1974.
[17] FELDMAN, A., AND KIRMAN, A. Fairness and envy.
American Economic Review 64, 6 (Dec. 1974), 995–1005.
[18] FLEURBAEY, M. Fairness, responsibility and welfare.
Oxford University Press, 2008.
[19] FLEURBAEY, M., AND MANIQUET, F. Fair social orderings
when agents have unequal production skills. Social
Choice and Welfare 24, 1 (feb 2005), 93–127.
[20] FLEURBAEY, M., AND MANIQUET, F. A theory of fairness
and social welfare. Cambridge University Press, 2011.
Econometric Society Monographs.
17

Electronic copy available at: https://ssrn.com/abstract=2357412

[21] FLEURBAEY, M., AND MONGIN, P. The new of the death of
welfare economics is greatly exagerated. Social Choice
and Welfare 25 (2005), 381–418.
[22] FOLEY, D. Resource allocation and the public sector. Yale
Econ. Essays 7, 1 (1967), 45–98.
[23] HAUSMAN, D. M., AND MACPHERSON, M. S. Economic
analysis and moral philosophy. Cambridge University
Press, 1996.
[24] HICKS, J. R. The foundations of welfare economics. The
Economic Journal 49 (december 1939), 696–712.
[25] HICKS, J. R. The rehabilitation of consumer’s surplus.
Review of Economic Studies 8 (1941), 108–116.
[26] KALDOR, N. Welfare propositions of economics and
interpersonal comparisons of utility. The Economic
Journal 49 (september 1939), 549–552.
[27] KOLM, S.-C. Justice et équité. CEPREMAP, Paris, 1971.
New ed. Paris, CNRS, Monographies du séminaire
d’économétrie, René Roy Dir., 1972.
[28] LAYARD, R. Happiness. Lessons from a new science.
Penguin books, London, 2005.
[29] LESSMANN, O. Effective freedom and combined
capabilities: Two different conceptions of capability.
Discussion paper n.152 of the Institute of
Socioeconomics, University of Hamburg, 2006.
[30] MISHAN, E. J. Economic efficiency and social welfare.
Selected essays on fundamental aspects of the economic
theory of social welfare. G. Allen and Unwin, London,
1981.
[31] MONGIN, P. Is there progress in normative economics? In
Is there progress in economics? Edward Elgar,
Cheltenham, 2002, ch. 10, pp. 145–169.
[32] MONGIN, P. Is there progress in normative economics?
Economics and Philosophy 22 (2006), 19–54.
[33] MYINT, H. Theories of welfare economics. Augustus M.
Kelley, Bookseller, New York, 1965. Published for The
London School of Economics and Political Science.
[34] NUSSBAUM, M. C. Nature, function and capability:
Aristotle on political distribution. Oxford Studies in
ancient Philosophy 6 (1988), 145–184.
[35] NUSSBAUM, M. C. Non-relative virtues: An Aristotelian
approach. In The quality of life. Clarendon Press, Oxford,
1993, pp. 242–276.
18

Electronic copy available at: https://ssrn.com/abstract=2357412

[36] PARETO, V. Manual of Political Economy, italian, 1906;
french transl., 1909; english transl., 1971 ed. 1906. Here,
pages of the French version, M. Giard et E. Brière: Paris,
1909.
[37] PAZNER, E., AND SCHMEIDLER, D. A difficulty in the
concept of fairness. Review of Economic Studies 41, 3
(1974), 441–443.
[38] PIGOU, A. C. Wealth and welfare. MacMillan, London,
1912.
[39] PIGOU, A. C. The economics of welfare. MacMillan,
London, 1920. 4ème édition, 1932; here 1962.
[40] ROBBINS, L. An essay on the nature and significance of
Economic Science, 3ème ed. Macmillan, London, 1932.
Voir également des extraits in D. M. Hausman, Ed. The
philosophy of economics. An anthology, Cambridge
University Press, chap. 4, pp. 113-140, 1984.
[41] ROBEYNS, I. The capability approach: A theoretical
survey. Journal of Human Development 6, 1 (march
2005), 93–114.
[42] ROBEYNS, I. Selecting capabilities for quality of life
measurement. Social Indicators Research 74, 1 (October
2005), 191–215.
[43] ROBEYNS, I. The capability approach in practice. Journal
of Political Philosophy 14 (2006), 351–376.
[44] SAMUELSON, P. A. Foundations of Economic Analysis.
Harvard University Press, Cambridge, 1947.
[45] SEN, A. K. Personal utilities and public judgements : Or
what’s wrong with welfare economics? The Economic
Journal 89, 355 (septembre 1979), 537–558.
[46] SEN, A. K. Utilitarianism and welfarism. The Journal of
Philosophy 76, 9 (1979), 463–489.
[47] SEN, A. K. The welfare basis of real income comparisons:
A survey. Journal of Economic Literature 17, 1 (1979),
1–45.
[48] SEN, A. K. Equality of what ? In The Tanner Lectures on
human Values, S. McMurrin, Ed., vol. 1. Cambridge
University Press, Cambridge, 1980. Also in A. K. Sen,
Choice, Welfare and Measurement, MIT Press,
Cambridge, pp. 365–366, 1982.
[49] SEN, A. K. Commodities and Capabilities. Oxford
University Press, 1985.

19

Electronic copy available at: https://ssrn.com/abstract=2357412

[50] SEN, A. K. The standard of living. Cambridge University
Press, Cambridge, 1987.
[51] SEN, A. K. Inequality re-examined. Clarendon Press,
Oxford, 1992.
[52] SEN, A. K. Capability and well-being. In The quality of
life. Clarendon Press, 1993, pp. 30–53.
[53] SEN, A. K. The possibility of social choice. American
Economic Review 89, 3 (june 1999), 349–378.
[54] SEN, A. K. The idea of justice. Harvard University Press,
Cambridge (MA), 2009.
[55] TINBERGEN, J. Redelijke Inkomensverdeling, 2 ed. N.
DeGuiden Pers, Haarlem, 1953.

20

Electronic copy available at: https://ssrn.com/abstract=2357412