Efficiency in economics means any state of resource allocation which maximizes social

welfare.
An economic state is Pareto-efficient if the allocation of resources in that state cannot be rearranged to make at least one individual better off without making any other worse off.

New Welfare Economics: Compensation Principle

Pareto formulated the concept of social optimum by using concept of ordinal utility and free
form interpersonal comparisons of utilities. Pareto criterion does not apply to those economic
changes which harm some and benefit others. For example, it doesnt say whether social
welfare increased or not form movement in either direction along a contract curve in terms of
Edgeworth box diagram. Since each point on the contract curve is Pareto optimum, Pareto
analysis leaves considerable amount of indeterminacy.

Kaldor-Hicks Welfare Criterion: Compensation Principle

Have made efforts to evaluate the changes in social welfare resulting from any economic
reorganisation which harms some and benefits others.

Assumptions
1.
2.
3.
4.

Satisfaction of an individual is independent of the others and he is best judge of his welfare.
No externalities of consumption and production.
Tastes of individual remains constant
Problems of production and exchange can be separated from problems of distribution.
Compensation principle accepts level of social welfare to be a function of level of production.
Thus it ignores the welfare effects of change in distribution.
5. Utility can be measured ordinally and interpersonal comparisons of utilities are not possible.
It is a value-free objective criterion of measuring social welfare.
If a certain change in economic organisation or policy makes some individual
better off and some other worse off, then that change will increase social welfare if
those who gain from the change could fully compensate the losers and still be
better off than before.
Thus those who gain evaluate their gains at a higher figure than the value the losers set upon
their losses.

In terms of utility possibility curve, movement down along DE will represent increase in utility of A
while decrease in utility of B. Suppose, utilities obtained by A and B from the distribution of income
or output between them is represented by point Q inside the utility possibility curve.

If due to an economic policy change, individuals move from Q inside PPF to T on PPF, Bs
utility has increased while As has decreased. Pareto criterion cannot evaluate such change as
it requires interpersonal comparison. Movement to any point on segment RS is socially
preferable according to Pareto criterion.
Acc. to Kaldor-Hicks compensation principle, if B gives some compensation to A for loss
suffered such that they move from T to R, A will be as well off as at Q but B will be still better
off. Therefore, according to Kaldor-Hick criterion, social welfare has increased with movement
from position Q to T because form T they can move to R by mere redistribution of income.
If they move from T to G, both will be better off than at Q. Thus T is superior to Q as by
redistribution they can move to G where both are better off compared with Q.
Compensation need not be actually paid, it is enough to know that gainers can compensate
losers and still be better off. Actual redistribution is left for government to decide.

It should be noted that such a compensation where the gainers are still better off after
compensation can only happen when change in economic policy leads to higher output
or income. Thats how they separate change in output from change in distribution. When the
economy has moved to a productively more efficient position, social welfare has increased.
Now whether actual redistribution takes place through compensation is a different matter.

Utility Possibility Curve moving Outwards due to Policy Change

Any change in the economy which moves individuals from a position on lower utility
possibility curve to a position on higher utility curve increase social welfare. Because in such a
case compensation leaves gain for
As a result of movement from Q on UV to R on UV, social welfare has increased as by mere
redistribution of income, individuals can move to S where B has been fully compensated for
his loss and A is still better off compared to position Q.

Scitovsky Paradox

Contradictory results obtained from KH criterion if following the policy change, new UPC
intersects old UPC.
For getting consistent results, if some position B has been shown to be an improvement over
an initial position A, then position A must not be preferred to position B on the same criteria.
Such contradiction exists and is known as Scitovsky Paradox.
Two intersecting utility possibility curves are drawn. Initial UPC is JK and as a result of policy
change it takes position GH (steeper). Movement from C on JK to D on GH is superior on basis
of Kaldor-Hicks criteria as movement o F on GH can be made by mere redistribution of
income, where B has been fully compensated and A is still better off.
But reverse movement from D on new UPC GH to C on JK also represent an improvement acc.
to Kaldor-Hicks criterion as by movement from C to E can be made by mere redistribution of
income and it will compensate A and B is still better off.

Scitovskys Double Criterion of Welfare

Double criterion requiring fulfilment of Kaldor-Hicks test and also the fulfilment of the
reversal test.
A change is an improvement if gainers in changed situation are able to persuade the losers to
accept the change and simultaneously the losers are not able to persuade the gainers to
remain in the original situation.

Critique
1. Only a definition of increase in wealth or efficiency and not a new welfare criterion as he
assumed welfare to be a function of increase in production or efficiency irrespective of
changes in distribution.
2. Not free from value-judgement, to say that a policy which meets the criterion increases
output or efficiency of a society is, in effect, to recommend it.
3. Compensation is hypothetical and is consistent with making the poor yet poorer.
4. Potential money compensation sets up interpersonal comparison on a money basis.
If the poor loser looses Rs. 100 and the rich gainer gains Rs. 500, we cannot say that welfare
has increased as the poor may value Rs. 100 more than the rich values Rs. 500. Might mean
major loss to welfare of poor and trivial gain to welfare of rich even though Kaldor will
recommend the change.
5. Does not take into account external effects on consumption and distribution. Welfare of an
individual doesnt depend on his own production and consumption but on the relative
economic position in society. An economic change leaving a person at same position as
before while making others better off will lead to him not feeling as well-off as in original
position and his welfare will decrease.
6. Separation of production change from distribution change: If as a result of policy change,
production of Coke increases and that of whisky decreases and X prefers the former and Y the
latter.
7. Welfare has dual aspect- absolute and relative. People are dissatisfied not only because they
are poor but because others are very rich. A lower output equitably distributed ensures
greater social welfare than a greater output inequitably distributed.
8. If compensation is actually made, KH criterion is quite unnecessary as in that case Pareto
criterion will be sufficient.
Arrows Theory of Social Choice

Construction of social welfare function reflecting preference of all individuals in a

society is an impossible task. Main contention of Arrow is that it is difficult to set up
reasonable democratic procedures for the aggregation of individual preferences into a social
preference for making a social choice.
Arrows condition for social choice: 5 conditions which must be met for an acceptable social
welfare function:
1. Transitivity or Consistency: If alternative A is socially preferred to B and alternative B is
socially preferred to alternative C, then alternative C will not be socially preferred to
alternative A.
2. Independence of Irrelevant Alternatives: If alternative A is socially preferred to B and
alternative B is socially preferred to alternative C, and if C is not available, it must not be
the case that B is preferred to A. Social ranking of any two alternatives is determined
exclusively by the individual rankings of these two alternatives alone and should not be
impacted by any other alternative which is not available.
3. Responsiveness to Individual Preferences: Social ordering must depict responsiveness to
individual preferences. The social welfare function is such that the social ordering
responds positively to alteration in individual value or at least not negatively. If an
alternative A is socially preferred to alternative B, and now due to economic change some
individual prefer A more strongly than before and no ones preference for A declines then A
must remain socially preferred to B.
4. Non-imposition: Social choices must not be imposed independently of individual
preferences. The choices must not be determined by anybody outside the community. If
nobody prefers B to A and one or few prefer A to B, then society must prefer A to B.
5. Non-dictatorship: Social choices must not be dictated by any one individual in the
community. A must not be preferred to B only because any one individual in society
prefers A to B irrespective of the preferences of other individuals. Implies that democratic
method of voting should determine social choices.

Arrow has shown it is impossible to make social choices without violating at least one of the
above 5 conditions. In other words, it is impossible to construct a social welfare function on
the basis of individual values that satisfy all the above conditions.
Specifically demonstrated that when the choice is between more than two alternatives,
then the individuals voting or expression of preferences for them would lead to inconsistent
or contradictory results so that no valid social choice can be made by the majority rule.

Arrows Impossibility Theorem

If we exclude the possibility of interpersonal comparisons of utility, then the only method of
passing from individual tastes to social preferences which will be satisfactory and defined for
a wide range of sets of individual ordering are either imposed or dictatorial.
Consistent social choices cannot be made without violating the consistency (transitivity)
principle.
Arrow has only assumed ordinal rankings and not the intensity or weights. If weights are
assigned or intensity for individual preferences is considered, then consistent and noncontradictory social choice can be made.
Amartya Sen resolved the problem of inconsistency involved in Arrows theorem by pointing
out that in such a situation for making a social choice by democratic mechanism, more
information is needed to make a rational social choice for maximising social welfare. With
additional information, it is possible to arrive at a rational social choice,