Scribd
Upload
19 views11 pages

Paretian Liberal Paradox-Lec-13

Uploaded by

abc def
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
19 views11 pages

Paretian Liberal Paradox-Lec-13

Uploaded by

abc def
Copyright
© © All Rights Reserved
We take content rights seriously. If you suspect this is your content, claim it here.
Available Formats
Download as PDF, TXT or read online on Scribd
You are on page 1/ 11

Aggregation of individual preference to reach a social

preference relation.
Arrow’s Theorem says that we cannot have a social welfare
function satisfying unrestricted domain, Pareto principle,
Independence of irrelevant alternatives, and non-dictatorship.
We know that simple majority violates transitivity.
Borda count violates independence of irrelevant alternatives.
Instead of social welfare function, if we consider social
decision rule - a function for which choice set is defined, we
get possibility result.
Social preference relation based on Pareto superiority and
Pareo non-superiority.
May’s Theorem- it characterizes simple majority rule as a
decisive collective choice rule.
By restricting the domain, number of possibility results can be
generated.
Paretian Liberal Paradox
Liberal value means that for some alternatives which are personal
to some individuals, those relevant individuals should be free to do
what they like. If relevant person is allowed to do whatever he likes
in case of his personal things, keeping everything else constant, the
social state will be at a better state.

Liberal values is defined as (condition L): For each person i there is


at least one pair of distinct alternatives (x , y ) such that he is
decisive in the social choice between them in either order that is
xPi y → xPy and yPi x → yPx .
Decisiveness is taken for every individual for a particular distinct
pair of alternatives. The weaker version of this definition can be to
restrict the number of decisive individual.
Minimal Liberalism ( condition L∗ ): there are at least two persons
k and j and two pairs of distinct alternatives (x , y ) and (z, w ) such
that k and j are decisive over (x , y ) and (z, w ) respectively, each
pair taken in either order.
It is obvious that condition L implies L∗ .
Theorem: There is no social decision function satisfying conditions
U, P and L∗ .

Proof: When (x , y ) and (z, w ) are same pair then condition L∗


does not hold. Suppose the pair has one of the elements in
common, x = z then let xPk y , wPj x and for all i yPi w . By
condition L∗ we have xPy and wPx . By condition P we have yPw .
This violates acyclicity. We have xPy &yPw and wPx . Therefore,
we dont any best element. Thus, choice set is empty. We cannot
have social decision function.
Suppose all four alternatives x , y and z, w are distinct. We assume
that xPk y , zPj w and for all i we have wPi x &yPi z.
These orderings are possible for example
k j l
w y w
x z y
y w x
z x z
By condition L∗ we have xPy and zPw . By condition P we have
wPx and yPz.
We have xPy & yPz & zPw and wPx . It violates acyclicity.
Therefore, there is no best element. Thus, choice set is empty. We
cannot have a social decision function.
In this theorem we dont have independence of irrelevant
alternatives and transitivity of social preference relation and still we
are getting a negative result.

The negative result implies that the four conditions U, P, L∗ and


SDF are not compatible together.
If we relax these conditions we may get possibility result.
We may drop acyclicity condition on social preference. If we drop
acyclicity on the social preference relation R, then R may not
satisfy condition α. We may have a situation like x is chosen from
(x , y ), y is chosen from (y , z) and z is chosen from (x , z) and x is
chosen from (x , y , z). This social choice is not consistent is the
way we have defined consistent choice based on property α and β.
Dropping Pareto principle means we have a social preference
relation which gives outcome oposite of what everybody likes. It is
very natural to have such assumption we want the social choice to
be based on individual preference ordering.
We may weaken the definition of minimal liberalism by reducing
the number of decisive individual. Already we have only two
individuals. If we make it one, then we may have that individual as
dictator. It will be against liberal values.

We may restrict the domain. It is often done in the literature. We


have to be careful in doing so.
Example:
k j l
x y y
y x z
w z x
z w w
Individual k is decisive over x , y and Individual j is decisive over
z, w . We get xPy and zPw . From Pareto we have yPz and xPw .
Thus, we have xPy , yPz, zPw and xPw . We have x as the best
element in this social preference.

You might also like