By Dr. Mnaku H. Maganya Department of Economics and Tax Management
By Dr. Mnaku H. Maganya Department of Economics and Tax Management
By Dr. Mnaku H. Maganya Department of Economics and Tax Management
1
PUBLIC CHOICE THEORY
1. Meaning of Public Choice Theory
2.Voting Paradox
3. Arrow’s Impossibility Theorem
4. Single Peaked Preference and Public
Spending
5. Logrolling
2
Public Choice Theory
Public choice theory uses economic theories to
analyze political decision processes. Public choice
is an integral part of the field of public economics.
It have been stressed that a lack of markets also
can mean a lack of information, so it is useful to
ask how well public choice mechanisms work to
translate individual preferences into publicly
chosen policies.
In particular, we may ask how well public choice
mechanisms fare in terms of (1) rationality and (2)
efficiency.
3
Voting Paradox
Voting occurs in many situations, such as in
elections of public officials, officers of a club, or
among a group of friends who have to decide in
which restaurant to eat. While elections involving
just two choices are quite simple, the opposite is
true for elections with three or more alternatives,
in which many complications and paradoxes arise.
Social choice theory was developed to analyze
the various types of voting methods, to discover
the potential pitfalls in each, and to attempt to
find improved systems of voting.
4
Voting Paradox
Voting paradox suggest that there exist no
majority voting equilibrium, even if there is
equilibrium it may not be Pareto efficiency. This
has important implication, since the outcome
depends on the order of voting process it becomes
important to control the agenda.
To begin discussing the voting paradox, suppose
that there are three individuals, 1, 2, and 3, and that
public decisions among three potential outcomes
(for example, a level of public spending), A, B, and
C, are made using majority voting.
5
Voting Paradox
Suppose that individual preferences satisfy:
1: 𝐴≻𝐵≻𝐶
2: 𝐶≻𝐴≻𝐵
3: 𝐵≻𝐶≻𝐴
We use pair wise voting to determine majority rule
winner among the three given alternatives:
A vs B ~ A is majority winner
B vs C ~ B is majority winner
A vs C ~ C is majority winner
Hence there is cycling meaning no overall majority
winner.
6
Voting Paradox
Diagrammatical illustration of the voters
preferences
Choices 1st Voter 2
2nd
Voter 3
3rd
Voter 1
A B C
Alternatives
7
Voting Paradox
For the above given preferences each individual
has well-defined preferences over the three
outcomes, but the ranking based on majority
voting is plagued by “cycling”:
A is chosen over B, B is chosen over C, and yet C
is chosen over A. This is the famous Voting
paradox, which suggests that majority voting is
not rational in the sense that it does not satisfy
the simple property of transitivity.
8
Arrow’s Impossibility Theorem
Can we do better? Arrow’s Theorem addresses
this question. The theorem starts with the
following axioms:(the characteristics of ideal
political mechanism according to Arrow)
1. Unrestricted domain.
2. The mechanism must satisfy the Pareto principle.
3. Independence of Irrelevant Alternatives.
4. There can be no a dictator in a society
5. Satisfying axiom of Transitivity.
9
Arrow’s Impossibility Theorem
Arrows Impossibility Theorem: There is no rule
that would satisfy all characteristics (axioms). No
mechanism based on individual preference
orderings satisfies all of these axioms, so we must
consider which of the axioms we might do
without.
It would seem that the ranking of outcomes
based on a standard social welfare function would
satisfy all of these axioms; while this is true, a
social welfare function is not based simply on
individual orderings of outcomes, but on the
strength of these orderings as well.
10
Single Peaked Preference and Public Choice
One possible route out of our dilemma is to
exclude the first axiom (unrestricted domain)
listed above from our list of necessary conditions
of ideal political mechanism.
For example, suppose all individuals have single-
peaked preferences over potential outcomes,
meaning that the outcomes can be ordered in a
single dimension in such a way that there is a
single, global maximum, with movements in
either direction associated with less and less
preferred outcomes.
11
Single Peaked Preference and Public Choice
The following figure illustrates single-peaked
preferences for three individuals over three
potential outcomes, A, B, and C, with valuations
on the vertical axis.
Joseph
1 st
Choices
2nd
John
3rd Jane
A B C Alternatives
12
Single Peaked Preference and Public Choice
Example 1: From the following Table, determine
if there is a winner using a majority winner. If so,
which one?
VOTERS
RANK 1 2 3 4
FIRST A B B C
SECOND B A C B
THIRD C C A A
13
Single Peaked Preference and Public Choice
Solution: You must determine the outcome of
three one-on-one competitions (pair-wise voting).
The outcome not considered in each one-on-one
competition can be ignored.
A vs B, A:1, B:3
A vs C, A:2, C:2
B vs C, B:3, C:1
Since B can defeat both A and C in a one-on-one
competition, B is the majority winner and hence
there is no voting paradox.
14
Single Peaked Preference and Public Choice
Example 2: Consider the following information
given below for three voters, three alternative and
the voter’s preferences.
CHOICES VOTERS
JUMA ROSE ASHA
FIRST A C B
SECOND B B C
THIRD C A A
a) Plot the above voter’s preference in a diagram
b) Determine if there will be majority vote rule or
there will be a voting paradox.
15
Single Peaked Preference and Public Choice
First lets determine the outcome of three one-on-
one competitions (pair-wise voting). The
outcome not considered in each one-on-one
competition can be ignored.
A vs B = B will win the contest
B vs C = B will win the contest
A vs C = C will win the contest
Hence B is the majority winner and there will be
no voting paradox.
16
Single Peaked Preference and Public Choice
First lets determine the outcome of three one-on-
one competitions (pair-wise voting). The
outcome not considered in each one-on-one
competition can be ignored.
A vs B = B will win the contest
B vs C = B will win the contest
A vs C = C will win the contest
Hence B is the majority winner and there will be
no voting paradox.
17
Single Peaked Preference and Public Choice
Plot the above given voter’s preference in a
diagram
1st Rose
2nd Asha
3rd
Juma
A B C
18
Double Peaked Preferences
Plot of the voter’s double peaked preference in a
diagram resulting into voting paradox
1st
2nd JOHN
3rd JOHNSON
A B C D E
19
Logrolling
In a majority voting system there is also
possibility to negotiate (logrolling) or simply vote
trading. This may lead to that efficient projects are
actually (that would not have been carried out if
no logrolling) carried out, but sometimes it may
cause oppression of minorities, and it can also
lead to inefficiency
Note that there is also possibility logrolling leads
to that projects that are inefficient for the society
may be carried out.
20
Logrolling
Example: Assume that we are about to vote for two
projects, a bridge and a tunnel. The net incomes for
three persons from the projects are as follows:
Jane John Paul Total
Bridge -10 -10 30 10
Tunnel -10 25 -10 5
Total (individual) -20 15 20 15
If vote separately for each project and there is no
logrolling, none of the projects will be carried out. If
John and Paul decided to co-operate; they vote for
both projects and society will benefit as well.
21
THANK YOU
FOR
YOUR ATTENTION.
22