Public Finance In Economics

(Econ 3122)
Chapter 3
Public choice and theories of Government
Expenditure Growth
Addis Ababa University
Department of Economics
Course instructor: Kefyalew Endale (Ph.D.)
April 2021
1
 Contents of this chapter
•Definition of public choice
•Voting models
•Direct democracy (unanimity, majority voting, median voter
theorem, log rolling, plurality voting, vote trading and so on )
•Representative democracy
•Politicians
•Bureaucrats
•Interest groups
•Iron triangle
•The issue network
•Public expenditure growth theories (Wagner, peacock
and Wiseman)
2
 Public choice theory
Objective of this chapter:
• Discussion of how individual preferences are translated in
terms of public choice through different voting models
Definition
Public Choice theory: is the study of political decision making
in the public sector.
• It consists of applying economic theory in collective decisions
and their processes and outcomes.
• Legislation (the process of making or enacting laws) and
policy decisions are public choices.
It attempts to understand
– how public policies come to be adopted using economic models and
the impacts of policies on households, firms, and individuals
– The behavior of politicians under a variety of political institutions.
– It tends to use rational choice models to do so
• The decisions are made through voting.
3
 Public choice theory cont’d
• Voting models helps to translate individual preferences
into group or social preferences.
• When the preferences of different individuals are
expressed through points in the diagram or in written
form to derive the group choice using voting models it is
known as Spatial Voting Models.
• Broad classification of votings
a. Direct voting (democracy): all people are either voting or
reach agreement. Voting procedures in Direct Democracy
• Unanimity rules
• Majority voting rules
• Log-rolling (vote trading) – intensity of preferences
b. Representative democracy (Indirect voting): involves
decision making through democratically elected representatives.
4
 Public choice theory cont’d
Direct democracy:
a) Unanimity rules: a type of political decision making where
all parties must agree for a policy/programme/project to be
implemented.
• Example: In principle, society could agree that a public good
should be provided rather than not being provided.
Limitations of unanimity
• It has large decision making costs, i.e., projects should be
formulated in such a way that all voters/parties benefit
• All voters have to be fully informed
• It encourages strategic behavior or
voting outcome=f(bargaining power)
• Some of the voters might have a special preference and may
not vote with the majority 5
 Public choice theory cont’d
b) Majority Voting rules: one more than half [(N/2)+1] of the
votes must favor a measure/a project to gain an approval.
– the most familiar/popular method
Optimal majority voting rule: a rule that minimizes the total
cost of voting. There are two types of costs
i. Decision making costs (C): are the costs of discussion,
organization, bargaining etc.
 increases with the size of the majority due to holdouts
(refusals/resistance) and transaction costs
ii. Externality cost on the minority (D): those not favoring the project
with the majority are affected by the decision negatively.
• The majority tyrannizing the minority
• The larger the size of the majority, the lower is the externality cost
• The optimal majority voting rule is obtained by minimizing the
sum of decision making and externality costs (i.e min C+D)
6
 Public choice theory cont’d
• The total cost of majority voting is C+D. These costs depend on the size of
voters. Min C(N)+D(N) st. N dC ( N )

dD( N )
0
d (N ) d (N )
F.O.C
dC ( N ) dD( N )
 
d (N ) d (N )
• The minimum of the C+D curve is not necessarily the crossing point!
• efficient majority voting rule is when slope of the two cost functions are equal
– K/N is the ideal voting rule as the two slope are equal

C+D Costs
Costs

Decision Making
Costs

Ext ernal
Costs

K N

Number of Individuals
Required to Take
Collective Action 7
 Public choice theory cont’d
• Limitations of majority voting
Cyclical voting: paired voting can go on forever without reaching a
decision. The cycling is a result of repeated paired voting.
• Assume three projects (A,B,C) and three voters (X, Y, Z)
– Project A: spending $100,000 on hospital
– Project B: spending $175,000 on hospital
– Project C: spending $250,000 on hospital
• Ranks by voters (Note > means more preferred to )
– Voter X: A>B>C
– Voter Y: B>C>A
– Voter Z: C>A>B
Outcomes from pair wise comparison
-2/3 (Majority) preferred A to B
-2/3 (Majority) preferred B to C
-2/3 (Majority preferred C to A
• The majority voting results are not transitive. From the first two pair wise
comparisons, A>C but for 2/3 of the voters (Voters Y &3 Z) C>A. 8
 Public choice theory cont’d
• Cause of cyclical phenomenon
• Single peaked preferences: in this case the voters utility falls as they move in
any direction from their most preferred outcome/project
• Double peaked preference: voters utility falls when it moves away from the
most preferred outcome but then it moves again.
• Voters X & Y have single peaks whereas voter z has two peaks (double peak).
– Solution: imposing voter Z to have a single peak (C>B>A) will solve the
cyclical problem
– In that case project B will be the winner

9
 Public choice theory cont’d
Majority voting and the median voter theorem
• The median voter theorem states that as long as all
preferences are single-peaked, the outcome of majority
voting reflects the preferences of the median voter.
• Consider the 5 voters each with single-peaked preferences as
given by the following preferred expenditure level on health.
Voters Preferred expenditure in
millions Birr
Samson 20
Gebrekidan 40
Yonas 80
Chala 90
Mohammed 120
10
 Public choice theory cont’d
Median voter theorem illustrated
• A movement from 20ml to 40ml would be approved by four
voters (except by Samson).
• A movement from 40ml to 80ml would be approved by Yonas,
Chala, and Mohammed.
• A movement from 80ml to 90 ml would be blocked by majority
of voters (Samson, Gebrekedan, and Yonas).
• A movement from 90ml to 120ml would be blocked by a majority
of the voters (Samson, Gebrekidan, Yonas, and Chala)
• Movement from 80ml to 40 ml would be blocked by a majority of
the voters (Yonas, chala, and Mohammed).
• Yonas is the median voter. Hence, his choice is favored by the
majority or by [n/2]+1 of the voters.

11
 Public choice theory cont’d
p

20 40 80 90 120
Ordinal utility index

20 40 80 90 120 12
 Public choice theory cont’d
• The preference of the median voter is choose with an
expenditure level of 80ml. This level corresponds to the
peak of the aggregate ordinal utility index (vertical
summation of individual indices
• Implications of median voter theorem
– Politicians and political parties must identify the
median voter in order to secure the majority of the
votes
– Political competition between major parties will
produce consensus politics. i.e. each party has a
similar program.

13
 Public choice theory cont’d
Other forms of direct democracy
a. Pluralistic voting: In the case of plurality rule weight is assigned
in the range n, n-1, n-2, ….,1 for the best preference and the least
preference among n, alternative outcomes in descending order.
• Then the sum of the weights enables in ranking the projects
accordingly.
• Suppose there are three individuals X, Y and Z and there are five
alternative outcomes such as a, b, c, d and e.
• X: a > b > c > d > e
• Y: a > b > c > d > e
• Z: c >d > e > a > b
• Applying the majority rule the group choice will be a > b > c > d >
e and ‘Z’ will be the minority.

14
 Public choice theory cont’d
• In terms of plurality rule weight age will be assigned as follows

• It is evident from the above table that the group preference

will be a>c>b>d>e
• In comparison to majority rule outcome ‘c’ the best preference
of ‘Z’ (the minority) is getting importance in the plurality rule
as it is placed on the second place in the group preference.

15
 Public choice theory cont’d
b. Point Voting: In the case of point voting rule a fixed number of
points (say, 100) is allotted to each voter.
– The voter castes number of points in accordance with his
intensity of preference for different alternative outcomes
aggregating the number of votes to the fixed number as allotted.
– Let us assume that there are three individuals X, Y and Z and
there are three alternative outcome a, b and c
X: a > b > c
Y: b > c > a
Z: c > b > a
• Suppose each voter is allowed to allocate 100 votes among a, b and
c in accordance with his intensity of preference (see the next slide).

16
 Public choice theory cont’d

Voters A B C

X 50 30 20

Y 30 60 10

Z 5 30 65

85 120 95

The groups choice results in b>c>a

Limitations of point voting:
– It gives scope for strategic voting because casting of the number of
points for an outcome on the basis of the intensity of preferences is
not definite but subjective.
– It might also give an indeterminate result in the group preference
when alternative outcomes get equal number of votes in aggregate
17
 Public choice theory cont’d
c. Logrolling: means vote trading. Individual X supports
the preference of Y in one case provided that Y
supports the preference of X in the other case. It is
illustrated in the following example.
• Suppose the Government is making expenditure in the
Education sector and also in the Health sector. Education is
explained by the size of school (S) and Health is explained
by the size of Hospital (H).
• Let H1, H2, H3 represent low size, medium size and big size
of the hospital. The preference patterns of individual X, Y
and Z is given as follows:
X: H1>H2>H3
Y: H2>H3>H1
Z: H3>H2>H1
• Applying majority rule the group choice will be H2>H3>H1.
18
 Public choice theory cont’d
Similarly, for school sizes, the preferences are as follow
X: S2>S3>S1
Y: S3>S2>S1
Z: S1>S2>S3
• Applying majority rule the group choice will be S2>S3>S1. It is
obvious that following majority rule the final outcome will be (H2,
S2) in the group choice if there are no vote trading.
• Suppose there is vote trading between individual X and Y. Y
supports the preference of X for H1 and in exchange X supports
the preference of Y for S3. This gives the group choice finally
according to the majority rule and (H1, S3).
• If again there is logrolling between individual Z and Y, Z supports
the preference of Y for S3 and in exchange Y supports the
preference of Z for H3. This leads to final group choice (H3,S3)
according to the majority rule.
– In this case logrolling fails to derive a unique group choice. For the success
of logrolling an individual has to go for vote trading only with another
individual, not with more than one individual simultaneously. 19
 Public choice theory cont’d
Advantages of Log rolling: voters get their preferred
project
Limitations of log rolling:
• It results in too much government expenditure because
the trading could favor the most expensive ones
• It encourages strategic behavior on voters. Voters might
report incorrect preferences strategically.

20
 Problems in direct democracy
• Can any ethically acceptable method for translating individual
preferences into collective preferences be free of difficulties
discussed so far?
• Reasonable criterias for decision making rule
– I – Independence of new alternatives: adding new option
should not affect original ranking
– N – Nondictatorship: all voters are equal
– P – Pareto criterion: all rankings shoul be Pareto efficient
– U – Unrestricted rankings: any possible individual rankings
should be received. Eg. Double peaked preference should
be allowed
– T – Transitivity: A>B; B>C; then A>C!
• Arrow’s Impossibility Theorem states that it is
impossible to find a decision rule that satisfies the 5
criteria's. 21
 Representative democracy
• In reality, government doesn’t simply aggregate people’s
preferences; rather the governing is done by politicians,
judges, bureaucrats, and so on.
• To understand the realities of public choice, one must
study the goals and behaviors of the people who govern.
– Because these players pursue their own objective functions
(their preferences) or interests during decision making.
Elected Politicians: are vote maximizers in election
Downs (1957)-argued that a vote maximizing
politician adopts the preferred program of the
median voter.
– Candidates move to middle of spectrum, because
voters support candidate with view closest to own,
and only one wins.
22
 Representative democracy
• Suppose two politicians (A and B) with single peaked preference
and with political view measured along a single dimension
(liberal/conservative)

• Implication:
– 2 party systems tend to be “stable” in the sense that both stake
out positions near the “center.”
– Replacement of direct referenda (e.g., direct democracy) by a
representative system has no effect on outcome. Both mirror
the preferences of median voter.
23
 Representative democracy
• Limitations of Down’s model
– Single-dimensional ranking assumption: the beliefs of
political parties may not be measured along a single spectrum.
– Ideology matters: politicians care about more about their
ideologies than just maximizing votes.
– Personality: voters may care more about candidate’s
personality than his/her policy/agenda
– Leadership: the model assumes that politicians are passive. In
reality the politicians themselves could influence the preference
of voters.
– Voter participation: may be affected by relative difference in
candidates. Voters might fail to vote due to
• Close position of candidates
• High cost of acquiring information
• Free rider problem: with the view that single vote doesn’t matter
24
 Bureaucrats/Public employees/Cadres
Bureaucrats: government employees.
– The legislation enacted by politicians are vague
– The program of the politicians are precisely implemented
by the bureaucrats
– They provide valuable technical expertise in the design and
execution of the programs
– They provide documented government activities for
improved decision making
• However, they are not passive or they are not only aiming to
fulfill the wishes of the electorate and its representatives.
• Niskanen (1971) argues that bureaucrats tend to focus on
maximizing perquisites of public office, public reputation,
power, etc., because opportunities for monetary gains are
minimal.

25
 Bureaucrats/Public employees/Cadres
Bureaucrats and allocative inefficiency-The Niskanen
Model
• Niskanen (1971) argues that bureaucrats maximize their
power, status, prestiges etc. correlated to the size of the
bureau/budget they run. The implications of this are:
– Bureaucrats exaggerate the benefits provided by their
bureau/department in order to get bigger budget
– They also tend to understate the costs (direct and indirect)
• Thus, bureaucrats have a tendancy to expand the output of
the public sector to twice the level of output that would be
regarded as socially optimum.
• Eg TB=aQ-bQ2 and TC=cQ+dQ2, a>0, b>0, c>0,d>0
Find
a. Profit maximizing level of output
b. Social optimum level of output
c. Output supplied by bureaucrats
26
 Bureaucrats/Public employees/Cadres cont’d
Marginal valuation=dTB/dQ=a-2bQ=Demand curve
Marginal cost=dTC/dQ=c+2dQ
TR=P.AR=aQ-2bQ2
MR=dTR/dQ=a-4bQ
Solutions
a. MC=MR: c+2dQ=a-4bQ QP*=(a-c)/[2(b+2d)]
b. MC=MB: c+2dQ=a-2bQ  QS*=(a-c)/[2(b+d)]
c. TB=TC: aQ-bQ2=cQ+dQ2  QB*=(a-c)/[(b+d)]
QB*=2QS* or (a-c)/[(b+d)]=2{(a-c)/[2(b+d)]}
• Bureaucrats produce an output larger than a socially
desirable level and causes too much government spending.
• Project doesn’t suffer losses, but is inefficient.
27
Bureaucrats/Public employees/Cadres cont’d

28
 Special Interest Groups
• Special interest groups can form coalitions and exercise
a disproportionate amount of power if they vote in blocks
or make campaign contributions.
• Interest groups formed based on many factors, including
rich vs. poor, gender (eg. Women and youth associations),
industries (eg. MEDROC), and demographics & ethnicity,
religion (Muslim, orthodox, protestant).
• Rent-seeking or transfer seeking by special groups
– when political links are used to obtain abnormal returns
(“rents”) from government services. This rent is political rent
and it is not economic rent
– Offering and making payments to office seekers and holders
and their political parties in order to obtain privileges such as
transfers of funds and protection from competition.
29
 Special Interest Groups
• Allegations of some pro-EPRDF businesses buying formerly state
businesses at a very low price in the name of privatization.

• The transfer seekers getting a lot.

• The costs to the public are thinly spread.
• Eg an import lobbying industry

30
 Special Interest Groups
• The competitive output would be at Qc.
• The company receiving the benefit could try to form
cartel to restrict output to Qcartel,
• Standard deadweight loss from reduced output is equal to
triangle cde.
• To the extent that the economic rents, abcd, are spent on
socially wasteful lobbying (rather than being a transfer to
producers), this is also considered deadweight loss.

31
 Iron Triangle
The “Iron Triangle”: is a model of the policy-making process that
emphasizes the mutually beneficial roles played by politicians,
interest Groups, & bureaucrats.
•Each of the players in the triangle influences, and is influenced by,
each of the other players.
The links are as follows (6 links)

1. Interest groups lobby politicians for favorable policy.

Politicians often listen, because interest groups provide campaign
funds and political advertising that can influence voters at election
time.
2. In return for their support, members of the parliament can provide
interest groups with friendly legislation. They can also, through
their oversight function, put pressure on bureaucrats to implement
programs and policies in a way that interest groups favor.

32
 Iron Triangle
3. Members of the parliament can influence the bureaucrats
through annual budgetary decisions.

3. Bureaucrats can choose to execute policies in a way that is

favorable to members of parliament who control their
budgets and oversee their operations.

5. Interest groups will often lobby bureaucrats directly for

favorable implementation of policies that affect their
members. Interest groups can also rally public support or
disfavor for bureaucrats and their agencies based on their
performance.

6. Bureaucrats can implement policies and programs in ways

that please interest groups to please the parlament
members.
33
 The Iron Triangle

Bureaucrats
3
4
5 6

Parliament members
Interest groups
1
2

• The iron triangle

34
 Issue Network
• The “Iron Triangle” model probably best illustrates how policies
are created and carried out but the public is largely ignorant.
• The Issue Network model assumes that an issue has been
brought to the attention of the public and that policy-
formulation is taking place in the public forum.
• These groups will try to influence public perceptions of the issue
by advertising their point of view to the public through mass
emails, TV and radio ads, sending their spokesmen out to appear
on political talk shows, etc..
• Policy experts at university think tanks and independent policy
institutes conduct studies on various policy alternatives and
report their findings to both the public and the parliament.
• Although these policy experts will often claim to be neutrals, in
reality, the results of their studies will often be biased.
• Interest groups and members of parliament often use the
findings of such policy experts to support their positions on the
issue at hand.
35
 Issue Network
Chart of the issue network

Interest groups

Policy experts like

Bureaucratic university think
Mass media
agency tanks, policy
research institutes

Members of the
parliament
36
 Theories of Government expenditure growth
• Public expenditure includes all government final consumption,
investment, and transfer payments on collective needs and wants such as
pension, public education, public health, roads, defense, and so on.
– General government final consumption expenditure includes all government
current expenditures for purchases of goods and services (including
compensation of employees) including defense spending.
– Public investment: includes the investment component of public expenditures
• Size of the public sector (as measured by GE/GDP) has
expanded substantially over time in almost all countries.
– For example it increased from 10% in 1929 to 35% in 2018 and for
UK it increased from 14.4% in 1900 to 38% in 2018
– Ethiopia’s GE as a share of GDP increased from 16.6% in 2012 to
18.3% in 2016
• The size of government during Derg regime was very large as the country
was under socialist system.

37
 Theories of Government expenditure growth
Determinants of the size of public sector
i. Perceived nature and extent of market failures: larger government
failures implies larger need of gov’t intervention.
ii. Public choice models pursued like logrolling, bureaucratic
process, and interest groups
iii. The fiscal process: in some cases there are hidden or disguised
taxes burdens which reduces the perceived costs of providing
public goods & services  raise demand  raise size of gov’t
spending

38
 Theories of Government expenditure growth
Methods of controlling size of government
When the size of government expenditure is perceived to have flaws
in the political process, bringing it under control is not easy. The
following are the suggested methods for this purpose
• Changing bureaucratic incentives - for example payments should
not be tied with size of the department
• Increasing the role of the private sector in the provision of public
sector goods and services
• Improving the discipline of the budget process.
– Budget process involves planning and forecasting implementation,
monitoring and controlling, and finally evaluating the performance of the
budget
• Introducing constitutional rules on budgetary process such as
fixing per diem rates
39
 Theories of Government Exp. growth cont’d
i. Citizens Preference or Wagner's Law/model
• Wagner, in his classic book, formulated a law of expanding state
activity.
• He asserted that there is a long run propensity for the scope of
government to increase with higher levels of economic
development due to the pressure of social progress. Wagner based
this generalization on two considerations
a) The income elasticity of demand for services provided by the
government (such as education and health) is high (larger than 1).
In other words, demand for public goods and services increases as
income increases.
b) The low price elasticity demand for public goods. Even if price of
such goods and service increases, demand do not decrease because
in the public sector there are limited research and developments to
reduce the cost of public goods (Baumol effect).
– The two effects pushes government expenditure over time.
40
 Theories of Government Exp. growth cont’d
ii. Marxist view: argue that the political economic system if the basis
for government expenditure growth
• The private sector over produces (such as armaments/tanks, buses
etc) and the public sector has to expand to absorb the excess
production (probably due to lobbying)
iii. Displacement Hypothesis (Peacock and Wiseman-1967)
• In normal periods there is a moderate growth in government
expenditure.
• Occasionally when an economic shock occurs, there will be a
demand for higher levels of government expenditure and novel
methods of financing.
• Even after the shock disappear, higher levels of government
spending continues due to inertia. GE

Years 41
 Theories of Government Exp. growth cont’d
iv. Changing Social attitudes
• Social attitudes have encouraged government expenditure growth
• Social trends encouraging personal self assertiveness lead people to
make an extravagant demands on the political system
• At the same time widespread television advertising causing people to
think their government programs do not have an opportunity cost
iv. Income distribution: the poor could use the political system to
redistribute income to themselves
• Suppose there are 5 individuals with incomes of $5000, $10000,
$15000, $20,000, and $40,000.
• The median voter is 15,0000. So a politician who supports
government programs that transfer income to those with less than
20,000$ will win the vote
– Government spending increases rapidly.
42
 Theories of Government Exp. growth cont’d
v. Development model (by Musgrave and Rostow)
– The economist, Musgrave, and the economic historian,
Rostow, (separately) suggested that the growth of public
expenditure might be related to the pattern of economic
growth and development in societies.
– The development models traces the development process
from industrialization.
– At early stages there are increases in externalities and like
pollution and crimes. Increasing proportion of expenditures
diverted towards control of externalities and crimes.
• At higher stages of development: high demand for private
goods and high demand for complementary public goods.
– Expenditure driven by the desire to react to issues such of
equity and transfer payments constitute a significant
proportion of government spending. 43