CHAPTER 4 - CANDIDATE
OBJECTIVES AND
ELECTORAL EQUILIBRIUM
By: Angel Me S.
Balondo
INTRODUCTIO
N
• Elections are central to representative democracy,
serving as the mechanism for choosing political
agents.
• Formal analysis of elections began with Hotelling
(1929) and Downs (1957).
• Focus: Spatial models of elections where
candidates adopt policy platforms and are
committed to them post-election.
• Abstracts complexities like primary elections,
ELECTORAL
FRAMEWORK
Model: Spatial model of politics (Davis, Hinich,
Ordeshook).
Assumptions:
• Policy space is Euclidean and compact.
• Two candidates announce policy positions; winner is
committed to promises.
• Voters (odd number, no abstention) cast ballots;
majority wins.
Two candidates (A & B) announce platforms
simultaneously in a defined policy space.
CANDIDATE
OBJECTIVES
[Link] Motivation:
• Candidates value victory (utility = 1 for
winning, 0 otherwise).
2. Vote Motivation:
• Candidates aim to maximize vote share
(utility = number of votes).
3. Office Motivation:
• Candidates are motivated by either win or
vote maximization.
VOTING
(a) Deterministic Voting (Downsian Model):
• Voters choose MODELS
candidates offering the closest policy
to their ideal.
• Median Voter Theorem (unidimensional case): Both
candidates adopt the median voter’s ideal policy.
• Multidimensional case: Equilibrium rarely exists due
to "radial symmetry" constraints.
(b) Probabilistic Voting:
• Voter behavior includes biases (e.g., partisanship).
• Introduces smoother candidate utility functions,
DOWNSIAN MODEL AND
MULTIDIMENSIONAL POLICY
SPACES
• Duggan and Fey (2005a) establish that in
a multidimensional policy space,
equilibria are rare and fragile.
• For equilibrium to exist where candidates'
gradients differ, their positions must
coincide (i.e., x=x∗x=x∗), and specific
voter gradients must be diametrically
opposed.
PROBABILISTIC VOTING AND
THE STOCHASTIC
PARTISANSHIP
Relaxation of Deterministic Voting:
• Voter behavior is treated MODEL
as probabilistic, reflecting candidates' uncertainty about voter preferences.
• Candidates have probabilistic beliefs about how voters choose based on policy and partisan
preferences.
Key Theorems:
• Theorem 4 (Hinich; Lindbeck and Weibull; Banks and Duggan):
⚬ In equilibrium, both candidates adopt the "utilitarian point" (the mean of voters' ideal policies).
⚬ Applies irrespective of the dimensionality of the policy space.
⚬ Quasi-concavity ensures equilibrium existence under specific conditions (e.g., uniform bias
distribution).
• Theorem 5 (Hinich, Ledyard, Ordeshook; Lindbeck and Weibull):
⚬ Conditions such as voter utility concavity and bias convexity guarantee the existence of
equilibrium.
Comparison with Downsian Model:
• Introducing even small probabilistic noise in voting behavior alters equilibria significantly.
WIN MOTIVATION VS. VOTE
Win Motivation MOTIVATION
(Duggan 2000a):
• Candidates aim to maximize their probability of
winning.
• Theorem 6: In equilibrium, candidates adopt the
utilitarian point under symmetric conditions and
convexity assumptions.
• Figure 4.4 illustrates how these conditions ensure
equilibrium through convex sets of achievable
probabilities.
Key Insight:
OVERALL IMPLICATIONS
• Multidimensional policy spaces complicate
equilibrium existence in deterministic
models.
• Probabilistic frameworks offer solutions but
introduce their challenges, such as reliance
on specific assumptions.
• The central tendency of candidates
(adopting the utilitarian point) reflects the
strong pull of median voter-like incentives,
CONCLUSION
• Equilibrium existence remains challenging in
multidimensional spaces.
• Probabilistic models offer insights but
require restrictive conditions.
• Dynamic and citizen-candidate models
present alternative approaches.
• Contributions blend theoretical foundations
with practical electoral insights.
THANK
YOU!
