Political Economy

Preference Revelation

Need to be able to elicit truthful revelation of preferences

  • Achieved through price mechanism in competitive private markets
  • However, in many contexts (e.g. public goods), this is infeasible

If one knows how much people value a public good, one can implement Lindahl tax. However, people may under-report their value because of the free-rider problem

Certain mechanisms (e.g. Clark-Groves mechanisms) can help by "internalising" the externality implicit in free-rider problem via side-payments

Gibbard-Satterthwaite Theorem
For any voting rule, one of the following must hold:

  • Only 2 options are considered
  • Voters will not truthfully state their preferences
  • Rule is dictatorial

Aggregating Preferences

Consistent aggregation needs to fulfill 3 conditions:

  • Pareto criterion: If all voters prefer A to B, social choice rule should choose A over B
  • Transitivity: If A is preferred to B, and B is preferred to C, then aggregation must be such that A is also preferred to C
  • Independence of Irrelevant Alternatives: If A is preferred to B, and C is added to the choice set, then A should still be preferred to B

Gibbard-Satterthwaite: Even before we confront preference aggregation, voting rules are subject to preference revelation concerns.

Condorcet Paradox: Majority voting does not lead to stable outcome.

  • Through majority outcome, we can get a majority cycle
  • We cannot find a stable equilibrium because we have non-single-peaked preferences, and thus, there is not one level that is preferred to all other levels

Single-peaked preferences
A group of agents is said to have single-peaked preferences if:

  • Each agent has an ideal choice in the set
  • For each agent, outcomes that are further from his ideal choice are preferred less.

With single-peaked preferences, you will select the median quantity for an outcome.

Arrow's impossibility theorem: The only choice rule that satisfies the Pareto criterion, transitivity, and IIA, is dictatorship


2 key assumptions to ensure existence of equilibrium

  • G is unidimensional (only choose between 2 options)
  • Preferences over G are single-peaked

Median voter theorem: With single-peaked preferences, majority voting produces equilibrium.

  • Equilibrium outcome is the median of the distribution of preferred G.
  • However, this equilibrium may not be Pareto efficient, i.e. it does not satisfy the Samuelson condition.

Example: Taxation and transfer systems

Theoretically, a poorer voter will prefer a higher tax rate.

  • If the median voter is "poor", then we will have a tax and transfer system.
  • Thus, if we have high inequality, we should end up with more redistribution and vice versa.

Support

  • Voting has become more inclusive over the years, changing the median voter
  • Median voter is now likely to be poorer and more disadvantage
  • This has led to more redistribution and thus we see a higher tax:GDP ratio

Contradiction

  • Theoretically, countries with higher inequality should have more redistribution, leading to higher tax:GDP ratio
  • However, Sweden has higher tax:GDP ratio compared to US/UK despite having lower inequality

Corruption

Olken (2006), Betrand et al (2007), Ferraz & Finan (2008)

Electoral Accountability

  • Theoretically, electoral accountability can reduce corruption
  • However, it depends a lot on the institutions and on politicians' motivations and incentives structures
  • Also, if people stay in office for too long, they can become more knowledgeable and work around the system

Special interest politics

  • Politicians can have the incentive to implement policies in which they have special interests (e.g. lobbying)
  • Many policies create concentrated benefits for a few well-defined groups, and with cost diffused in society at large

Scope of Government

  • Incomplete contracts are prevalent
  • Ownership is defined by residual right of control: who gains the benefits/bears the costs of actions taken outside scope of a contract

New Contract Theory

Emphasises importance of ownership for incentives to invest:

  • To reduce costs
  • To improve quality and innovate

Problem lies in non-contractible quality

Case for privatisation is stronger when:

  • Quality-reducing cost reductions can be controlled by contract
  • Innovation is important and competition effective
  • Government suffers from patronage and inefficiency
  • E.g. mail services

Case for privatisation is weaker when:

  • Cost reductions likely to lead to a non-contractible deterioration of quality
  • Innovation is relatively unimportant. Competition is weak and consumer choice is ineffective
  • Reputational mechanisms are ineffective
  • E.g. police

Hart, Shleifer & Vishny (1997) has a good chart on privatisation. Look at lecture notes.