Condorcet's Jury Theorem with Abstention

Overview

• Condorcet's Jury theorem states that as the population size increases, the majority rule will select the best alternative between two options with probability 1.
• This study examines this result under an asymmetric two-candidate setup, where supporters of both candidates have different participation costs.
• The paper proposes and analyzes a boundedly rational model where voters overestimate their influence on the outcome, and shows that this can lead to non-trivial equilibria where the winning probability of both candidates is bounded away from 1.

Plain English Explanation

In a typical election, voters have to decide between two candidates. Condorcet's Jury theorem suggests that as the number of voters increases, the majority rule will always select the better of the two candidates.

However, this paper looks at a more nuanced scenario. Imagine that supporters of the two candidates have different costs associated with voting. For example, it may be more convenient for some voters to get to the polling station than others.

When voters rationally consider their chance of being the deciding vote (called "pivotality"), the paper shows that the only possible outcome is a trivial one where everyone abstains except those with zero voting cost.

The paper proposes a more realistic model where voters overestimate their influence on the outcome. Under this model, the paper demonstrates that non-trivial equilibria can emerge, where neither candidate is assured of winning, even with a large number of voters.

The key insight is that when voters' perception of their pivotality strongly depends on the margin of victory, neither candidate is guaranteed to win, in contrast to Condorcet's assertion. Only when this perception is weakly dependent on the margin does Condorcet's Jury theorem hold.

Technical Explanation

The paper examines Condorcet's Jury theorem, which states that as the population size increases, the majority rule will select the best alternative between two options with probability 1.

The study looks at an asymmetric two-candidate setup, where supporters of the two candidates may have different participation costs. When the decision to abstain is fully rational, based on the probability of being the deciding vote (pivotality), the paper shows that the only possible equilibrium is a trivial one where all voters except those with zero voting cost abstain.

To address this, the authors propose a boundedly rational model where voters overestimate their pivotality. Under this model, the paper demonstrates that non-trivial equilibria can emerge, where the winning probability of both candidates is bounded away from 1, in contrast to Condorcet's assertion.

The key finding is that when voters' perception of their pivotality strongly depends on the margin of victory, neither candidate is guaranteed to win, regardless of population size. Only when this perception is weakly dependent on the margin does Condorcet's Jury theorem hold.

Critical Analysis

The paper provides an interesting alternative to the classic Condorcet's Jury theorem, by incorporating asymmetric participation costs and a more realistic model of voter behavior. The boundedly rational model where voters overestimate their pivotality is a plausible assumption and leads to more nuanced outcomes than the fully rational case.

One potential limitation is that the paper focuses on a two-candidate scenario. It would be valuable to explore how the results extend to multi-candidate elections, which are more common in practice. Additionally, the paper does not address factors like strategic voting or information asymmetries, which could also impact the equilibrium outcomes.

Further research could investigate the sensitivity of the results to the specific functional forms and parameters used in the model. Empirical studies validating the model's predictions would also strengthen the conclusions.

Overall, this paper challenges the conventional wisdom of Condorcet's Jury theorem and offers a more realistic perspective on the dynamics of electoral competition. It encourages readers to think critically about the assumptions underlying theoretical models of voting behavior.

Conclusion

This paper provides a thoughtful critique of Condorcet's Jury theorem, a well-known result in the field of voting theory. By introducing asymmetric participation costs and a boundedly rational model of voter behavior, the study demonstrates that the classic theorem may not hold in more realistic scenarios.

The key insight is that when voters' perception of their pivotality is strongly tied to the margin of victory, neither candidate is assured of winning, even with a large population. This contrasts with Condorcet's assertion that the majority rule will select the better candidate with probability 1 as the population size increases.

This work highlights the importance of incorporating realistic assumptions into theoretical models of voting and political competition. It encourages further exploration of alternative frameworks that can better capture the complexities of human decision-making in the electoral process.