The Equilibrium Effects of Costly Participation in Elections

38 concepts13 papers3 proposals

In classical voting theory, casting a ballot, running for office, or expressing an opinion is often treated as a frictionless, costless act. This assumption simplifies mathematical models but introduces a glaring flaw known as the "lazy voter" problem: when participation is free, agents will vote or act even if their input has zero impact on the final outcome. This yields highly unrealistic Nash equilibria populated by inconsequential actions.

To bridge this gap, a growing body of research integrates a simple but powerful mechanism: participation costs. By treating participation as an investment of effort, time, or capital, these models introduce a pivotal filter. Agents rationally choose to step forward only if the expected utility of changing the outcome outweighs the personal cost of showing up.

The Pivotal Filter and Equilibrium Instability

How does a small cost change the strategic landscape? In a simultaneous plurality vote, if voting is free, everyone might as well vote. But as Desmedt demonstrates in Equilibria of Plurality Voting with Abstentions, adding even a minor ballot cost filters out non-pivotal behavior. Voters rationally choose strategic abstention if their preferred candidate is already winning or completely unviable. While this refines the equilibrium set to reflect meaningful choices, it also introduces instability: in simultaneous voting games, stable equilibria may disappear entirely unless the electorate is close to a near-consensus.

This instability extends from voters to the candidates themselves. Traditional models assume political candidates are purely power-hungry and will run under any circumstances. However, Obraztsova et al. introduce "lazy candidates" in Strategic Candidacy Games with Lazy Candidates, modeling the reality that campaigning incurs significant financial and reputational costs. A lazy candidate prefers to stay home unless their entry actively alters the election winner.

The authors show that while every stable equilibrium in a costly game is also an equilibrium in a cost-free game, the reverse is not true; adding costs collapses many traditional stable states. When a redundant candidate withdraws to save on costs, it shifts the vote distribution, which can trigger a cascading cycle of withdrawals and entries. In later work, Obraztsova et al. expand this to "reluctant leaders" in Strategic candidacy with reluctance, proving that determining whether a stable election outcome even exists under these realistic constraints becomes an NP-complete computational problem. Similarly, Vasselai argues in Iterative Calculus of Voting under Plurality that dynamic models must incorporate these strategizing costs to accurately map real-world voter turnout and stability.

Why Abstention Reduces Distortion

Intuitively, one might think that when voters stay home, the democratic quality of an election degrades. If a large portion of the electorate abstains, doesn't the outcome become unrepresentative?

Remarkably, game-theoretic modeling reveals the exact opposite. In social choice theory, "distortion" measures how far an elected winner deviates from the socially optimal choice (the candidate minimizing total social cost). In a standard model where everyone is forced to vote, the worst-case distortion is a well-known constant of 3. However, in a series of papers, Ghodsi et al. (On the Distortion Value of the Elections with Abstention) and Ghodsi et al. (On the Distortion Value of the Elections with Abstention) prove that when you allow for realistic abstention, the worst-case distortion drops dramatically to approximately 1.5.

Why does this happen? The authors, alongside Seddighin et al. in On the Distortion Value of Elections with Abstention, formalize this by modeling two types of abstention:

  • Indifference: when candidates are too similar, meaning the voter has no strong preference.
  • Alienation: when all candidates are too ideologically distant from the voter.

When participation is sensitive to these factors, highly unmotivated or centrist-indifferent voters drop out. This effectively filters out the "noise" of voters who do not strongly care, ensuring that the final tally is driven by those with strong, well-defined preferences. This behavioral nuance prevents the worst-case, highly distorted outcomes that occur when indifferent voters randomly tip the scales. Acknowledging this, Bhaskar et al. note in Nearly Tight Bounds on Approximate Equilibria in Spatial Competition on the Line that moving away from the classic assumption of inelastic, mandatory voting toward models incorporating voter apathy remains a vital frontier for spatial competition.

Information Costs and the Breakdown of Collective Wisdom

Participation costs are not always physical or financial; they are often cognitive. The famous Condorcet Jury Theorem states that if each individual is even slightly more likely to be right than wrong, the "wisdom of crowds" ensures that a larger group will almost certainly make the correct collective decision.

But what if becoming informed requires effort? Michelini et al. explore this in Group Wisdom at a Price: Jury Theorems with Costly Information. When acquiring information incurs a cost, individuals face a dilemma: expend effort to become competent, or abstain and free-ride on others. Because an individual only wants to pay the information cost if they believe their vote will be pivotal, the incentive to participate actually shrinks as the group grows larger. Consequently, the wisdom of crowds breaks down, capping the effective size of the informed voting pool.

This cognitive bottleneck is precisely why alternative democratic structures like liquid democracy and proxy voting have emerged. In liquid democracy, voters can either vote directly or delegate their voting power to a trusted peer. Bloembergen et al. analyze this in On Rational Delegations in Liquid Democracy, showing that delegation is a rational mechanism to bypass the personal cost of information acquisition by "borrowing" an expert's labor.

However, delegation itself isn't free. As Alouf-Heffetz et al. demonstrate in The Cost Perspective of Liquid Democracy: Feasibility and Control, finding an optimal, representative delegation network under strict individual cognitive budgets introduces severe computational constraints. When designed correctly, however, endogenizing these participation choices can rescue collective accuracy. Cohensius et al. show in Proxy Voting for Better Outcomes that proxy voting—where inactive citizens automatically delegate to the nearest active voter—overcomes the noise of low-sample participation, driving down decision errors quadratically faster than traditional systems.

Designing Mechanisms Defiant to Manipulation

Finally, introducing costs can serve as an intentional tool for mechanism designers to prevent manipulation. In open online voting environments, systems are highly vulnerable to Sybil attacks, where a single malicious actor creates multiple false identities to swing an election.

To combat this, Wagman et al. investigate mechanism design in Optimal False-Name-Proof Voting Rules with Costly Voting. While traditional theory suggested that making a voting rule completely "false-name-proof" required ignoring voter preferences almost entirely, the authors prove that introducing even a tiny cost for acquiring additional identities (such as solving CAPTCHAs) changes everything. This marginal cost acts as a barrier that penalizes malicious duplication while allowing the voting rule to remain highly responsive to the genuine majority.

Even outside formal voting, costs shape expression: Huberman shows in Opinion formation under costly expression that when expressing an opinion online is costly, later participants naturally moderate the prevailing sentiment to appear discerning, fundamentally altering how collective consensus evolves over time.

Key Foundational Work

Equilibria of Plurality Voting with Abstentions (2010) — contributed 3 concept elements to this direction.
Optimal False-Name-Proof Voting Rules with Costly Voting (2008) — contributed 2 concept elements to this direction.
Opinion formation under costly expression (2010) — contributed 1 concept elements to this direction.
History & citationVersion 2 · 2026-10-04

Current · Permanent link

  1. Version 2 (latest) · 2026-10-04 · 14 papers
  2. Version 1 · 2026-10-04 · 10 papers
Cite this direction

Latest version (a living page):

@misc{lacuna_voter_abstention_modeling_a357dd74,
  title = {voter abstention modeling},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/voter-abstention-modeling/rdr_a357dd74698c4aa7956699191e302d83},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 2, frozen text):

@misc{lacuna_voter_abstention_modeling_a357dd74_v2,
  title = {voter abstention modeling},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/voter-abstention-modeling/rdr_a357dd74698c4aa7956699191e302d83/v/2},
  note = {Version 2, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake