Transitive Vote Delegation in Social Networks

16 concepts8 papers3 proposals

In a traditional direct democracy, every citizen votes on every issue, which often introduces significant noise due to varying levels of voter information and low participation rates. Conversely, representative democracy mitigates this noise by electing centralized agents, but at the cost of individual granularity and flexibility.

Liquid democracy emerges as a dynamic middle ground. It allows voters to either cast a ballot directly or transitively delegate their voting power to a peer. If Voter A delegates to Voter B, and Voter B delegates to Voter C, Voter C wields the accumulated voting weight of all three. By allowing votes to flow recursively through a social network, this mechanism aims to organically concentrate decision-making power in the hands of well-informed experts.

[Voter A] ──> [Voter B] ──> [Voter C (Guru)]  <── [Voter D]
  (Wt: 0)       (Wt: 0)        (Wt: 3)              (Wt: 0)

The Intuition: Routing Weight Along Competence Gradients

The fundamental promise of transitive delegation is its ability to aggregate expertise dynamically. Instead of requiring a centralized authority to identify experts, the system relies on local, decentralized choices.

As explored by Shiri Alouf-Heffetz et al., when individuals pass their voting weight to neighbors who possess higher relative competence, the system iteratively pushes influence "uphill" toward expertise. This competence-based delegation filters out the noise of uncalculated guesses without requiring any single voter to have a global view of who the most qualified experts are.

When inactive or less-informed citizens pass their ballots to the "nearest" active, informed voter, the collective decision-making process improves. Gal Cohensius et al. demonstrated that this weighting of representative opinions overcomes the small-sample participation problem inherent in direct voting, accelerating the reduction of collective error quadratically faster than traditional direct democracy.

The Pathological Concentration of Power

Despite its theoretical appeal, transitive delegation introduces a structural vulnerability: the emergence of "super-voters." Because delegation chains can grow arbitrarily long, a few highly visible individuals can accumulate massive amounts of voting weight.

[Voter 1] ──┐
[Voter 2] ──┼─> [Super-Voter] (Massive Weight)
[Voter 3] ──┤
[Voter 4] ──┘

This phenomenon creates a severe mathematical paradox. Anson Kahng et al. proved that simple, localized delegation rules can actually make collective decisions less accurate than direct democracy. If a super-voter with a vast pool of accumulated votes happens to make an incorrect choice, they can single-handedly drag down the accuracy of the entire system. The authors showed that avoiding this pitfall requires non-local mechanisms, such as centrally capping the maximum weight any single proxy can hold.

Furthermore, evaluating the optimal way to delegate is not trivial. Ioannis Caragiannis et al. provided a rigorous critique showing that finding an optimal delegation strategy across a network is an NP-hard problem. They demonstrated that local heuristics can behave counterintuitively, to the extent that delegating to a less informed neighbor can sometimes yield a better collective outcome than delegating to a more informed one, depending on how the broader network structure aggregates terminal nodes.

Quantifying Influence in Multi-Step Chains

Traditional voting systems rely on classic power indices (like the Banzhaf index) to measure a voter's capacity to alter an election outcome. These traditional metrics assume flat, independent voting structures where each agent directly casts a ballot. In a fluid delegation network, however, these tools break down because they fail to account for the structural influence of intermediaries who act as conduits in multi-step chains.

To address this gap, researchers have developed new mathematical frameworks to quantify influence within delegation graphs:

  • The Delegative Banzhaf Index: Introduced by Yuzhe Zhang 0001 et al., this index measures the structural power of both terminal voters and intermediate proxies. By modeling voters as strategic agents, the authors uncovered a surprising behavioral paradox: when individuals act out of a desire for personal power ("power-greedy" agents), the resulting network dynamics actually yield a more egalitarian distribution of power than when agents act purely altruistically.
  • I-Voting Power: Developed by Rachael Colley et al., this metric focuses on a voter's ability to influence the final outcome based on their specific stance rather than just their presence in a winning coalition. The authors established that a voter's in-degree—the number of peers who can directly or indirectly delegate to them—serves as a primary driver of voting power. While calculating this power is generally #P-hard, it can be computed efficiently in specific network topologies like bipartite or complete graphs.

This modern focus on network-dependent influence builds upon foundational probability theory. Olle Häggström et al. previously extended the law of large numbers to non-independent voting scenarios by defining a voter's "effect"—the expected difference in election outcomes when a single voter's choice is toggled between choices. In weighted majority systems, keeping individual effects small ensures that the collective decision reliably tracks the underlying ground truth, a principle that directly informs how delegation weights must be managed to prevent systemic instability.

Strategic Stability and Network Topology

Because liquid democracy relies on voluntary choices, the stability of the entire system depends on the incentives of individual actors. Voters constantly balance a game-theoretic trade-off: the cognitive effort required to acquire information and vote accurately versus the utility of "borrowing" a peer's expertise through delegation.

Daan Bloembergen et al. formalized this dynamic, proving that stable delegation patterns (Nash equilibria) exist under various conditions. Crucially, they demonstrated that the underlying topology of the social network—whether it exhibits "small-world" properties or follows a "scale-free" distribution—profoundly dictates whether rational individual choices will converge into an accurate collective decision. The architecture of the social graph determines whether transitive delegation successfully channels wisdom upward, or collapses under the weight of its own centralized hubs.

Key Foundational Work

A law of large numbers for weighted majority (2006) — contributed 2 concept elements to this direction.
Proxy Voting for Better Outcomes (2016) — contributed 1 concept elements to this direction.
History & citationVersion 1 · 2026-10-04

Current · Permanent link

View frozen version 1

Cite this direction

Latest version (a living page):

@misc{lacuna_transitive_vote_delegation_1d902e58,
  title = {transitive vote delegation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/transitive-vote-delegation/rdr_1d902e5841544f6aa541f3e27d20a0f3},
  note = {Latest version of a living research direction; cite a version for a fixed text. Accessed 2026-11-19},
}

This exact version (version 1, frozen text):

@misc{lacuna_transitive_vote_delegation_1d902e58_v1,
  title = {transitive vote delegation},
  author = {{Lacuna}},
  howpublished = {Lacuna research direction},
  year = {2026},
  url = {https://lacuna.tiptreesystems.com/direction/transitive-vote-delegation/rdr_1d902e5841544f6aa541f3e27d20a0f3/v/1},
  note = {Version 1, frozen 2026-10-04. Accessed 2026-11-19},
}
About

These directions overlap with this one through shared papers.

See something wrong? Report a mistake