Futarchy in DAOs: Governance by Prediction Markets
In 2000, economist Robin Hanson proposed a governance system that has haunted political theorists and crypto governance designers ever since. His idea was deceptively simple: vote on values, bet on beliefs. A society would democratically select its goals — maximise GDP, minimise carbon emissions, increase life expectancy — and then use prediction markets to determine which policies would best achieve those goals. The policies favoured by the market would be implemented. The policies disfavoured would be discarded.
Hanson called this system futarchy, and for two decades it remained an intellectual curiosity — theoretically compelling but practically impossible to implement. Blockchain technology and DAO governance have changed that equation. On-chain prediction markets can be deployed permissionlessly, governance tokens provide natural collateral for market positions, and smart contracts can automatically execute the policies that markets endorse.
The Theoretical Foundation
Futarchy rests on two intellectual pillars: the efficient market hypothesis and the separation of values from beliefs.
The efficient market hypothesis holds that market prices aggregate dispersed information more accurately than any individual or committee. If thousands of participants trade on the question “Will Policy A increase protocol revenue by more than Policy B?”, the resulting market prices reflect the collective judgement of everyone with relevant information, expertise, or insight. Markets are not infallible, but they are demonstrably better than alternative aggregation mechanisms across a wide range of domains.
The separation of values from beliefs addresses a fundamental limitation of democratic voting. When voters choose between policies, they are simultaneously expressing what they want (values) and what they think will work (beliefs). These two dimensions are entangled in a standard vote, making it impossible to distinguish between a proposal that failed because voters disagreed with its goals and one that failed because voters doubted its efficacy.
Futarchy separates these dimensions. The community votes on objectives — what success looks like. Markets then determine the means — which proposals are most likely to achieve the defined objectives. This separation allows each mechanism to do what it does best: democracy for normative questions, markets for empirical ones.
How Futarchy Works in Practice
A futarchic governance process follows a structured sequence.
Step 1: Define the objective metric. The community selects a measurable outcome that proposals will be evaluated against. For a DeFi protocol, this might be total value locked, revenue, or a composite governance health score. For a public goods DAO, it might be a measure of community impact or user growth.
Step 2: Submit proposals. Participants submit alternative governance actions — different parameter configurations, treasury allocations, or strategic initiatives.
Step 3: Open conditional prediction markets. For each proposal, a prediction market is created that trades on the expected value of the objective metric conditional on that proposal being implemented. If two proposals are competing, there are two markets: “What will TVL be if Proposal A is implemented?” and “What will TVL be if Proposal B is implemented?”
Step 4: Market resolution. After a trading period, the proposal whose conditional market predicts the highest value for the objective metric is selected for implementation. The winning market is settled at the actual outcome after implementation. The losing market is voided — participants get their collateral back regardless of their position.
Step 5: Execution and measurement. The selected proposal is implemented, and after a defined measurement period, the actual value of the objective metric is determined. The winning market is settled based on this actual outcome.
DAO Implementations
Several projects have brought futarchic principles into operational DAO governance, with varying degrees of fidelity to Hanson’s original conception.
MetaDAO on Solana has built the most faithful implementation of futarchy for DAO governance. Proposals are evaluated through conditional prediction markets that trade on the expected impact on the protocol’s token price. The system has processed real governance decisions, using market-determined outcomes to guide protocol development.
MetaDAO’s experience has demonstrated both the promise and the challenges of futarchy. Markets have shown the ability to distinguish between value-creating and value-destroying proposals. However, liquidity constraints in prediction markets have limited the mechanism’s reliability for high-stakes decisions.
Gnosis explored futarchy through its prediction market platform and governance experiments. The Gnosis chain’s governance incorporated elements of futarchic decision-making, using market signals to inform — though not deterministically control — governance outcomes.
Polymarket and Augur, while not governance systems per se, have demonstrated the viability of on-chain prediction markets at scale. Their success in predicting political outcomes, sporting events, and economic indicators provides empirical support for the information aggregation properties that futarchy relies upon.
Advantages of Futarchic Governance
When functioning properly, futarchy offers several compelling properties.
Information aggregation is the primary advantage. Governance decisions in DAOs are often made by participants who may be passionate but poorly informed. Prediction markets create financial incentives for informed participants to reveal their knowledge through trading. A whale who knows that a proposed oracle change will introduce systemic risk can profit by betting against the proposal — and in doing so, steers governance away from the dangerous choice.
Reduced tribalism follows from the separation of values and beliefs. In traditional governance, proposals become tribal markers — participants vote based on who proposed it, which faction supports it, or how they feel about the proposer. Futarchy bypasses this dynamic by letting markets evaluate proposals on their expected outcomes rather than their political associations.
Alignment of incentives occurs because market participants profit by being right and lose by being wrong. Unlike in voting, where expressing a poorly informed opinion costs nothing, futarchic participation carries financial consequences. This filters governance signal towards participants with genuine expertise and skin in the game.
Resistance to manipulation is stronger in markets than in voting systems. To manipulate a vote, an attacker needs only to acquire tokens and cast them. To manipulate a prediction market, an attacker must maintain a losing position against potentially thousands of profit-motivated counter-traders. Market manipulation is possible but economically costly — and the more important the decision, the more liquidity flows in to correct mispricings.
Fundamental Challenges
Futarchy faces several challenges that range from practical to philosophical.
Metric selection is arguably the hardest problem. Futarchy requires a clearly defined, objectively measurable outcome metric. But most governance goals are multidimensional and difficult to quantify. “Protocol health” involves security, decentralisation, revenue, user experience, and developer ecosystem vitality — collapsing these into a single number inevitably loses important information.
The choice of metric also creates perverse incentives. If the metric is token price, proposals that increase short-term price at the expense of long-term sustainability will be favoured. If the metric is TVL, proposals that attract mercenary capital through unsustainable incentives will win. Goodhart’s law — when a measure becomes a target, it ceases to be a good measure — applies with full force.
Liquidity constraints limit the mechanism’s reliability. Prediction markets function well when they are liquid — when many participants trade significant volumes. Thin markets can be manipulated cheaply and produce noisy signals. Most DAO governance decisions are insufficiently interesting to attract deep liquidity, leaving the markets vulnerable to manipulation or simply unreliable.
Temporal mismatch creates evaluation difficulties. Many governance decisions have effects that unfold over months or years. A prediction market that settles in thirty days cannot capture the long-term consequences of a strategic decision. Extending the settlement window increases capital lockup costs, reduces liquidity, and introduces confounding variables that make attribution difficult.
Reflexivity complicates market interpretation. If the market’s prediction influences the outcome — for example, if a negative market signal causes developers to abandon a proposal before it is implemented — the counterfactual becomes impossible to evaluate. Did the market correctly predict failure, or did it cause failure? This reflexivity problem undermines the epistemic foundation of futarchy.
The Token Price Problem
Many futarchic implementations use token price as the objective metric — the proposal that the market expects to produce the highest token price is implemented. This choice is pragmatic: token prices are objective, continuously observable, and naturally capture aggregate market sentiment about protocol value.
However, token price is a deeply flawed governance metric. Prices reflect speculative dynamics, market conditions, and macroeconomic factors that have nothing to do with the governance decision being evaluated. A proposal implemented during a bull market will appear successful regardless of its governance merits. A proposal implemented during a bear market will appear to have failed.
More fundamentally, optimising for token price biases governance towards short-term extractive strategies and away from investments in public goods, community development, and long-term infrastructure. The proposals that maximise token price over a thirty-day window may be precisely the proposals that undermine the protocol’s long-term health.
Alternative metrics — revenue, active users, developer commits, or composite indices — address some of these concerns but introduce their own measurement and manipulation challenges.
Futarchy and Other Governance Mechanisms
Futarchy can function as either a standalone governance system or a component within a larger governance architecture.
As a decision support tool, prediction markets inform but do not determine governance outcomes. A DAO might use markets to generate expected impact estimates for competing proposals while reserving final decision authority for a delegated voting system or quadratic vote. This approach captures the information aggregation benefits of markets without ceding control to them entirely.
As a veto mechanism, markets can flag proposals whose expected outcomes fall below a defined threshold. Rather than selecting between proposals, the market simply answers: “Is this proposal expected to be better or worse than the status quo?” If worse, the proposal is automatically vetoed. This negative filter is more robust to market manipulation than positive selection because the manipulator must maintain a losing position indefinitely.
As a component of optimistic governance, prediction markets can serve as the challenge mechanism during the review period. If the market for a queued proposal shows negative expected impact, the proposal is automatically escalated to a full community vote.
Design Recommendations
DAOs exploring futarchic governance should consider several design principles.
Start with non-binding markets. Deploy prediction markets alongside existing governance mechanisms and track their predictive accuracy before giving them decision-making authority. This calibration period allows the community to build confidence in the mechanism and identify systematic biases.
Choose metrics carefully. Invest significant community deliberation in selecting the objective metric. Consider composite metrics that capture multiple dimensions of protocol health, and build in mechanisms to update the metric as the protocol’s priorities evolve.
Ensure market liquidity. Subsidise prediction markets through matching funds, liquidity mining, or protocol-funded market makers. Thin markets produce unreliable signals, and unreliable signals produce bad governance.
Implement circuit breakers. Define conditions under which futarchic outcomes are overridden by human governance — extreme market conditions, suspected manipulation, or metric manipulation. Markets are powerful information aggregation tools, but they are not infallible.
Futarchy remains the most intellectually ambitious governance mechanism in the DAO design space. Its practical challenges are substantial, but the organisations that solve them will have access to a decision-making tool that is fundamentally more informed than any voting system can be.
Donovan Vanderbilt is a contributing editor at ZUG DAO, the decentralised governance intelligence publication of The Vanderbilt Portfolio AG, Zurich. His work examines the intersection of governance design, institutional economics, and on-chain coordination.