Author: Zack Pokorny, Galaxy Research Fellow
Translation: Yangz, Techub News
Prediction markets achieved a "from 0 to 1" breakthrough by the end of 2024 to 2025, establishing their position as a mechanism that aggregates dispersed information into probability signals. Future applications may extend this capability, not only answering whether an event will occur but also further revealing what these events mean for asset prices and organizational decision-making.
We anticipate this expansion will take two forms: Impact Markets and Decision Markets. Impact Markets present the collective view of the market on the conditional valuation of assets (e.g., if the Federal Reserve cuts rates by 75 basis points, what will Bitcoin's trading price be; or if a specific candidate wins the election, what will Nvidia's valuation be). Decision Markets go a step further, using these conditional valuations to automate organizational governance, allowing markets to directly decide actions based on expected economic outcomes. Currently, a few projects are working toward these concepts, and we expect their numbers to grow over the next year and beyond, as products and user ecosystems around them flourish.
Both mechanisms are built on the predictive market's ability to aggregate information, but they shift the output from event probabilities to actionable intelligent information. These markets no longer stop at "there is a 65% chance this will happen," but answer "what does this specifically mean for your portfolio" and "what actions should your organization take." While infrastructure and incentive structures are similar, these markets can explicitly reveal a broader range of questions.
Limitations: Event Probabilities and Economic Outcomes
The current key constraint is that prediction markets only provide binary payoffs for discrete events, separated from asset price outcomes. For example, users can bet on the likelihood that the Federal Reserve will cut the benchmark rate by 25, 50, or 75 basis points, each with its corresponding "yes/no" market and a set of probability outcomes. But they cannot directly trade or explicitly communicate signals indicating how bonds, Bitcoin, or any other assets will be affected if these rate cuts occur. Prediction markets capture the probability of events but do not reveal how these events will influence asset prices. Such conditional pricing information is valuable not only for hedging but also for strategic decision-making.
The current utility of prediction markets as hedging tools and information dissemination channels should not be underestimated. They provide a direct mechanism to hedge risks of discrete events (e.g., the timing of the next-generation ChatGPT model release or whether military actions will occur before a certain date), risks that are often difficult or impossible to hedge in traditional markets. Moreover, by aggregating dispersed knowledge from real capital participants, prediction markets can rapidly synthesize scattered information into continuously updated probabilities with capital bias, outperforming expert forecasts and polls in historical performance.
Impact Markets
Impact Markets address a fundamental information gap. Current prediction markets cannot tell you the value of an asset under specific event conditions. This information does not exist in an explicit, discoverable form. While prediction markets reveal event probabilities and spot markets reveal prices, there is no mechanism to present the collective view of the market on the conditional valuation of assets tied to specific events, such as: if the Fed cuts rates by 75 or 50 basis points, at what price will Bitcoin trade; or if a candidate warning about AI wins a major election, how will Nvidia's stock perform.
In Impact Markets, users do not trade probabilities with "yes/no" tokens but directly trade the assets themselves under conditional states. For example, a position might be: "I am willing to buy Bitcoin at a price of $110,000 (a 10% premium over the market price) only if the Federal Reserve cuts rates by 75 basis points." This fundamentally enhances the information revealed by the market. We no longer need separate markets for "probability of Fed rate cut" and "Bitcoin price," but can directly discover the price of "Bitcoin | Fed rate cut 75 basis points." This concept can extend to any "asset | event" pairing, such as "GOOGL | GPT 6 release before Gemini Series 4" or "Gold | asteroid mining before 2030," and so on.
The key distinction is: an event is one thing, but its impact on companies and assets is another—these are fundamentally different. Prediction markets aggregate the probability of whether an event will occur. Impact Markets answer the next question: "If this event occurs, what will happen to this company or asset?" This separation allows each market type to operate professionally while creating a more complete information set.
This architecture solves the multi-step reasoning problem that plagues markets: traders must gather probability odds from prediction markets, input them into proprietary models to estimate asset impacts, and then execute independent trades. When traders see a 25% probability of a 75 basis point rate cut and a 65% probability of a 25 basis point cut, they must independently interpret what these probabilities mean for their Bitcoin holdings, then pray that their correlation assumptions hold when executing (or hope the Fed doesn't unexpectedly hike rates and blow up their positions). Impact Markets simplify the entire workflow into direct price discovery of conditional valuations, with settlement only occurring if the event actually happens. The benefits are substantial:
-
Revealing Hidden Information: Currently, no mechanism can present the collective view of the market on the conditional valuation of assets tied to specific events. While prediction markets reveal event probabilities and spot markets reveal current prices, Impact Markets answer the question neither can directly address: if the event occurs, at what price will the asset trade?
-
Genuine Economic Hedging: Bitcoin holders worried about election impacts can lock in a conditional price for their assets under specific concerns. This is fundamentally different from prediction market "hedging." They do not need to hedge event probabilities while managing separate asset positions; instead, they execute a single trade that guarantees their economic outcome if the event occurs. This minimizes the basis risk between their event outlook and asset exposure.
-
Reducing Model Risk: Users do not need to build correlation models or estimate how events influence prices. Markets automatically aggregate these views through display preferences. Reasoning issues are addressed by allowing market participants to directly express conditional valuations.
The most impactful insight is that Impact Markets reveal the market-implied joint distribution of events and asset prices, information that is hidden under current prediction market architectures or requires complex model assumptions to extract. This is valuable not only for hedging but also for any decision-making that depends on understanding how events influence economic outcomes.
Subsequently, organizations can effectively leverage this model to glimpse conditional futures and act accordingly. We call this the Decision Market.
Decision Markets
Decision Markets originated from economist Robin Hanson's 2000 working paper "Shall We Vote on Values, But Bet on Beliefs?" extending the impact market mechanism from information revelation to governance automation. They do not merely reveal conditional valuations for individual decision-making reference but directly and bindingly determine whether an organization takes a particular action, based on which outcome is priced higher in the market. This mechanism has been implemented in "futarchy" DAOs, with total trading volume reaching several million dollars. In typical setups, an organization proposes a decision (e.g., whether to dilute its token supply by 5% to fund a new product line), and the market trades two conditional "states": "pass" and "fail." Each state assigns a value to the organization's tokens, with the price serving as the objective function—the target the market seeks to optimize. If the market prices tokens higher in the "pass" state, the organization executes the decision. If the "fail" state tokens are priced higher, the proposal is rejected, and no action is taken. Market participants collectively decide which action maximizes expected value, with their trades conditionally executed based on the winning outcome. Galaxy Research has reported on these markets and their organizational applications in reports on future politics and their on-chain implementations, as well as in our annual forecasts for 2025 and 2026.
However, this structure reveals a critical constraint: the effectiveness of the objective function. For decision markets to produce meaningful signals, the traded assets must have a causal relationship with the outcomes the decision aims to optimize; ideally, holders should have substantial ownership rights over the economic value generated by the relevant application. Many (possibly most) crypto tokens fail this test. When a governance token's value is structurally decoupled from the growth of on-chain applications, or token holders do not substantially benefit from the economic gains of the chain, it cannot effectively guide ecosystem grants. If "performance" (or harm) cannot flow back to the traded assets, the market cannot reveal whether a grant will succeed. Worse, rational traders may refuse to participate altogether. When the success of an application does not accrue value to the token, why buy governance tokens (and increase their economic exposure) to support grants? When the objective function and decision domain are misaligned, the market's ability to accurately predict the best decision collapses.
This effectiveness issue makes it very difficult to adapt decision markets to existing organizations.
Mature DAOs have established specific governance processes, social norms, informal power structures, and (most critically) token and ownership structures that may not align token value with organizational outcomes. Introducing binding decision markets requires fundamental restructuring—such as how proposals are formed, how rights are exercised, and how contributors act post-decision—and often changing the organization's own structure, as token ownership may need to be reorganized to fully link economic outcomes with governance decisions. The human, coordination, and capital/legal costs to achieve this may be prohibitive for some organizations.
Newly formed organizations are not constrained by these issues. They can design from the outset with decision markets as the default coordination mechanism, specifically engineering tokenomics and governance processes to ensure the objective function and governance expectations remain aligned. Tokens can be designed so that organizational success necessarily flows into token value, avoiding the disconnection issues faced by legacy organizations restructured for decision markets. Therefore, decision markets are more likely to be adopted through the creation of new, inherently market-governed organizations rather than grafted onto existing structures.
This points a path for the broader development of decision markets. They do not replace existing coordination mechanisms but enable entirely new organizational forms that were previously impossible. These entities can aggregate information, coordinate incentives and ownership, and execute decisions through direct market mechanisms rather than political processes.
This offers advantages but also constraints. Decision markets are best suited for assessing risk and return, maximizing resource deployment efficiency, and pricing decisions under high uncertainty. They are especially powerful for capital allocation, resource deployment, and strategic choices where economic impact is the primary concern. However, they are less suitable for decisions involving qualitative factors, such as value alignment, community social capital, or standards that cannot be reduced to token prices. In decision markets, the mechanism must optimize for economic value to generate meaningful signals. This is a feature, not a flaw. The strength of decision markets lies precisely in their ability to compress complex trade-offs into a single economic value dimension. As a result, organizations relying on this structure may govern themselves differently from today's typical on-chain organizations. They might not micromanage application details but focus governance on monthly contributor stipends and burns, treasury expenditures, and asset control (including intellectual property), while operators retain autonomy over creativity and daily operations.
Summary
Prediction markets have demonstrated that markets can aggregate information about whether events will occur. Impact Markets and Decision Markets are poised to extend this understanding, revealing the value of these events and, in some cases, allowing markets to directly decide what actions organizations should take. Impact Markets can fill critical information gaps by presenting conditional asset valuations, enabling true economic hedging and event-driven price discovery. Decision Markets will further leverage these valuations to guide capital allocation and strategic decisions. Together, they mark a shift from merely predicting events to pricing their consequences and acting accordingly.
While these next-generation markets are still in their infancy, the groundbreaking success of prediction markets over the past two years suggests a promising new chapter in information finance.
