Hook
Prediction markets have spent years asking whether they can forecast reality. The more consequential question now is whether reality can be traded at institutional scale.
Cantor Fitzgerald, a major financial services firm, is preparing to connect institutional clients with Kalshi, the US Commodity Futures Trading Commission regulated event-contract exchange. Susquehanna International Group, one of the world's largest quantitative trading firms, is also entering the market as a liquidity provider and pricing specialist. The announcement is significant because it addresses the central weakness of prediction markets: not a lack of public interest, but a lack of reliable depth for large orders.
The immediate story is being framed as a major Wall Street endorsement of prediction markets. That is accurate, but incomplete. The more important development is structural. Kalshi is moving away from the image of a retail wagering venue and toward the operating model of a regulated derivatives marketplace, where brokers introduce capital, market makers absorb risk, and institutions use event contracts as instruments of exposure management.
This is not a breakthrough in blockchain engineering. It is a test of whether financial intermediation can make a controversial market usable for institutions without removing the uncertainty that gives the market its value.
Context
Kalshi operates as a regulated designated contract market under the CFTC framework. Its products are event contracts: instruments whose payout depends on whether a defined real-world outcome occurs. A contract may settle according to an economic indicator, a policy decision, a weather condition, or a political event, subject to the exchange's rules and regulatory permissions.
The basic structure is simple. A contract trades between zero and one dollar, with the market price often interpreted as an approximate probability of the outcome. A contract priced at forty cents implies that buyers and sellers are collectively assigning a roughly forty percent probability to settlement at one dollar. This simplicity is deceptive. The contract still requires a rulebook, an authoritative data source, a dispute process, collateral controls, surveillance, and final settlement procedures.
For small traders, an electronic order book may be sufficient. For a hedge fund or asset manager, it is not. Institutional participation requires predictable execution, credit processes, legal documentation, reporting, compliance integration, and a way to transact without revealing a large position to the entire market. Thin order books create a direct cost: a large order moves the price against the buyer or seller, while an attempted exit can become impossible during a volatile event.
Cantor's role is therefore more than customer acquisition. As an introducing broker, it can connect its institutional network to Kalshi's regulated venue and help translate event contracts into a familiar workflow. Susquehanna adds another missing component: professional two-sided pricing. A prediction market cannot become a risk-management tool if the quoted price disappears whenever the underlying event becomes politically or economically sensitive.
The division of labor resembles established markets. The exchange provides the legal and operational framework. The broker provides access. The market maker supplies continuous prices and manages inventory. Institutions provide demand. The blockchain sector has often treated disintermediation as the highest form of financial progress. This arrangement suggests that, for some products, institutional adoption may begin with carefully regulated intermediation instead.
Core Insight
The central innovation is not a new ledger, token, or consensus mechanism. It is the transplantation of block-trade logic from equities and fixed income into event contracts.
A public order book is efficient when orders are small, information is widely distributed, and the market can absorb ordinary volatility. It becomes fragile when an institution needs to trade a position large enough to expose its intentions. In that situation, a negotiated block trade can reduce information leakage and limit visible price impact. The trade may still carry risk, but the risk is handled through an established financial process rather than left to the mechanical depth of an open book.
This distinction matters because prediction markets are often discussed as if their primary problem were participation. In practice, they have a capacity problem. A market can attract thousands of users and still fail the institutional test if ten million dollars cannot be executed without severe slippage. Headline volume is not the same as executable liquidity. Liquidity is a mirage when it exists only at the quoted size.
Susquehanna's participation directly addresses that weakness. A specialist market maker does not merely add more orders. It contributes models, risk limits, event interpretation, and the ability to price correlated outcomes. If a firm is trading contracts on interest rates, inflation, elections, or energy supply, the positions may not be independent. A shock that changes the probability of one event can alter the value of several others. Professional liquidity depends on understanding those relationships, not simply matching buyers with sellers.
That creates an important information gain for observers of this market. The institutional value of prediction contracts may not come from their standalone accuracy. It may come from their correlation with exposures that institutions already hold. A fund with interest-rate risk might use a macroeconomic event contract as a small, transparent hedge. An asset manager with political or regulatory exposure might use contracts to express a view that is difficult to implement through conventional securities. An insurer might study event prices as an external signal for risks its existing products do not cover.
In that setting, the market price is not only a forecast. It is a tradable risk coordinate. The contract tells a portfolio manager how other participants are pricing a discrete uncertainty, while the ability to transact allows the manager to adjust exposure. This is closer to derivatives infrastructure than to a public poll.
My own experience auditing financial transaction systems has made me skeptical of announcements that describe distribution as innovation. In 2017, while examining high-volume transaction flows in Hangzhou, I saw how a system could process enormous throughput while remaining dependent on a narrow set of operational bottlenecks. Performance metrics concealed concentration. The same warning applies here. Cantor and Susquehanna can improve access and depth, but they may also concentrate pricing power in a small group of intermediaries.
Code is law, but who writes the law? In a regulated event market, the answer is not only the software team. It is also the exchange rulebook, the broker's compliance department, the market maker's risk committee, the data provider, and ultimately the regulator. Every one of these actors influences what counts as a valid event, which evidence settles a contract, and how disputes are resolved.
That governance stack offers institutional confidence, but it changes the source of trust. A decentralized prediction market asks users to trust transparent contracts, public settlement logic, and self-custody. Kalshi asks participants to trust a regulated operator, a legal framework, and a controlled settlement process. Neither model eliminates trust. They relocate it.
This is why the event should not be misclassified as a blockchain adoption milestone. The source material provides no evidence that Kalshi's core infrastructure depends on a decentralized ledger, permissionless validation, or a native token. There is no token economy to analyze, no staking system securing settlement, and no community governance mechanism determining market rules. The business model is more likely to depend on transaction fees, data services, and institutional execution than on inflationary incentives.
That absence is itself informative. During the previous cycle, many platforms attempted to manufacture liquidity through token rewards. The resulting activity often disappeared when subsidies ended. A regulated exchange with professional intermediaries is pursuing a different model: make participation economically useful, then charge for the infrastructure that supports it. This may produce slower growth, but it is easier to evaluate.
The arrangement also reveals a three-layer institutional migration. First, prediction markets need legal recognition as legitimate financial venues. Second, they need brokers that can carry the product into existing investment organizations. Third, they need market makers that can survive the uneven distribution of information surrounding major events. Kalshi supplies the first layer, Cantor the second, and Susquehanna the third. If any layer fails, the institutional thesis weakens.
The most difficult test will be settlement integrity. Event contracts can appear objective until the relevant question is defined precisely. What constitutes an election victory? Which agency's release controls an economic result? How are revisions, delays, recounts, or conflicting data sources handled? In ordinary securities markets, the asset exists independently of the settlement rule. In prediction markets, the settlement rule is part of the asset. Ambiguity is not a footnote; it is counterparty risk.
Your data is not yours anymore. That phrase is usually applied to personal information, but it also describes the institutional future of prediction markets. A broker, exchange, market maker, and regulator will each generate records about who traded, when they traded, and what information they may have possessed. For institutions, this surveillance supports compliance. For individual participants, it may mean less privacy than they associate with crypto markets.
The result is a platform that may be less technologically novel than a decentralized alternative but more legible to banks, funds, and auditors. Financial markets are often adopted through legibility before they are adopted through elegance. Institutions need to explain a product to a compliance officer before they can place it in a portfolio.
Contrarian Angle
The obvious conclusion is that institutional support will solve prediction markets' liquidity problem and launch a new asset class. That conclusion may be premature.
Professional market makers can quote deeper prices, but they cannot create unlimited willingness to take the other side of a politically or economically explosive event. During extreme uncertainty, their models may converge rather than diverge. If every risk model identifies the same adverse scenario, displayed liquidity can retreat precisely when users need it most. A negotiated block trade reduces market impact; it does not abolish information risk or settlement risk.
The regulatory advantage is also conditional. CFTC oversight is a powerful institutional signal, but it is not a permanent guarantee that every event contract will be permitted. Political contracts are especially vulnerable to changing interpretations of market integrity, public interest, and the boundary between hedging and wagering. A platform can be compliant today and still face a narrower product set tomorrow.
There is another blind spot. The institutional model may validate prediction markets while weakening the decentralization narrative that first made them attractive to crypto users. A regulated, centrally operated exchange can provide better reporting and execution, yet it can also restrict access, collect more user data, and reserve the most efficient trading channels for approved counterparties. Code is law, but who writes the law? Here, access policy and supervisory discretion may matter more than smart-contract transparency.
Polymarket and similar platforms therefore should not be dismissed simply because institutions favor Kalshi. They may serve different demand. A permissionless platform can generate faster experimentation, broader market creation, and global participation. Its weaknesses are regulatory exposure, settlement governance, and uneven institutional access. The two models may coexist, but they will compete for attention, data, and legitimacy. The winner may not be the platform with the highest peak volume. It may be the one that retains meaningful volume after the election cycle and other headline events have passed.
Takeaway
Cantor Fitzgerald and Susquehanna are testing whether prediction markets can become instruments for managing uncertainty rather than merely trading curiosity. The answer will depend on post-event behavior: sustained institutional volume, transparent settlement, resilient quotes, and evidence that contracts remain useful when no election is dominating the news cycle.
Liquidity is a mirage until it survives stress. The next phase should be judged by executable depth, concentration of pricing power, and the quality of risk transferred. If Kalshi can meet those tests, regulated prediction markets may enter the broader derivatives landscape. The question for the next cycle is not whether institutions can enter, but which uncertainties they will finally be willing to price.