The Number Wall Street Can’t Ignore
In January 2026, prediction markets recorded $12 billion in notional volume in a single month. This was not an electoral spike or an isolated event; it confirmed something that has been building since 2024, when annual volume reached $15.8 billion, only to soar to over $63 billion in 2025, marking over 300% growth in just twelve months.
These figures are sufficient to catch the attention of any serious trading desk. And that’s exactly what happened. Susquehanna International Group, one of the world’s most sophisticated options firms, began hiring traders specifically to operate in these markets. Goldman Sachs CEO David Solomon revealed during an earnings call that he met with teams from Polymarket and Kalshi in January to explore opportunities. Meanwhile, the Federal Reserve published a report indicating that Kalshi is useful for forecasting economic events.
This is not hype. It is a sequence of institutional signals pointing in one direction.
What Institutions Are Really Buying
The superficial narrative paints prediction markets as sophisticated betting platforms. This reading is incorrect, or at least incomplete. What Susquehanna and Goldman Sachs are evaluating is not gambling but real-time price signals on events not covered by traditional instruments.
Standard derivatives allow for expressions of views on rates, currencies, or indices. However, there is no liquid futures contract available to accurately bet on whether the Fed will raise rates at a specific meeting under specific conditions, or on the outcome of a regulatory vote affecting an entire sector. Prediction markets fill that void with a simple mechanism: binary contracts whose price reflects the probability the market assigns to an event.
When a firm like Susquehanna incorporates that price into its risk model, it is not speculating in the colloquial sense. It is integrating a data source that did not previously exist in liquid form. The operational difference is significant: moving from reading surveys or making self-estimations to observing a probability with real money behind it changes the quality of information that enters the model.
GSR's Chief Risk Officer, Alex Taaffe, described it as a turning point for institutional adoption. Marek Sandrik from RockawayX points to the maturation of these platforms as tools for real-time sentiment. Both readings converge on the same point: the asset being bought is market intelligence, not speculative exposure.
There’s also a dynamic of cost structure worth auditing. For a firm with options capability like Susquehanna, entering prediction markets doesn’t require building new infrastructure from scratch. It requires adapting existing models and hiring specific talent. The marginal cost of entry is low compared to potential volume. This is the arithmetic that justifies the movement, regardless of the narrative enthusiasm surrounding the sector.
The Problem That Volume Alone Doesn’t Solve
The 300% growth in volume is an attractive figure, but there are operational variables that this number does not capture. Liquidity in prediction markets is deeply heterogeneous. Contracts on Fed decisions or large political events may have reasonably dense markets. Contracts on more specific or niche events remain thin, with wide spreads that make any sizable position costly.
This creates a structural problem for institutional participants: they cannot scale positions without moving the price. A firm that moves tens or hundreds of millions in traditional derivatives cannot operate under the same logic in a market where the order book quickly dries up. The result is that, for now, institutional participation operates more as an analytical tool exploration than as a large-scale capital deployment.
The regulatory framework adds another layer of complexity. Platforms like Kalshi have had to fight legal battles to operate in the United States. Polymarket, which operates from outside the U.S. jurisdiction, has its own access restrictions. Goldman Sachs can’t just open an account and start trading without a thorough legal analysis of how its capital is classified, how it's reported, and under what regulatory schema the activity falls. Solomon was careful in his language: exploration, not commitment.
This caution makes operational sense. Institutions that arrive first and build positions before regulatory clarity exists assume a costly compliance risk. Those that wait too long concede an advantage to early players. Timing entry into these nascent markets is one of the most strategically challenging problems to solve without perfect information.
The Pattern That Matters to Read What’s Ahead
There is a recognizable historical pattern in how Wall Street absorbs new asset classes or instruments. First come firms with greater tolerance for operational risk and higher modeling capacity. Then, the larger banks arrive with careful explorations. Following that, the regulatory framework institutionalizes access. Finally, the instrument integrates into standard workflows.
Prediction markets are clearly in the second phase of that cycle. Susquehanna fits the profile of firms entering the first phase: high quantitative sophistication, appetite for under-arbitraged markets, agile structure. Goldman represents the second: public recognition of the issue, exploratory meetings, without announced capital commitment. The Fed’s validation for Kalshi is an early signal that the third phase, regulatory, has actors working within it.
What this implies for the native platforms of these markets is straightforward: institutional entry solves the long-term liquidity problem but redistributes price-setting power. Currently, native operators of Polymarket or Kalshi are the ones determining market probabilities. If Susquehanna starts operating with its own models and significant capital, those same operators face a counterparty with informational and computational advantages. The internal competitive dynamics of the market change before regulation changes.
Sustained volume beyond electoral cycles is the most relevant signal from the available data. It indicates that liquidity does not depend on a specific event but rather on recurring contracts linked to monetary policy decisions, macro results, and regulatory catalysts. That is what turns these markets into potential infrastructure, not a seasonal instrument. And it is exactly the type of property that institutional risk models need to permanently integrate a data source.
