Comment Text:
Re: ANPRM on Prediction Markets (RIN 3038-AF65)
Dear Secretary Kirkpatrick,
Amplifi submits this comment in response to the Commission's Advance Notice of Proposed Rulemaking on prediction markets. Amplifi is a decentralized protocol that provides leverage for prediction market positions through isolated lending pools. Our comments focus on Questions 2.f and 3.c regarding margin trading.
Why Margin Matters for Prediction Markets?
Liquidity is the most important efficiency driver for any market. Full collateralization, while eliminating counterparty credit risk, creates a capital efficiency bottleneck that constrains liquidity, widens spreads, and slows price discovery. Margin allows participants to deploy capital more efficiently, directly improving market depth and the speed at which prices reflect new information. These improvements serve the core purposes of CEA Section 3(a).
As an additional benefit, deeper liquidity raises the cost of market manipulation. When order books are thin, small amounts of capital can move prices significantly. Margin-driven liquidity makes this harder, supporting the Commission's mandate under CEA Section 3(b).
Response to Question 2.f
Retail vs. institutional. Margin should be available to both. Restricting it to institutional participants would undermine Core Principle 2's impartial access requirements and create a two-tiered market. The goal should be open access with robust risk controls, not access restrictions by participant class.
Disclosure. Retail participants must understand two risks specific to leveraged prediction markets: (1) they may lose their entire equity even if they correctly predict the event outcome, because interim volatility can trigger liquidation before resolution; and (2) event contracts are uniquely susceptible to sudden information shocks (rumors, preliminary reports) that cause temporary price spikes and trigger irreversible liquidations, even when the price later recovers.
Margin calculation. Requirements should be dynamic and market-specific, not static or uniform. Key inputs include real-time price volatility, order book depth and depth volatility, and the realistic ability to liquidate a position in current conditions. Liquidation conditions should adjust in real time as market conditions change. A statistical, risk-based approach is preferable to flat percentage rates, which are inevitably too loose for volatile markets and too tight for stable ones.
Time to resolution. Short-duration events are significantly more volatile and require more conservative margin, but capital is locked for less time. Long-duration events have lower short-term volatility but capital is tied up longer, and the risk of fundamental shifts in the underlying probability is higher. Both profiles argue against one-size-fits-all margin; risk should be priced per-market based on each contract's characteristics.
Cross-margin. Correlated contracts should be eligible for cross-margining. For example, a participant long on "Candidate A wins" and short on "Candidate B wins" in a two-candidate race holds naturally offsetting positions. Requiring full margin on both ignores this and forces capital against risk that does not exist in aggregate.
Response to Question 3.c
Margin models calibrated for traditional derivatives may not capture event contract dynamics: binary payoffs, information shocks, and variable resolution timelines. The Commission should consider isolated clearing approaches where risk from one category of event contracts does not affect another, preventing contagion across unrelated markets. Cross-margin arrangements should be permitted where genuine correlation exists, supported by stress-tested analysis.
Amplifi's Approach: Isolated Lending Pools
Amplifi has developed a novel approach to margin infrastructure that we believe is relevant to the Commission's considerations. We create a permissionless lending pool for each individual prediction market. Rather than a centralized system setting margin by fiat, lenders supply capital to market-specific pools and price risk independently based on each market's characteristics.
This architecture provides: risk isolation (bad debt in one market cannot cascade into others); market-driven risk pricing (lenders naturally restrict leverage in thin, volatile markets and provide it in deep, stable ones, creating a self-regulating mechanism); on-chain transparency (all margin ratios, collateral levels, and liquidation thresholds are publicly auditable in real time, directly serving the Commission's surveillance goals under Core Principles 3 and 4); and open, non-discriminatory access (anyone meeting collateral requirements can participate, consistent with Core Principle 2).
This model demonstrates that the margin question need not be binary between full collateralization and traditional brokered margin. The Commission's framework should be flexible enough to accommodate such innovation, consistent with CEA Section 3(b)'s directive to promote responsible innovation.
Respectfully,
Can Kisagun
Co-founder and CEO of t1 Labs
