Every generation of markets eventually produces instruments that make risk harder to smooth over and easier to see. Predictive and event contracts are doing precisely that.
These novel markets take exposures that were once implicit and spread thinly across futures, options, and swaps, and concentrate them into contracts whose outcomes resolve as a scheduled all-or-nothing outcome.
For risk managers, the risk itself is not new. The visibility is, and visibility changes the game.
Historically, the risk manager’s remit has been grounded in monitoring continuous price processes and managing risk through well-understood abstractions: volatility, correlation, sensitivity, incremental loss.
Enterprise Risk Management frameworks evolved around these assumptions, supported by Value-at-Risk, Expected Shortfall, and margin methodologies calibrated to gradual repricing and orderly liquidation.
Event contracts disrupt this equilibrium. They jump, rather than drift, into loss and they do so at known resolution times, in a manner that renders diffusion-based models and volatility snapshots insufficient for capturing future exposure.
Risk management, then, must move from inference to anticipation. In portfolios that include predictive contracts, the central question is no longer “How volatile is this book?,” but “Which outcomes are we exposed to, when do they resolve, and what happens if they resolve together?”
This reframes risk around outcomes, rather than instruments. One inflation print, one policy decision, one weather disruption can cut across desks, asset classes, and clearing regimes at once. The risk manager must think at the level of the enterprise, not the individual positions that reference it. Only a firm-wide view reveals true net exposure.
Losses crystallize instantly, variation margin demands spike intraday, and liquidity management plans are tested not over days but over minutes. Margin models calibrated to historical price paths struggle to demonstrate 99th percentile coverage when the distribution itself is discontinuous. Default management procedures must confront the reality that event-driven portfolios may offer little time or flexibility for intervention.
Compounding this challenge is regulatory bifurcation: predictive contracts may be classified as futures under one set of circumstances and as swaps under another, invoking different margin methodologies, collateral segregation rules, and reporting obligations. The economic exposure is the same; the clearing treatment may not be. Offsetting positions can end up margined separately, liquidity fragmented, and natural netting benefits lost.
This pushes the risk manager’s role beyond measurement into governance, as CROs must now advise on contract design, manipulation risk, and work closely with surveillance functions to detect abnormal positioning and outcome-driven crowding.
Technology, too, must rise to meet this expanded role. Systems built to aggregate risk by instrument and shock prices by percentage are poorly suited to portfolios organized around discrete outcomes. Events must become first-class objects in risk architecture – modeled with their definitions, resolution mechanisms, timing, payoff structures, and correlations made explicit. Instruments become expressions of event exposure, not the primary units of analysis.
Risk systems must also operate across clearing and regulatory boundaries without being constrained by them. They must be clearing-aware, economically coherent, and capable of stress testing against adverse event resolution, correlated outcomes, and intraday liquidity demands rather than generic price shocks.
Explainability becomes non-negotiable, and technology must support this scrutiny with transparent models, auditable assumptions, and clear mappings to regulatory principles.
In this setting, the risk manager’s role becomes one of stewardship rather than surveillance. Success depends on technology that aligns with the way risk actually behaves in event-driven markets — technology that respects timing as much as magnitude, outcomes as much as instruments, and governance as much as computation.
