Criterica Intelligence — production models trained on real court records, not synthetic data
For Funders

The Secondary Market for Legal Claims

A litigation position can be sold before it resolves — to another fund, a specialty buyer, or as part of a portfolio transfer — but the market stays thin because there is no shared reference price two counterparties can negotiate from.

Why the secondary market stays illiquid

Litigation positions are transferable in principle — a fund can sell a funded claim, a portfolio of positions, or a participation interest to another capital provider before the underlying case resolves — and this happens in practice, particularly when a fund needs liquidity ahead of an LP redemption, wants to de-risk a concentrated exposure, or is restructuring around a fund's wind-down. What keeps this market thin compared to other asset classes with active secondary trading is the absence of a shared reference price: the seller's valuation is built on proprietary underwriting the buyer cannot fully verify, and the buyer's counter-valuation is built on an independent re-underwriting that may use entirely different assumptions about outcome probability and duration. Without a common distribution both sides can inspect, every transaction becomes a full re-underwriting negotiation, which is slow and expensive relative to the size of many individual positions.

What a shared calibrated framework changes about the negotiation

A calibrated, jurisdiction- and judge-aware outcome model that both counterparties can reference — even if they disagree on the specific inputs for a given case — narrows the negotiation to a smaller set of disputed variables instead of requiring the buyer to reconstruct the entire underwriting thesis from scratch. If both sides agree on the base rate for the jurisdiction and case type, and disagree only on how a specific piece of case-specific evidence should move the position relative to that base rate, the negotiation is faster and the resulting price is more defensible to each side's own investment committee. This does not eliminate disagreement about a specific case's facts — it removes disagreement about the baseline those facts are being compared against.

It also changes what a portfolio transfer looks like in practice: a seller can present a book's concentration profile (judge, venue, defendant, causation theory) and duration distribution using the same framework a buyer's own underwriting team uses, which shortens diligence on a multi-position transfer from a case-by-case re-underwrite to a targeted review of where the seller's and buyer's models actually diverge.

What it does not solve

A shared framework does not resolve the fundamental illiquidity premium that comes from the position being genuinely hard to exit on a specific timeline — a case does not resolve faster because it changed hands, and a buyer taking on a position mid-litigation still needs to underwrite whatever has changed since the seller's original intake assessment, including anything the seller's monitoring should have caught and priced in already. What it does is remove the artificial illiquidity that comes purely from information asymmetry and incompatible private models, leaving the residual illiquidity that comes from the underlying asset actually being a multi-year, binary-resolution legal claim — which is a real, structural feature of the asset class, not a market inefficiency a data layer can price away.

Standardized reporting is the other precondition for a real market

A shared calibrated model narrows disagreement about the baseline, but it does not by itself make a position easy to transact — that also requires a standardized, position-level reporting format a buyer can review quickly, analogous to a loan tape in credit markets: case-level status, current probability and duration estimate versus the original intake estimate, monitoring history, and any triggered early-warning signals, presented in a consistent structure across every position a seller might offer. Without that standardization, even two counterparties using the same underlying model still spend most of a transaction's timeline reconstructing basic position status from unstructured case files and outside counsel updates.

Funds that maintain this kind of standardized, always-current reporting internally — not built specially for a transaction, but as the ordinary output of their monitoring framework — are the ones positioned to transact quickly when a liquidity need or a portfolio rebalancing opportunity arises, because the diligence package a buyer needs already exists rather than needing to be assembled under time pressure.

Standardization also protects the seller, not only the buyer. A seller offering a well-documented, standardized position invites less aggressive re-underwriting from a prospective buyer than one offering a thin or inconsistent file, because the buyer has less unresolved uncertainty to price a discount against. In a market where information asymmetry is the primary source of illiquidity, the party that reduces its own information asymmetry first is usually the party that captures more of the transaction's value, not less.

What to ask for from an intelligence provider

  • 01A jurisdiction- and judge-aware base rate that both a seller and a buyer can independently verify against the same methodology.
  • 02Portfolio-level concentration and duration reporting formatted for transfer diligence, not only for internal fund reporting.
  • 03A documented monitoring history for any position being transferred, so the buyer inherits the seller's post-intake signal rather than starting from zero.
  • 04A clear statement of what a shared framework does and does not resolve — it narrows disputed inputs, it does not manufacture liquidity the underlying asset does not have.
  • 05A standardized, always-current, complete position report maintained as the normal, routine output of ongoing monitoring, not assembled specially and under real time pressure once a transfer opportunity actually appears on the desk and a buyer is already waiting.

Statistics shown reflect historical or illustrative model outputs derived from real case data. They are not predictions or guarantees of any individual outcome. Litigation results depend on facts, jurisdiction, judge, and counsel, and vary case by case. Model accuracy is subject to selection effects and changing legal dynamics.

Prepare a book for secondary transfer.

Start a Portfolio Audit
← All Funder Guides