Maryland now faces roughly 12,000 childhood sexual abuse claims filed under its 2023 Child Victims Act, concentrated on state agencies and local school systems that carry the exposure on their own balance sheets. The Maryland Association of Counties published a July 15, 2026 status update confirming the count and the tiered damage structure. A 2025 amendment (HB 1378) rolled back the caps, setting $400,000 per claimant for previously time-barred claims not filed before June 1, 2025, and $890,000 for new claims and those filed earlier. The Department of Juvenile Services and county school boards are the primary named defendants; counties and municipalities have seen far fewer filings.
The per-claimant cap gets the headlines, but the cap is the severity ceiling, not the reserve driver. Multiply 12,000 claims by even the lower $400,000 tier and the aggregate runs past $4.8 billion before defense costs. What actually resets the reserve is the reported-claim count landing in a compressed window against accident years that span decades. That is a pure-IBNR problem, not a case-reserve problem.
Who It Affects
Self-insured Maryland public entities and their risk pools carry this directly: the state’s general fund, county boards of education operating self-insurance funds, and the pools that front excess coverage for smaller districts. Judgments and settlements must be funded through existing governmental resources, insurance coverage, or self-insurance reserves, with no dedicated payout plan in place. That last point matters for any finance director signing off on a fund balance: a claim population this large with no appropriated funding source becomes a booked liability the moment reserving actuaries can bracket it.
The exposure also travels. From tracking revival-window filings across New York, New Jersey, and California over the past year, the pattern holds: the per-claim cap draws the coverage, but it is the reported-claim count in the first 18 months that resets the reserve.
The Reserve Mechanism
Two things distinguish this from an ordinary liability reserve. First, occurrence dates span decades while report dates cluster in a two-year window, so pure IBNR dominates and there is almost no clean frequency or severity history to anchor it. Case adequacy is nearly impossible this early because most files are freshly reported and unadjudicated. Second, the tail is long and now bends on litigation, not injury. A June 25, 2026 Maryland Supreme Court ruling (Board of Education v. Sturm) held that county school boards retain sovereign immunity for conduct before July 1, 1971, dismissing that cohort without prejudice. The Court agreed to hear the parallel state question in its fall 2026 term: whether the state is immune for claims predating July 1, 1982, when the Maryland Tort Claims Act took effect. Roughly 1,300 pre-1982 claims and more than $1 billion in exposure ride on that answer, with the Attorney General having filed more than 200 motions to dismiss.
That pending ruling is the single largest swing factor in the tail assumption. A defense win carves out a large, old cohort; a plaintiff win revives it. Either way, the covered accident-year window is not yet fixed, which is exactly the condition under which pure IBNR, not broad IBNR, should dominate the load.
Where This Shows Up in Your Reserves
Look on the public entity general liability report at the report-year IBNR line for revived-claim cohorts, not the accident-year triangle: the triangle has no credible development history for 1970s and 1980s occurrences reported in 2025 and 2026. On the pool or excess side, check whether reinsurance responds on an occurrence-year or report-year basis, because decades-old occurrence years may sit below or outside current attachment terms. Fund this at a stated confidence level rather than an expected value; the count uncertainty is wide enough that a central estimate understates the reasonable range.
Decision-Maker Checklist
- Confirm the pure-IBNR load your actuary carries for revived-claim cohorts, and stress it against reported-claim count, not per-claim severity.
- Ask whether your excess or pool reinsurance responds on an occurrence-year or report-year basis for pre-1982 accident years.
- Model both outcomes of the fall 2026 sovereign immunity appeal as explicit tail scenarios, not a single point.
- Verify your TPA flags claim-level occurrence date so pre-1971 and pre-1982 cohorts can be isolated as the rulings land.
- Reconcile any booked reserve against the absence of a dedicated state payout plan before certifying the fund balance.
Sources
- MACo Conduit Street: Where Do Things Stand With Maryland’s CVA Cases (July 15, 2026)
- The Daily Record: Child Victims Act, Public Schools Immune From Some Claims, MD Supreme Court Rules (June 25, 2026)
- Maryland Matters: State Claims Immunity From Some Sexual Abuse Lawsuits (July 9, 2026)
- Maryland Child Victims Act (2023), Peoples Law Library