If you administer a public entity risk pool or sit on the board of a joint powers authority that self-insures workers compensation, liability, or property, you have been handed a reserve number labeled IBNR and asked to sign off on it. This article explains what the number is for a pooled governmental structure, why the failure modes are different from a private captive, and what to expect from the annual reserve study.
A public entity pool is a self-insurance vehicle formed by two or more governmental units under a state interlocal cooperation statute. The California version is the joint powers authority created under Government Code section 6500 and following. The Texas version is the interlocal cooperation contract authorized by Texas Government Code Chapter 791. Most other states have a parallel enabling statute. The pool is not a licensed insurance company, it is not regulated by a state insurance department in most domiciles, and it does not file a statutory annual statement. It is a governmental cooperative that finances risk on behalf of its member governments, and it answers to GASB rather than NAIC.
IBNR is the sum of claims that have already occurred but have not yet been reported to the pool (pure IBNR) plus the development still expected on claims the adjuster already knows about (IBNER). Actuaries use that broad definition almost universally (Friedland, p. 388). If the letters are new to you, start with the plain-English IBNR explainer and come back for the pool-specific translation. If your pool writes general liability and you want the line-specific deep dive on tort, police, and revival-window exposures, the public entity GL IBNR explainer is the companion piece to this one.
Why a public entity pool is not a group captive
A public entity pool looks like a group captive from a distance. Both pool risk across multiple owners. Both retain a working layer and reinsure the excess. Both produce one aggregate reserve number that gets allocated back to members. Up close, four things make a pool a different reserving problem.
Regulatory regime. A group captive is a licensed insurer in its domicile and signs a Statement of Actuarial Opinion under ASOP 36 every year. A public entity pool in most states is not licensed and does not file an SAO. The pool’s reserve number flows through the member governments’ financial statements under GASB Statement No. 10, Accounting and Financial Reporting for Risk Financing and Related Insurance Issues, which requires recognition of an estimated liability for claims that are probable and reasonably estimable, including IBNR. The regulatory floor is lower, but the financial-reporting consequence is just as real.
Member type. A group captive’s members are private companies whose governance answers to shareholders and whose loss-control decisions flow through commercial budgets. A pool’s members are cities, counties, school districts, transit authorities, and special districts whose governance answers to elected officials and whose loss-control budgets are subject to political cycles. When the pool’s actuary recommends an assessment to restore surplus, the member finance director has to defend that assessment in a public council meeting. That is a fundamentally different political economy from a captive board vote.
Exposure mix. Group captives typically write one or two lines in a single industry: trucking auto liability and workers compensation, for example. A public entity pool often covers the full spectrum of governmental risk: workers compensation, general liability (premises and police), employment practices, vehicle liability, property, and sometimes cyber and crime. The aggregate reserve combines all of these. The allocation back to members has to respect the very different loss profiles across both lines and member types.
The immunity layer. This is the wrinkle no captive faces. Many governmental units retain sovereign or qualified immunity for certain categories of claim, and the immunity attaches to the member, not always to the pool. When immunity applies, the case reserve a pool carries for the disputed claim may collapse to zero (or close to it). The Texas appellate ruling that the TAC risk pool was a governmental unit immune from a deputy’s death benefits claim is a recent example of how a single decision can reshape the reserve profile for an entire category of pool exposure overnight.
The plumbing is the same as a group captive. The pressure points are not.
What most public entity pools actually finance
The canonical pool book is some mix of four lines. Each behaves differently in the triangle, and the aggregate reserve is the sum of four very different reserving problems sitting under one roof.
Workers compensation. The longest tail in the book. Indemnity claims for permanently disabled public safety workers can stay open for decades, and medical exposure on lifetime-medical jurisdictions runs even longer. The California public self-insured medical severity surge of 13 percent per year for three years running is the kind of trend that compounds across every open accident year. Friedland identifies payroll as the standard exposure base (Friedland, p. 132), and pools normalize on payroll the same way private self-insureds do.
General liability. The line where latent claims and revival windows dominate everything else. Civil rights cases, abuse cases, environmental contamination claims, and police misconduct cases can emerge ten, twenty, or thirty years after the underlying conduct. The public entity GL IBNR explainer goes through this in detail. The summary version is that a triangle built from the last fifteen years of data assumes claims were substantially reported within that window, and a state revival-window statute breaks that assumption.
Property. The shortest tail. Most property claims report and pay within twelve to twenty-four months. The reserving problem is dominated by catastrophe exposure rather than development pattern: a single hurricane, wildfire, or earthquake event can reshape a year. Pools that carry significant property tend to reinsure aggressively and reserve the property line on a different cadence than the casualty lines.
Auto liability. A mid-length tail. Settlements on serious bodily injury cases run three to seven years from accident date. The exposure base is vehicle count or vehicle miles. Pools with transit-authority members have a meaningfully higher per-occurrence severity than pools of small cities, and a competent reserve study segments by member type where data supports it.
A pool that lumps all four lines into a single triangle is producing the wrong answer for at least one of them. The reserve report should show separate triangles by line, even when the aggregate number is what appears on the balance sheet.
The aggregate-versus-member tension
A pool’s single loss triangle contains the claim experience of every member. That is computationally convenient, and it hides the same editorial tension that runs through every group captive reserve report: the actuary is reserving for the pool as a whole because that is what the auditor and the pool’s balance sheet require, but the board is thinking about individual members because the pool’s assessments, dividends, and member retention decisions all run through member-level allocations. A good actuary produces both views and reconciles them. A mediocre one delivers the aggregate number, lets the member allocation happen somewhere off to the side in a separate spreadsheet, and does not explain how the two views line up.
That tension is sharper in a public entity pool than in a private group captive for two reasons. First, member assessments are politically visible. A city council that has to approve a special assessment because the pool’s reserves developed adversely is going to ask hard questions about whether the assessment is fair to its city specifically. Second, member exit is heavily constrained. A private member of a group captive can leave at the next renewal, taking their experience with them. A member of a JPA usually cannot leave cleanly because the interlocal agreement assigns continuing liability for prior accident years, and the mechanics of unwinding that liability are governed by the JPA’s bylaws and the relevant state statute. Members are stuck with each other in a way captive members are not, and the allocation conversation matters more as a result.
If your reserve study gives you only the aggregate side, you are missing the governance conversation that is arguably the reason your pool exists in the first place.
The methods and a pool-specific wrinkle
Public entity pool reserving uses the same core toolkit as any other property and casualty reserving engagement. Chain ladder is the workhorse for mature accident years, built on the assumption that claims already recorded will continue to develop in the same pattern (Friedland, p. 84). Bornhuetter-Ferguson is the workhorse for recent and intermediate years, anchoring the unreported portion to an expected-claims estimate rather than to a leveraged development factor (Friedland, p. 152). Expected claims is the right answer for a brand-new line, a brand-new cohort of members, or the very latest accident year (Friedland, p. 131). Friedland states directly that no single method can produce the best estimate in all situations (Friedland, p. 345), and a competent reserve study shows you which method was selected for each accident year and line.
The pool-specific wrinkle is that membership has almost always changed materially over the experience period. A pool that had thirty members in 2015 and now has forty-five, with eight of the new ones joining after 2022, does not have a clean ten-year history of the current book. The oldest accident years in the triangle reflect a pool that no longer exists. Friedland’s general principle on a changing mix of business applies directly: any change in the composition of the underlying risk invalidates the chain ladder assumption that past development patterns project the future. The standard fixes are to reweight historical data to approximate the current member mix, apply expected claims to the current membership for recent years, or segment the triangle by membership cohort. Whichever approach is selected should be visible in the report. If the study treats ten years of history as a single pool, ask how membership change was handled.
The retention and reinsurance structure
Most public entity pools retain a working layer per occurrence and buy excess-of-loss reinsurance above the retention. Many also buy an aggregate stop-loss that caps the pool’s total annual liability at a multiple of expected losses. Some pools layer additional reinsurance above the stop-loss through a public risk pool reinsurance market that specializes in governmental cedents.
That three-layer structure is exactly the one Friedland addresses when she describes the reserving pattern for a self-insured pool with a combination of per-occurrence excess-of-loss and aggregate stop-loss coverage. The standard approach is to estimate ultimate claims limited to the per-occurrence retention first, then apply the stop-loss limit as a final adjustment (Friedland, p. 332). Trying to model the entire gross distribution in one step gives the wrong answer because the stop-loss is triggered by aggregate outcomes, not individual large claims, and the per-occurrence layer is already removing the tail claims that would otherwise dominate the gross number.
The pool-specific twist is that the immunity layer described earlier sits inside the gross-to-net bridge. A claim that is initially reserved at full exposure may, after an immunity ruling, collapse to a much smaller amount or disappear entirely from the pool’s net liability. The case reserve falls; the development pattern reflecting that fall arrives in the triangle several months later. If the actuary does not know which case reserve movements were driven by immunity rulings versus settlement outcomes versus reserve adequacy adjustments, the reported triangle is mixing three different signals, and the projected ultimate is harder to interpret.
What the pool board needs to see in the report is a clear bridge: gross losses of the underlying member exposures, capped losses at the per-occurrence retention, the aggregate stop-loss adjustment on top, and a separate note on how immunity rulings and similar legal defenses were treated in the case reserves that fed the triangle. The gross, ceded, and net mechanics article walks through the layering for captive structures; the pool version adds the immunity wrinkle but the rest of the math is the same.
The member allocation question
Most public entity pools allocate aggregate reserves back to individual members at each reserve study. The allocation drives next-year contribution rates, surplus-distribution decisions when the pool is overfunded, special-assessment calculations when it is not, and occasionally membership review when one member’s loss experience is materially worse than the pool average. Allocation is a practitioner convention rather than a method Friedland covers directly, so this section names practitioner convention explicitly.
The allocation approaches you will see in practice fall into a few common buckets. Pro rata to contributions is simple and auditable but unresponsive to member-level loss experience. Pro rata to reported losses rewards clean years and penalizes noisy ones, though it tends to double-count the noise from a single large claim. A mini loss-development exercise on each of the largest members, with the residual allocated to smaller members by formula, is more actuarially defensible but more expensive. Many pools use an experience-modification factor calculated on a multi-year window, which dampens the single-large-claim problem while still being responsive to genuine loss-experience differences.
The governance question every pool board should be asking, and that most do not ask clearly enough, is this: does the sum of the member-level allocations reconcile to the actuary’s aggregate estimate? The two views should match because they are describing the same liability. If they do not, the board should know why. A gap of one or two percent is usually an administrative rounding artifact or a difference in valuation date. A larger gap almost always signals that one side or the other is using a different method, a different cohort, or a different set of assumptions, and the board needs to know which one is load-bearing for the financials and for the upcoming contribution-rate cycle.
Four things that make a pool IBNR estimate go wrong
Friedland’s middle chapters systematically run each reserving method against the same portfolio under different kinds of operational change, and four of the failure modes are especially common in public entity pools. None of these is exotic; all are the predictable consequence of pooling governmental risk.
A revival-window statute reopens accident years the triangle treated as closed. This is the failure mode that distinguishes a public entity pool from almost any other self-insurance vehicle. When a state enacts a revival window for childhood sexual abuse claims, decades-old accident years that showed zero open exposure suddenly start receiving claims. The California AB 218 expansion of the abuse lookback window and the New York General Municipal Article 8 lookback for public-entity abuse claims are two recent examples that hit school-district and municipal pools especially hard. The development triangle cannot project these claims because they did not exist in the historical pattern. The actuary needs to add an explicit loading for latent exposure that sits outside the standard projection, and the loading requires assumptions about claim frequency, severity, and the scope of the specific statute that cannot be derived from the pool’s own triangle.
Membership change shifts the underlying mix. A pool that lost a large member two years ago, added three new ones last year, or absorbed a sister pool through a merger is not the same risk-financing vehicle that produced the oldest accident years in the triangle. Friedland’s general warning on a changing mix of business applies directly, and the fix is reweighting historical data, applying expected claims to the current mix for recent years, or segmenting by cohort. Most pool reserve reports handle membership change poorly because the operational data about which members were in the pool when each loss occurred is harder to assemble than the dollar triangle itself.
Case reserve philosophy changes at the TPA or in-house claims unit. The same Berquist-Sherman case-adequacy distortion that affects every self-insured book hits pools too. If the pool’s TPA changed its approach to setting case reserves, or the pool brought claims in-house and the new in-house team reserves differently, the reported chain ladder will be biased. Friedland’s recommended fix is the Berquist-Sherman case adjustment, restating historical reported claims to a consistent case adequacy level (Friedland, p. 283). The pool-specific complication is that case reserve practice can also be moved by immunity rulings, as described above, and the actuary needs to separate the two effects to adjust them correctly.
Settlement speed shifts after an operational change. Paid chain ladder has the mirror problem. A new claims initiative pushing faster settlements compresses the paid pattern. A TPA transition introducing processing delays stretches it. A state-level change to the workers compensation medical fee schedule shifts both the timing and the size of medical payments. Friedland’s recommended fix is the Berquist-Sherman disposal-rate adjustment, restating the paid triangle on a constant-disposal-rate basis (Friedland, p. 287). If your pool changed TPAs during the experience period and the reserve report does not mention it, that is a gap worth a direct question.
Underneath all four is the same principle that runs through every reserve diagnostic: a lag-based reserve estimate is a statement about stability, and when the environment changes the estimate needs to be adjusted rather than just re-run.
What the report should deliver
A reserve study that earns its engagement fee gives the pool, at a minimum, seven things in writing.
An aggregate point estimate for each line and for the pool overall. The single numbers that flow into the pool’s audited financial statements. ASOP 43 governs this estimate and requires the actuary to produce an actuarial central estimate, meaning an expected value over the range of outcomes reasonably possible given the data and methods (Friedland, p. 10).
A reasonable range around the aggregate point. Not a statistical confidence interval, but the window of estimates a different qualified actuary could defend on the same data. The point estimate versus range explainer walks through the distinction for a non-actuarial reader. Public entity pools typically have wider ranges than private captives for the same volume of business because latent exposure adds an uncertainty band that private commercial books do not carry.
A member-level allocation of the aggregate number, with a written reconciliation showing how the member pieces sum to the aggregate. See the earlier section on why this matters for governance and for contribution-rate setting.
A gross-to-net bridge for every accident year, walking through the retention, the per-occurrence excess reinsurance, and the aggregate stop-loss in enough detail that a board member can follow it without a calculator. The bridge should explicitly note how immunity rulings and similar legal defenses were reflected in case reserves.
Separate triangles by line of business at the pool level. A combined triangle that blends workers compensation with general liability with property is producing one number from four very different reserving problems and is not auditable.
Latent exposure disclosure stating whether the reserve includes a loading for latent abuse, civil rights, or environmental claims, what the basis for the loading is, and which revival-window statutes affect the pool’s member jurisdictions. If the loading is excluded, the report should explain why.
Operational change disclosure documenting any TPA change, in-house claims transition, retention change, membership change, or material legal-environment shift during the experience period. The report should state how each change was reflected in the data adjustments and method selections.
One limitation worth naming. Most public entity pools do not file Statements of Actuarial Opinion under ASOP 36 because they are not licensed insurers, but the pool’s auditor will still expect actuarial support for the IBNR liability under GASB Statement No. 10. The substance of what the actuary delivers should look much like an SAO even if the document does not carry the formal label.
Cadence and what to expect to pay
Annual reserve studies are the floor for almost every public entity pool, driven by the audit cycle rather than by an insurance-department requirement. A pool with a June 30 fiscal year-end typically receives a preliminary actuarial estimate in late summer for accrual purposes and a final report in early fall for the audited financials. Most pools above a meaningful size now also receive quarterly roll-forwards, in part because the contribution-rate cycle and the political calendar do not always line up with the audit cadence.
A quarterly roll-forward is typically a week of actuarial time, not a month, because the existing method selections are reapplied to a refreshed triangle rather than rebuilt from scratch. For a pool with a fast-moving exposure environment (a state with active revival-window legislation, for example, or a transit-authority member that just experienced a serious incident), more frequent visibility is almost always worth the marginal cost relative to the audit risk of discovering a material reserve miss at fiscal year end.
Engagement fees vary enough that any single dollar figure here would mislead. Calibrate against the pool’s annual contributions and surplus position, not against a generic benchmark table.
Three concrete actions
If you have read this far and want to turn it into something your pool can do this quarter, three moves will get you most of the value.
Ask your actuary for both the aggregate reserve and the member-level allocation, and verify the two reconcile. This is the governance question that most pool boards never quite ask cleanly. A crisp reconciliation confirms the pool’s actuarial process and its contribution-rate process are using the same data. A vague answer or a material gap tells you exactly what needs fixing before the next contribution cycle.
Ask specifically how latent abuse and civil-rights exposure is treated in the reserve, given the revival-window statutes in your member jurisdictions. If a loading is included, ask for the quantitative basis and the assumed scope of the statute. If a loading is excluded, ask for the written rationale. Both answers are defensible in the right circumstances. No answer at all is not.
Ask how the per-occurrence retention, the aggregate stop-loss, and any immunity rulings are sequenced in the reserve estimate. Friedland’s recommended sequence is to estimate ultimate claims limited to the per-occurrence retention first, then apply the stop-loss as a final adjustment (Friedland, p. 332). The immunity layer sits inside the gross-to-net bridge and should be addressed explicitly. If your actuary cannot walk the board through that sequence in plain English, that is a board-level conversation worth having before the next renewal.
None of these steps requires you to become an actuary. They require questions your actuary should be happy to answer, and an independent second opinion is often the cleanest way to benchmark the answers you get back. The independent reviews note has more on why a second set of eyes matters for pooled structures, and for pools specifically the benefit is that an independent actuary has no relationship with the pool’s TPA, its broker, or its excess reinsurer and can speak to the reserve number on its own merits.