ibnr-allocation-analysis-memorandum.docx

1025-00005 · Quorum Insurance Holdings · Case Assessment

ibnr-allocation-analysis-memorandum.docx DOCUMENT Litigation

4 / 37 ← → ✕

PRIVILEGED AND CONFIDENTIAL — ATTORNEY-CLIENT COMMUNICATION / ATTORNEY WORK PRODUCT This memorandum is protected by the attorney-client privilege and the attorney work product doctrine. It is intended solely for the use of the attorneys and client representatives identified herein and may not be disclosed to any third party without the prior written consent of Calderwood & Harkness LLP.

CALDERWOOD & HARKNESS LLP One Liberty Plaza, 55th Floor New York, NY 10006

MEMORANDUM

To: Douglas Carmichael, Responsible Partner; Jeffrey Park, Lead Associate

From: Yuki Nakamura, Associate

Date: March 12, 2026

Re: IBNR Allocation Analysis — Article 5 "Losses Incurred" Definition, Treaty No. PA-QCS-2019-EXL-0044

Matter No.: 1025-00005

Client: Quorum Insurance Holdings Corp. / Quorum Casualty & Surety Company

Right-click to update Table of Contents

I. Executive Summary

This memorandum analyzes Pinnacle Atlantic Reinsurance Ltd.'s ("Pinnacle Atlantic") second ground for dispute in connection with the ongoing reinsurance recovery dispute with Quorum Casualty & Surety Company ("QC&S"): the allegation that QC&S improperly allocated $22,300,000 in incurred but not reported ("IBNR") reserve strengthening to Treaty Year 2022 (July 1, 2022 – June 30, 2023) ("TY 2022") that Pinnacle Atlantic contends should have been attributed to Treaty Year 2021 (July 1, 2021 – June 30, 2022) ("TY 2021"). The dispute arises under Treaty No. PA-QCS-2019-EXL-0044 (the "Treaty"), an aggregate excess-of-loss reinsurance treaty between QC&S, as cedent, and Pinnacle Atlantic, as reinsurer.

If Pinnacle Atlantic's reallocation theory prevails, the financial impact would be a reduction in TY 2022 ceded losses from $69,700,000 to $47,400,000 — a decrease of $22,300,000 in the total disputed reinsurance recoverable. In its pre-suit demand letter dated February 14, 2026, Pinnacle Atlantic does not acknowledge that the reallocation of $22,300,000 to TY 2021 would increase TY 2021 aggregate net incurred losses above the $120,000,000 retention, thereby creating a potential $6,900,000 cession for TY 2021 that QC&S could assert as an offset. The net exposure on this allocation issue is therefore $15,400,000, not $22,300,000.

Based on the analysis set forth below, this memorandum concludes that QC&S has a strong but not dispositive position that the IBNR allocation to TY 2022 is consistent with Article 5.2 of the Treaty, the actuarial methodology employed by Velthorn Actuarial Advisors LLC ("Velthorn"), applicable actuarial standards of practice, and New York reinsurance case law. However, certain ambiguities in Article 5.2 and the factual record regarding the timing of loss development create a moderate risk that an arbitration panel or court could adopt Pinnacle Atlantic's interpretation. The risk assessment is sensitive to the precise treaty language — in particular, whether Article 5.2 anchors the "losses incurred" definition to the accident year of underlying claims or to the reporting period in which reserves are recognized — and to the factual question of whether any portion of the $22,300,000 can be traced to identifiable accident-year 2021 occurrences.

This memorandum should be read in conjunction with the companion memoranda on the late notice issue (Meridian Docks claims) and the follow-the-fortunes doctrine under Article 11 of the Treaty, as these issues intersect with the IBNR allocation question in several important respects.

II. Background and Factual Summary

A. The 2019 Treaty Structure

Treaty No. PA-QCS-2019-EXL-0044, effective July 1, 2019, is an aggregate excess-of-loss reinsurance treaty between QC&S, as cedent, and Pinnacle Atlantic, as reinsurer. The Treaty was placed through Broadreach Intermediaries Inc. ("Broadreach"), with Jamie Castellano serving as the Account Executive responsible for the placement. On behalf of QC&S, the Treaty was negotiated primarily by Tobias Engström, Senior Vice President, Reinsurance & Capital Markets, with actuarial support from Velthorn and claims input from Diana Kessler-Hoff, Vice President, Claims & Reinsurance Recoveries.

The Treaty provides aggregate excess-of-loss reinsurance coverage for three lines of business written by QC&S: (i) commercial general liability, (ii) excess and umbrella liability, and (iii) professional liability. These lines constitute a significant portion of QC&S's casualty book and are characterized by long-tail loss development patterns, particularly the professional liability and excess/umbrella lines. The long-tail nature of these coverages is central to the IBNR allocation dispute, as it means that losses arising from insured events occurring in a given accident year may not be fully reported or reserved for several years following the accident year.

Under the Treaty, QC&S retains the first $120,000,000 in aggregate net incurred losses per treaty year. Pinnacle Atlantic's coverage applies to aggregate net incurred losses in excess of the $120,000,000 retention, up to a cession limit of $200,000,000 per treaty year. The maximum potential recovery from Pinnacle Atlantic for any single treaty year is therefore $200,000,000, representing losses from $120,000,000 to $320,000,000 in aggregate net incurred losses.

Each treaty year runs from July 1 through June 30 of the following calendar year. The Treaty is governed by New York law, as specified in Article 21, and contains an arbitration clause in Article 22 providing for disputes to be resolved through arbitration conducted under the rules of ARIAS-U.S. (the AIDA Reinsurance and Insurance Arbitration Society — U.S.).

The Treaty does not have a fixed termination date but is subject to annual renewal by mutual agreement. The Treaty has been renewed for each treaty year from TY 2019 (July 1, 2019 – June 30, 2020) through TY 2024 (July 1, 2024 – June 30, 2025), although the parties have been in a standstill regarding TY 2025 renewal pending resolution of the current dispute.

B. Treaty Years at Issue

The instant dispute involves cessions under two treaty years, with the IBNR allocation issue specifically implicating TY 2022 and potentially TY 2021.

Treaty Year 2022 (July 1, 2022 – June 30, 2023). QC&S reported aggregate net incurred losses for TY 2022 of $189,700,000. After application of the $120,000,000 per-treaty-year retention, QC&S claimed ceded losses of $69,700,000 ($189,700,000 minus $120,000,000). This represents the largest cession in the history of the Treaty. The aggregate net incurred loss figure of $189,700,000 includes the disputed $22,300,000 in IBNR reserve strengthening that is the subject of this memorandum. It also includes case reserves on reported claims, paid losses, and other IBNR components that are not currently disputed by Pinnacle Atlantic.

Treaty Year 2023 (July 1, 2023 – June 30, 2024). QC&S reported aggregate net incurred losses for TY 2023 of $128,700,000, resulting in a ceded loss of $8,700,000 ($128,700,000 minus $120,000,000). While the TY 2023 cession is part of the broader dispute (Pinnacle Atlantic challenges the entire $78,400,000 recoverable), the IBNR allocation issue addressed in this memorandum concerns only TY 2022. The TY 2023 cession is addressed primarily in the companion memoranda on late notice and follow the fortunes.

Treaty Year 2021 (July 1, 2021 – June 30, 2022). QC&S reported aggregate net incurred losses for TY 2021 of $104,600,000. Because this figure is below the $120,000,000 retention, no cession attached for TY 2021, and no billings were submitted to Pinnacle Atlantic for this treaty year. This is significant because Pinnacle Atlantic's reallocation theory would, by its own logic, increase TY 2021 aggregate net incurred losses to $126,900,000 — above the retention — and thereby create a new cession obligation for TY 2021, a consequence that Pinnacle Atlantic's demand letter fails to address.

The total disputed reinsurance recoverable across the two treaty years for which cessions were billed is $78,400,000, consisting of $69,700,000 for TY 2022 and $8,700,000 for TY 2023. This entire amount was paid by Pinnacle Atlantic across seven billing installments submitted between October 15, 2022 and March 29, 2024. Pinnacle Atlantic's February 14, 2026 demand letter seeks recovery of $22,300,000 on the IBNR allocation ground alone, plus additional amounts on the late notice and follow-the-fortunes grounds, potentially up to the full $78,400,000.

C. The IBNR Dispute

The IBNR allocation dispute was first formally raised by Pinnacle Atlantic in a reservation-of-rights letter dated December 12, 2025, sent by its external counsel, Hargrove, Fitch & Calloway LLP (lead partner Nathaniel R. Calloway III, with senior associate Priya Subramanian). That letter reserved Pinnacle Atlantic's right to challenge the IBNR allocation methodology used by QC&S and Velthorn in computing the TY 2022 aggregate net incurred loss. The reservation-of-rights letter was followed by a formal pre-suit demand letter dated February 14, 2026, also from Hargrove, Fitch & Calloway, setting forth three grounds for Pinnacle Atlantic's claim for recovery of ceded loss payments: (1) late notice under Article 8 (the Meridian Docks claims), (2) improper IBNR allocation under Article 5.2 (the subject of this memorandum), and (3) breach of the follow-the-fortunes/follow-the-settlements obligations under Article 11.

Pinnacle Atlantic's demand letter alleges, in substance, that $22,300,000 in IBNR reserve strengthening included by QC&S in the TY 2022 aggregate net incurred loss figure improperly relates to accident-year 2021 underlying exposures — that is, to insured events that occurred during the period July 1, 2021 through June 30, 2022 — and should have been allocated to TY 2021 under the terms of Article 5.2. Pinnacle Atlantic's theory, as articulated in the demand letter, is that Article 5.2 requires losses to be allocated to the treaty year corresponding to the accident year in which the underlying insured events occurred, and that IBNR reserve development is merely a lagging recognition of losses that were already "incurred" in the accident year, not new losses "incurred" in the later treaty year when the reserves were booked or strengthened.

QC&S's position, supported by Velthorn's actuarial certificates and methodology, is that the IBNR development relates to accident-year 2022 claims whose development factors crystallized during TY 2022 reporting periods, and that even to the extent some of the underlying insured events may have occurred during accident-year 2021, Article 5.2 allocates IBNR to the treaty year during which the reserve development is recognized on the cedent's books in accordance with the cedent's actuarial methodology. This recognition-based approach, QC&S contends, is the methodology contemplated by Article 5.2 and is consistent with the actuarial methodology applied by Velthorn across all treaty years without prior objection from Pinnacle Atlantic.

The IBNR analysis and actuarial certificates underlying the disputed cessions were prepared by Velthorn Actuarial Advisors LLC, an independent actuarial consulting firm located at 75 Ashbury Asset Advisors, Suite 1400, Boston, Massachusetts 02109. The principal actuary responsible for the QC&S engagement is Dr. Stephen Tanaka, FCAS, MAAA. Dr. Tanaka has served as the appointed actuary for QC&S's reinsurance reserve analyses since the inception of the Treaty in 2019 and has provided actuarial certificates supporting each of the billing statements submitted by QC&S to Pinnacle Atlantic through Broadreach.

D. Financial Impact of Reallocation

The financial consequences of Pinnacle Atlantic's reallocation theory are significant and merit detailed analysis. The following numerical modeling illustrates the impact under various scenarios.

Current QC&S-Reported Position (Status Quo):

Pinnacle Atlantic's Reallocation Scenario (Without TY 2021 Offset):

If $22,300,000 is reallocated from TY 2022 to TY 2021:

Under Pinnacle Atlantic's theory as stated in its demand letter, the net reduction in the disputed recoverable would be $22,300,000 ($78,400,000 current total minus $56,100,000 revised total). However, this figure does not account for the TY 2021 cession that would result from the reallocation.

Reallocation Scenario With TY 2021 Offset:

A critical and apparently overlooked consequence of Pinnacle Atlantic's reallocation theory is that the reallocation of $22,300,000 to TY 2021 would increase TY 2021 aggregate net incurred losses from $104,600,000 to $126,900,000 — an amount that exceeds the $120,000,000 retention by $6,900,000. This would create a $6,900,000 cessed loss for TY 2021 that QC&S never billed because, under QC&S's allocation methodology, TY 2021 aggregate net incurred losses remained below the retention.

If QC&S asserts this offset (either as a counterclaim or as a setoff against Pinnacle Atlantic's recovery):

The net reduction from the current total of $78,400,000 would be $15,400,000 ($78,400,000 minus $63,000,000), not the $22,300,000 that Pinnacle Atlantic claims on this ground. Pinnacle Atlantic's demand letter does not acknowledge this arithmetic consequence of its own reallocation theory. This is a significant gap in Pinnacle Atlantic's position and should be preserved as a defensive and counter-recovery measure in any response, arbitration demand, or counterclaim.

Impact on QC&S Statutory Surplus:

As of December 31, 2025, QC&S's statutory surplus stood at approximately $1,420,000,000. The Cartwright National Bank credit agreement contains a minimum surplus covenant of $1,100,000,000, providing current headroom of approximately $320,000,000. If QC&S were required to write down $22,300,000 in reinsurance recoverables (the gross exposure on the IBNR allocation issue), QC&S's statutory surplus would be reduced to approximately $1,397,700,000, with covenant headroom of approximately $297,700,000. While this reduction is material from a financial reporting perspective, it would not trigger a covenant breach under the credit agreement. If the net exposure (accounting for the TY 2021 offset) of $15,400,000 is used, the surplus impact is correspondingly reduced. Neither scenario presents an existential financial risk to QC&S, but the reputational and precedential implications of an adverse ruling on this issue are significant, particularly for future treaty year renewals and other reinsurance relationships.

III. Article 5.2 — "Losses Incurred" Definition

A. Treaty Language

Article 5 of the Treaty is titled "Losses Incurred" and provides the framework for calculating the aggregate net incurred loss for each treaty year — the figure that determines whether the $120,000,000 retention has been breached and the amount, if any, of ceded losses owed by Pinnacle Atlantic. The critical operative provision for purposes of this dispute is Article 5.2, which defines the components of "losses incurred" and specifies the methodology for calculating the aggregate net incurred loss.

Article 5.2 provides, in relevant part:

"For purposes of this Treaty, the term 'Losses Incurred' shall mean, with respect to each Treaty Year, the sum of: (a) all loss and loss adjustment expense payments made during such Treaty Year on claims arising under Covered Policies, net of salvage, subrogation, and other recoveries; (b) case reserves established or maintained as of the end of such Treaty Year on reported claims arising under Covered Policies; and (c) reserves for losses incurred but not reported ('IBNR') as determined by the Cedent's appointed actuary in accordance with generally accepted actuarial principles and applicable actuarial standards of practice, as recognized and reported by the Cedent during such Treaty Year."

Several features of this provision are critical to the analysis:

First, Article 5.2(a) addresses paid losses. The allocation of paid losses to treaty years is straightforward: they are allocated to the treaty year during which the payment was made. This is a cash-basis allocation that creates no ambiguity with respect to treaty year assignment.

Second, Article 5.2(b) addresses case reserves on reported claims. These are allocated to the treaty year during which the reserves are "established or maintained as of the end of such Treaty Year." Again, the temporal anchor is the treaty year in which the reserve exists, not the accident year of the underlying claim. This is consistent with a recognition-based or reporting-period allocation methodology.

Third, and most critically, Article 5.2(c) addresses IBNR reserves. The provision states that IBNR is to be "determined by the Cedent's appointed actuary in accordance with generally accepted actuarial principles and applicable actuarial standards of practice." This language vests significant discretion in the cedent's actuary (here, Velthorn) and anchors the determination to actuarial standards. The trailing clause — "as recognized and reported by the Cedent during such Treaty Year" — is the key phrase in dispute. QC&S reads this clause as allocating IBNR to the treaty year in which the IBNR is "recognized and reported" by QC&S on its books, regardless of the accident year of the underlying claims. Pinnacle Atlantic reads the clause differently, arguing that it merely specifies the timing of reporting and does not override what it contends is an implicit accident-year allocation framework embedded in the concept of "losses incurred."

The Treaty does not contain a separate definition of "accident year" or an explicit provision requiring the mapping of IBNR reserves to specific accident years for purposes of treaty year allocation. Article 5.1 defines "Treaty Year" as the twelve-month period from July 1 through June 30, and Article 5.3 defines "Covered Policies" as the policies incepting or renewing during the Treaty Year. Article 5.4 addresses "allocated loss adjustment expenses" and "unallocated loss adjustment expenses," neither of which is relevant to the IBNR allocation issue.

B. Ambiguity Analysis

The central interpretive question is whether Article 5.2(c) allocates IBNR to treaty years on a recognition basis (the treaty year in which the IBNR reserve development is recognized and reported by the cedent) or on an accident-year basis (the treaty year corresponding to the accident year in which the underlying insured events occurred). The language of Article 5.2(c) is susceptible to both readings, creating a genuine ambiguity that may require extrinsic evidence to resolve.

Arguments supporting QC&S's recognition-based reading. The phrase "as recognized and reported by the Cedent during such Treaty Year" most naturally reads as a temporal allocation mechanism: IBNR is assigned to the treaty year during which the cedent recognizes and reports the reserve. This reading is consistent with the structure of Article 5.2 as a whole, which allocates paid losses to the treaty year of payment (Article 5.2(a)) and case reserves to the treaty year of establishment or maintenance (Article 5.2(b)). All three components — paid losses, case reserves, and IBNR — are allocated on a consistent reporting-period basis, creating a coherent framework. Under this reading, the aggregate net incurred loss for any treaty year is a snapshot of the cedent's financial position as of the end of that treaty year, reflecting all losses recognized and reported during the period. This is a practical and administratively workable approach that is commonly used in reinsurance accounting.

Arguments supporting Pinnacle Atlantic's accident-year reading. Pinnacle Atlantic may argue that the phrase "losses incurred" inherently refers to losses arising from insured events that have occurred (been "incurred") and that the concept of "incurred" is tied to the occurrence of the insured event, not to the timing of reserve recognition. Under this reading, "as recognized and reported by the Cedent during such Treaty Year" merely describes the process by which the cedent communicates its IBNR estimates to the reinsurer — it is a reporting requirement, not an allocation mechanism. Pinnacle Atlantic may further argue that the reference to "generally accepted actuarial principles" in Article 5.2(c) incorporates the actuarial convention of attributing IBNR to the accident year of the underlying claims, because that is how IBNR is typically analyzed in actuarial loss reserving (i.e., by constructing development triangles organized by accident year). This reading finds some support in the fact that Article 5.3's definition of "Covered Policies" is organized by inception or renewal date, suggesting that the Treaty contemplates a temporal mapping of losses to the periods in which the underlying risk was assumed.

Contra proferentem analysis. Under New York law, the principle of contra proferentem provides that ambiguous contractual provisions are construed against the drafter. The identity of the drafter of Article 5.2 is a factual question that requires further investigation. The Treaty was placed through Broadreach Intermediaries Inc., and it is typical in reinsurance placements for the intermediary to prepare the treaty wording based on a combination of cedent-provided terms, market-standard language, and reinsurer-negotiated modifications. If the treaty language was principally drafted by Broadreach at QC&S's direction, contra proferentem would favor Pinnacle Atlantic's reading. Conversely, if the language was derived from Pinnacle Atlantic's standard terms or was the product of mutual negotiation, the principle may carry less weight. We should review the slip history, correspondence files, and any mark-ups exchanged during the placement process to determine the drafting history of Article 5.2. Tobias Engström and Jamie Castellano should be able to provide context on this point.

Express reference to actuarial methodology. Article 5.2(c) contains a general reference to "generally accepted actuarial principles and applicable actuarial standards of practice" but does not specify a particular actuarial methodology for IBNR allocation (e.g., accident-year allocation versus recognition-based allocation). This silence cuts both ways. It supports QC&S's position that the actuary has discretion in selecting the allocation methodology. But it also supports Pinnacle Atlantic's position that the Treaty did not intend to create a novel or idiosyncratic allocation method and instead intended to rely on what Pinnacle Atlantic may characterize as the "default" actuarial practice of accident-year allocation.

C. QC&S's Interpretation

QC&S reads Article 5.2 as allocating IBNR to the treaty year in which the reserve development is recognized in the cedent's books — that is, the treaty year during which actuarial development factors indicate the need for increased IBNR. This is a "recognition-based" or "reporting-period" allocation that treats IBNR as a component of the cedent's loss position as of the end of each treaty year, rather than as a historical attribution of losses to prior accident years.

Under this reading, the $22,300,000 in IBNR development was properly booked during TY 2022 because Velthorn's actuarial analysis determined that the development factors crystallized in the reporting periods falling within TY 2022 (July 1, 2022 through June 30, 2023). The development factors used by Velthorn are derived from QC&S's historical loss triangles and reflect the pattern of loss emergence over time. As development data accumulated during TY 2022, Velthorn's analysis indicated that additional IBNR reserves were warranted, and these reserves were recognized and reported by QC&S during TY 2022. Under QC&S's interpretation of Article 5.2(c), this recognition event is the trigger for allocating the IBNR to TY 2022.

QC&S's interpretation is supported by several considerations:

First, the plain language of Article 5.2(c) — "as recognized and reported by the Cedent during such Treaty Year" — expressly ties IBNR allocation to the act of recognition and reporting during a specific treaty year.

Second, the structural consistency of Article 5.2 supports a reporting-period allocation: paid losses are allocated to the treaty year of payment, case reserves to the treaty year of establishment, and IBNR to the treaty year of recognition. Each component is allocated based on when it appears on the cedent's books during the treaty year, not based on the accident year of the underlying claims.

Third, a recognition-based approach is the most practical and administrable methodology for an aggregate excess-of-loss treaty. IBNR reserves, by their very nature, relate to claims that have not yet been reported. Until the actuarial analysis indicates the need for reserves, the losses cannot be identified at the individual-claim level, making accident-year attribution impractical or speculative. The recognition-based approach avoids the need for retroactive reallocation of IBNR as additional claim information emerges.

Fourth, QC&S's interpretation is consistent with the economic structure of the Treaty. The retention of $120,000,000 per treaty year is designed to capture the cedent's expected loss level for each period. If IBNR development is allocated to the treaty year in which it is recognized, the aggregate net incurred loss for each treaty year reflects the cedent's actual financial experience during that period, including the impact of emerging development. This alignment between financial experience and cession calculations is consistent with the commercial purpose of the Treaty.

D. Pinnacle Atlantic's Interpretation

Pinnacle Atlantic reads Article 5.2 as requiring allocation of IBNR to the accident year of the underlying claims to which the reserves relate. Under this "accident-year" allocation, the $22,300,000 relates to accident-year 2021 exposures — losses arising from insured events that occurred during the period July 1, 2021 through June 30, 2022 — and must be attributed to TY 2021 regardless of when the IBNR development was recognized on QC&S's books.

Pinnacle Atlantic's theory, as articulated in the February 14, 2026 demand letter, rests on the following propositions:

First, Pinnacle Atlantic contends that "losses incurred" is an industry term of art that refers to losses arising from insured events that have actually occurred. The word "incurred," in Pinnacle Atlantic's view, refers to the occurrence of the underlying loss — not the recognition of a reserve for that loss. Under this reading, a loss is "incurred" when the insured event happens, not when the cedent's actuary determines that a reserve is warranted.

Second, Pinnacle Atlantic argues that IBNR reserve development merely reflects the maturation of already-existing losses that were "incurred" in a prior treaty year. The fact that the losses were not reported or reserved until a later period does not change when they were "incurred." The reserve strengthening during TY 2022 was simply the belated recognition of losses that were already incurred during TY 2021, and those losses should be attributed to TY 2021 for purposes of the Treaty.

Third, Pinnacle Atlantic contends that Velthorn's own actuarial workpapers likely organize the IBNR analysis by accident year, using loss development triangles that attribute emerging development to the accident year in which the underlying claims arose. If Velthorn's internal methodology attributes the $22,300,000 to accident-year 2021 claims, Pinnacle Atlantic argues, the Treaty requires the same attribution for cession calculation purposes.

Fourth, Pinnacle Atlantic argues that the phrase "as recognized and reported by the Cedent during such Treaty Year" in Article 5.2(c) is a procedural reporting requirement — it specifies when the cedent must report its IBNR estimates to the reinsurer, not how the IBNR should be allocated among treaty years. The allocation question, Pinnacle Atlantic contends, is governed by the concept of "losses incurred," which inherently ties to accident-year attribution.

Pinnacle Atlantic's interpretation, while not without textual support, faces significant challenges. As noted above, the structural consistency of Article 5.2 — allocating all three components (paid losses, case reserves, and IBNR) on a reporting-period basis — undermines the argument that IBNR alone should be subject to accident-year allocation. Moreover, Pinnacle Atlantic's interpretation would require retroactive reallocation of IBNR across treaty years whenever emerging development data allows for more precise accident-year attribution, creating administrative complexity and uncertainty that is inconsistent with the practical operation of a treaty that uses rolling twelve-month reporting periods.

IV. Velthorn Actuarial Advisors' Methodology

A. Overview of Velthorn's Analysis

Dr. Stephen Tanaka, FCAS, MAAA, is the Principal Actuary at Velthorn Actuarial Advisors LLC who performed the loss development and reserve analysis underlying the TY 2022 cession. Dr. Tanaka holds the Fellow of the Casualty Actuarial Society (FCAS) designation and is a Member of the American Academy of Actuaries (MAAA). He has over twenty years of experience in property and casualty actuarial consulting, with particular expertise in reinsurance reserving, loss development analysis, and actuarial opinions for regulatory and contractual purposes. Dr. Tanaka has served as the appointed actuary for QC&S's reinsurance reserve analyses since the inception of the Treaty in 2019.

For each treaty year, Velthorn prepares an actuarial analysis of QC&S's aggregate net incurred losses across the three covered lines of business. This analysis includes a review of paid loss data, case reserve data, and IBNR reserve estimates. Based on this analysis, Velthorn prepares actuarial certificates that are attached to the billing statements submitted by QC&S to Pinnacle Atlantic through Broadreach. The actuarial certificates represent Dr. Tanaka's professional opinion that the aggregate net incurred loss figures are reasonable and have been determined in accordance with generally accepted actuarial principles and applicable actuarial standards of practice, as required by Article 5.2(c) of the Treaty.

Velthorn's analysis employs a blend of three recognized actuarial methodologies for estimating ultimate losses and IBNR:

1. Paid Loss Development Method: This method projects ultimate losses by applying development factors to paid loss data organized by accident year. The development factors are derived from QC&S's historical loss triangles, which track the emergence of paid losses over successive reporting periods. This method is most reliable for mature accident years with stable payment patterns but may be less reliable for recent accident years with limited payment history.

2. Incurred Loss Development Method: This method is similar to the paid loss development method but uses incurred loss data (paid losses plus case reserves) as the starting point. Incurred losses typically exhibit less volatility than paid losses in the early development periods because case reserves provide an early estimate of ultimate loss values. Velthorn applies development factors to the incurred loss data to project ultimate losses by accident year.

3. Bornhuetter-Ferguson Method: This method is a credibility-weighted blend of the development method and an a priori expected loss estimate. It is particularly useful for recent accident years where the loss development data is immature and the development method alone may produce volatile or unreliable estimates. The Bornhuetter-Ferguson method tempers the development method projections with a prior expectation of loss levels, providing a more stable estimate in the early stages of development.

Velthorn applies these three methods to each covered line of business separately and then selects a point estimate of ultimate losses and IBNR for each accident year based on professional judgment, considering the relative reliability of each method given the maturity of the data and the characteristics of the line of business.

B. IBNR Development Allocation Approach

The IBNR allocation methodology employed by Velthorn is central to this dispute. Velthorn allocated IBNR development to treaty years based on the reporting period in which development factors indicated that reserve changes were appropriate — consistent with QC&S's interpretation of Article 5.2(c). Under Velthorn's approach, when the actuarial analysis performed during a given treaty year indicates that IBNR reserves should be increased (or decreased), the resulting change in IBNR is allocated to the treaty year in which the analysis is performed and the reserve change is recognized on QC&S's books.

The $22,300,000 in disputed IBNR represents reserve strengthening that was recognized during the period July 2022 through June 2023 (TY 2022). This reserve strengthening was driven by emerging claim frequency and severity data received by QC&S and analyzed by Velthorn during that period. Specifically, as QC&S's claims department received new claim reports and updated information on existing claims during TY 2022, Velthorn incorporated this data into its actuarial models and determined that the previously estimated IBNR was insufficient. The resulting $22,300,000 increase in IBNR was reflected in the actuarial certificates supporting the TY 2022 billing statements.

Velthorn's position, as communicated by Dr. Tanaka in his written actuarial analyses and supporting documentation, is that while some of the underlying insured events contributing to the $22,300,000 in IBNR development may have occurred during accident-year 2021 (i.e., events between July 1, 2021 and June 30, 2022), the loss development patterns did not support IBNR recognition until the actuarial evaluations performed during TY 2022. In other words, the data available as of June 30, 2022 (the end of TY 2021) did not indicate the need for the additional $22,300,000 in reserves; it was only as additional loss data emerged during TY 2022 that the actuarial analysis supported the reserve increase.

This temporal distinction is critical. In long-tail casualty lines such as commercial general liability, excess/umbrella liability, and professional liability, it is common for significant loss development to occur well after the accident year. Claims may not be reported for months or years after the insured event, and even reported claims may undergo substantial reserve changes as litigation or investigation progresses. The loss development factors used by Velthorn are based on QC&S's historical loss triangles for each covered line and reflect the typical pattern of loss emergence over time. These development factors are not static; they are updated periodically as additional data becomes available. The $22,300,000 in IBNR development reflects the application of these updated development factors to the loss data available during TY 2022.

C. Consistency with Prior Practice

A significant element of QC&S's defense is the consistency of Velthorn's allocation methodology across treaty years. If the same recognition-based allocation methodology was used in TY 2019, TY 2020, and TY 2021 without objection from Pinnacle Atlantic, this establishes a course of dealing that supports QC&S's interpretation of Article 5.2 and undercuts Pinnacle Atlantic's claim that the methodology is improper.

We need to verify with Velthorn and with QC&S's internal records that the same approach was consistently applied. Based on preliminary conversations with Diana Kessler-Hoff, the recognition-based approach has been used since the inception of the Treaty. In each prior treaty year, Velthorn allocated IBNR to the treaty year in which the reserve development was recognized, and the resulting aggregate net incurred loss figures were reported to Pinnacle Atlantic through Broadreach in the same manner as the TY 2022 figures. Pinnacle Atlantic paid the TY 2020 cessions (TY 2020 was the first treaty year in which aggregate net incurred losses exceeded the $120,000,000 retention, with a cession of approximately $14,200,000) without raising any concerns about the IBNR allocation methodology. Similarly, no IBNR allocation issues were raised during TY 2021, a year in which aggregate losses fell below the retention.

If this consistency can be confirmed through documentary evidence — specifically, through comparison of Velthorn's actuarial certificates and supporting workpapers across treaty years — it will provide a strong foundation for a course-of-dealing argument. The argument would be that Pinnacle Atlantic acquiesced in the recognition-based allocation methodology for several years, received and reviewed the actuarial certificates that reflected this methodology, paid the cessions without objection, and cannot now retroactively challenge the methodology for TY 2022.

However, we should also consider whether Pinnacle Atlantic might argue that the IBNR allocation methodology was not transparent in the actuarial certificates — that the certificates simply reported aggregate IBNR figures without disclosing the allocation methodology in sufficient detail for Pinnacle Atlantic to identify and object to the approach. This would require review of the certificates themselves to assess how much detail was provided about the allocation methodology.

D. Potential Vulnerabilities in Velthorn's Approach

While Velthorn's recognition-based allocation methodology is a legitimate actuarial approach, several potential vulnerabilities should be noted and investigated:

Traceability of IBNR to accident-year 2021 claims. If any of the $22,300,000 in IBNR development can be traced to specific underlying claims that are clearly accident-year 2021 occurrences — for example, if Velthorn's actuarial workpapers organize the IBNR by accident year and show that a substantial portion of the $22,300,000 relates to accident-year 2021 exposures — this could undermine the recognition-based allocation theory. The distinction between "the development factors crystallized during TY 2022" and "the underlying losses occurred during accident-year 2021" may be difficult to maintain if Velthorn's own internal analysis attributes the IBNR to accident-year 2021 claims. We need to review Velthorn's complete workpapers, particularly the accident-year triangles and development factor selections, to assess this vulnerability. If the workpapers show that the $22,300,000 is principally attributable to development on accident-year 2022 claims (i.e., claims arising from events occurring after July 1, 2022), this vulnerability is significantly diminished. If the workpapers show a mix of accident-year 2021 and accident-year 2022 claims, the analysis becomes more nuanced.

Timing concerns and proximity to retention threshold. Pinnacle Atlantic may argue that the timing of the IBNR recognition during TY 2022 — a year in which aggregate losses were approaching and ultimately exceeded the $120,000,000 retention — is suspicious and suggests that QC&S or Velthorn engaged in strategic reserve manipulation. This argument would be that QC&S had an economic incentive to allocate IBNR to TY 2022 (where it would contribute to a cession above the retention) rather than to TY 2021 (where aggregate losses were well below the retention and the IBNR would be absorbed entirely by QC&S). While this is a serious allegation, it would require Pinnacle Atlantic to present evidence of bad faith or manipulation, which is a heavy burden. Dr. Tanaka's credentialed professional opinion carries significant weight, and absent evidence that the actuarial analysis was influenced by non-actuarial considerations, the timing argument is speculative. Nevertheless, we should investigate whether there were any communications between QC&S's management and Velthorn regarding the allocation of IBNR to specific treaty years, any discussions about the impact of the allocation on cession calculations, or any changes in reserving philosophy that coincided with the TY 2022 reporting.

Claims department reclassifications. We should also assess whether any of the $22,300,000 in IBNR development was influenced by claims department reclassifications or changes in coding practices during TY 2022. If claims that were originally coded to accident-year 2022 were reclassified to accident-year 2021 (or vice versa) during the period, this could complicate the allocation analysis and provide fodder for Pinnacle Atlantic's argument.

Completeness of workpapers. The importance of obtaining Velthorn's complete workpapers and data exhibits cannot be overstated. The actuarial certificates provided to Pinnacle Atlantic are summary documents that do not contain the level of detail necessary to evaluate the IBNR allocation. The workpapers — including loss development triangles, development factor selections, expected loss ratio assumptions, claim count data, and any memoranda documenting Velthorn's professional judgment — are essential to understanding the actuarial basis for the allocation and to assessing whether the allocation can withstand scrutiny from an opposing actuarial expert.

V. Actuarial Standards of Practice

A. Applicable Actuarial Standards

The actuarial analysis performed by Velthorn is governed by several Actuarial Standards of Practice ("ASOPs") promulgated by the Actuarial Standards Board (the "ASB"), which is the body responsible for establishing binding standards for actuaries in the United States. Article 5.2(c) of the Treaty expressly requires that IBNR reserves be determined "in accordance with generally accepted actuarial principles and applicable actuarial standards of practice." The following ASOPs are directly relevant to the IBNR allocation analysis:

Actuarial Standard of Practice No. 43 — "Unpaid Claim Estimates." ASOP No. 43 sets forth the standards for actuaries when estimating unpaid claims, including IBNR. It applies to actuaries who issue statements of actuarial opinion on unpaid claim estimates for property and casualty insurance and reinsurance. ASOP No. 43 addresses the selection of methods, the use of data, the evaluation of assumptions, and the communication of results. Importantly, ASOP No. 43 recognizes that the actuary may use multiple methods and should exercise professional judgment in selecting a point estimate of unpaid claims. The standard does not prescribe a single allocation methodology for attributing IBNR to accident years or reporting periods; rather, it provides a framework within which the actuary exercises judgment based on the characteristics of the data and the purpose of the analysis.

Actuarial Standard of Practice No. 36 — "Statements of Actuarial Opinion Regarding Property/Casualty Loss and Loss Adjustment Expense Reserves." ASOP No. 36 is relevant to the actuarial opinions that Velthorn provided in connection with QC&S's loss reserve certifications. This standard addresses the actuary's responsibilities when preparing a statement of actuarial opinion regarding loss reserves, including the scope of the opinion, the data reviewed, the methods used, and the disclosures required. ASOP No. 36 requires the actuary to disclose any limitations on the scope of the analysis and any significant assumptions that materially affect the results.

Actuarial Standard of Practice No. 23 — "Data Quality." ASOP No. 23 sets forth standards for the data relied upon by the actuary in performing actuarial analyses. It requires the actuary to evaluate the data for reasonableness, completeness, and consistency, and to disclose any limitations in the data that may affect the results. In the context of the IBNR allocation dispute, ASOP No. 23 is relevant to the question of whether the data used by Velthorn to allocate IBNR to treaty years was sufficient and appropriate for that purpose. If the data does not support a reliable attribution of IBNR to specific accident years, this supports the use of a recognition-based allocation that relies on aggregate development factors rather than accident-year-level data.

B. Treatment of IBNR Allocation in Actuarial Practice

A fundamental point in the actuarial analysis is that the applicable ASOPs do not mandate a single method for allocating IBNR to specific accident years or reporting periods. The selection of methodology is within the actuary's professional judgment, subject to the constraints of reasonableness, consistency, and disclosure. Both accident-year allocation and recognition-based allocation methods are recognized in actuarial practice, and neither is inherently superior to the other in all circumstances.

In general actuarial practice, IBNR is most commonly estimated at the accident-year level — that is, the actuary constructs development triangles organized by accident year and estimates ultimate losses and IBNR for each accident year separately. This is the standard approach to loss reserving for purposes of financial reporting, regulatory filings, and internal management of the insurance enterprise. However, this accident-year organization of the actuarial analysis does not necessarily dictate how the resulting IBNR estimates should be allocated among contractual reporting periods (such as treaty years in a reinsurance treaty) for purposes of calculating ceded losses. The allocation to contractual periods is a distinct question that depends on the terms of the contract.

In the context of reinsurance treaties, the allocation method is typically governed by the treaty language, and the actuary's role is to apply the method consistent with the contractual framework. Where the treaty language is ambiguous or silent on the allocation method, the actuary may exercise professional judgment in selecting a methodology, considering the purpose of the analysis (cession calculation under an aggregate excess-of-loss treaty), the characteristics of the data, and any applicable industry conventions.

Both the recognition-based approach (allocating IBNR to the treaty year in which the reserve development is recognized) and the accident-year approach (allocating IBNR to the treaty year corresponding to the accident year of the underlying claims) are legitimate methodologies that find support in actuarial practice. The recognition-based approach is more commonly used in reinsurance contexts where the treaty language defines "losses incurred" by reference to the cedent's reported loss position, as it avoids the need for retroactive reallocation of IBNR among treaty years as development data matures. The accident-year approach is more commonly used in regulatory and financial reporting contexts, where the goal is to attribute losses to the periods in which the underlying risk was assumed.

C. Implications for QC&S's Defense

The actuarial standards landscape has several important implications for QC&S's defense:

If Article 5.2 is determined to be ambiguous, actuarial custom and practice may inform the interpretation. The fact that Velthorn's recognition-based approach is a legitimate actuarial methodology — not a novel or idiosyncratic method — supports QC&S's position. An arbitration panel or court considering extrinsic evidence of actuarial practice would likely find that both allocation methods are reasonable, which may lead to a finding that Article 5.2 is genuinely ambiguous and must be resolved based on other interpretive tools (contra proferentem, course of dealing, etc.).

However, if Pinnacle Atlantic retains its own actuarial expert who opines that accident-year allocation is the industry-standard method for aggregate excess-of-loss treaties and that recognition-based allocation is disfavored or unusual in this context, this could create a battle of experts. An opposing expert might argue that the recognition-based approach, while technically permissible under the ASOPs, is not the approach contemplated by the Treaty or customarily used in the aggregate excess-of-loss treaty market. The credibility and qualifications of the competing experts would be critical, and QC&S should be prepared to retain an independent actuarial expert (separate from Velthorn, which is a fact witness) to provide an opinion supporting the reasonableness of the recognition-based approach.

It is also worth noting that the ASOPs provide procedural protections for the actuary's professional judgment. ASOP No. 43 requires the actuary to document the rationale for the selected methodology and to disclose any significant assumptions. If Dr. Tanaka's workpapers and documentation demonstrate that he considered the accident-year allocation approach and concluded that the recognition-based approach was more appropriate given the Treaty language and the characteristics of the data, this would be a significant point in QC&S's favor.

VI. New York Reinsurance Case Law on IBNR Allocation

A. General Principles of Treaty Interpretation Under New York Law

New York courts interpret reinsurance treaties as contracts, applying the standard principles of contract construction that govern commercial agreements generally. These principles are well-established and may be summarized as follows:

The primary objective of contract interpretation is to give effect to the parties' intent as expressed in the language of the agreement. Where the contract language is clear and unambiguous, the court will enforce the provision according to its plain meaning, without resort to extrinsic evidence. See W.W.W. Assocs., Inc. v. Giancontieri, 77 N.Y.2d 157, 162 (1990) ("A familiar and eminently sensible proposition of law is that, when parties set down their agreement in a clear, complete document, their writing should as a rule be enforced according to its terms."). The question of whether a contractual provision is ambiguous is a question of law for the court to determine. A provision is ambiguous if it is reasonably susceptible of more than one interpretation. Id. at 162–63.

When a reinsurance treaty provision is found to be ambiguous, New York courts may consider extrinsic evidence to resolve the ambiguity, including the parties' course of dealing, industry custom and usage, the circumstances surrounding the formation of the contract, and the reasonable expectations of the parties. See Greenfield v. Philles Records, Inc., 98 N.Y.2d 562, 569 (2002). The court may also apply the doctrine of contra proferentem, construing ambiguous provisions against the party that drafted them. See Ace Fire Underwriters Ins. Co. v. ITT Indus., Inc., 15 A.D.3d 263, 264 (1st Dep't 2005). This doctrine has particular application in reinsurance disputes where one party prepared the treaty wording.

In the reinsurance context, New York courts have recognized that reinsurance treaties are specialized commercial contracts that should be interpreted in light of the customs and practices of the reinsurance industry. See Unigard Sec. Ins. Co. v. Greaves Specialty Insurance, 79 N.Y.2d 576, 581 (1992). Courts are generally reluctant to construe reinsurance treaties in a manner that produces commercially unreasonable results or that is inconsistent with the economic structure of the transaction.

The "losses incurred" definition is a critical term in aggregate excess-of-loss treaties because it directly determines the amount of ceded losses and, consequently, the reinsurer's financial obligation. As such, it has been the subject of significant reinsurance litigation, both in court and in arbitration.

B. Relevant New York Precedent on IBNR Allocation in Reinsurance

New York case law addressing the specific question of IBNR allocation between treaty years in aggregate excess-of-loss reinsurance treaties is limited, reflecting the fact that many such disputes are resolved through arbitration under ARIAS-U.S. or similar rules. However, several categories of precedent provide useful guidance:

Cedent deference in loss allocation. New York courts have generally afforded deference to the cedent's reasonable loss allocation methodology in reinsurance disputes. The principle underlying this deference is that the cedent, as the party with primary responsibility for administering the underlying insurance claims, is in the best position to determine how losses should be allocated and reported. Courts have held that a cedent's allocation methodology will be upheld unless it is shown to be unreasonable, arbitrary, or made in bad faith. See, e.g., Greaves Specialty Insurance v. ACE Am. Reinsurance Co., 361 F.3d 134, 140 (2d Cir. 2004) (applying New York law and affirming the principle that a reinsurer is generally bound by the cedent's reasonable claims-handling decisions). This principle of cedent deference is relevant to the IBNR allocation dispute because it places the burden on Pinnacle Atlantic to demonstrate that QC&S's allocation methodology is unreasonable, rather than requiring QC&S to prove that its methodology is the only correct approach.

IBNR allocation and the "losses incurred" definition. Courts and arbitration panels have addressed the treatment of IBNR in the "losses incurred" calculation under various reinsurance treaty formulations. In Employers Reinsurance Corp. v. Mid-Continent Cas. Co., No. 07 Civ. 6009 (S.D.N.Y. 2009), the court considered a dispute over the inclusion of IBNR reserves in the calculation of losses under an aggregate excess-of-loss treaty. While the specific allocation issue was not identical to the one presented here, the court emphasized the importance of the treaty language in defining "losses incurred" and held that the cedent's approach to including IBNR in the loss calculation was proper where it was consistent with the treaty's definition. Similarly, in various ARIAS-U.S. arbitrations that have been reported or discussed in secondary sources, arbitration panels have generally upheld the cedent's IBNR methodology where it was consistent with the treaty language, was applied consistently across reporting periods, and was supported by a credentialed actuary's opinion.

Follow-the-fortunes and IBNR. The "follow the fortunes" doctrine — which requires a reinsurer to follow the cedent's reasonable claims-handling and underwriting decisions — has been applied in some cases to IBNR allocation decisions. Under this doctrine, a reinsurer is bound by the cedent's good-faith determination of losses, including the allocation of IBNR, absent evidence of bad faith, fraud, or manifest error. See Vireo Life Insurance v. Certain Underwriters at Sundial Indemnity Co., 96 N.Y.2d 583, 596 (2001) (discussing the follow-the-fortunes doctrine in the reinsurance context). The application of this doctrine to the IBNR allocation issue is analyzed in greater detail in the companion memorandum on Article 11 (follow the fortunes / follow the settlements). For present purposes, it is sufficient to note that the follow-the-fortunes doctrine may provide an additional layer of protection for QC&S's IBNR allocation, even if Article 5.2 is found to be ambiguous, by requiring Pinnacle Atlantic to accept the cedent's reasonable allocation absent bad faith.

Arbitration forum considerations. Given that the Treaty contains an arbitration clause providing for resolution under ARIAS-U.S. Rules, it is important to recognize that the ultimate forum for this dispute is likely to be a reinsurance arbitration panel rather than a court. Arbitration panels in reinsurance disputes are composed of experienced industry professionals who are familiar with actuarial practices, treaty interpretation conventions, and the commercial realities of the reinsurance market. These panels tend to give significant weight to industry custom and actuarial practice in resolving treaty interpretation disputes. An arbitration panel is likely to be receptive to testimony from actuarial experts and to evaluate the reasonableness of the cedent's allocation methodology in light of industry norms, rather than engaging in a purely textual analysis of the treaty language. This forum consideration favors QC&S, as its recognition-based allocation methodology is supported by actuarial practice and was applied consistently without prior objection.

C. Application to the Treaty Dispute

Applying the foregoing legal principles to the facts of this dispute, we conclude that QC&S's allocation of the $22,300,000 in IBNR to TY 2022 based on Velthorn's recognition-based methodology is defensible under New York law, provided that three conditions are satisfied:

(i) The methodology is consistent with Article 5.2. As discussed in Section III, QC&S's interpretation of Article 5.2(c) — that IBNR is allocated to the treaty year in which it is "recognized and reported by the Cedent" — is a reasonable reading of the treaty language and may well be the best reading when considered in the context of Article 5.2 as a whole.

(ii) The methodology was applied consistently across treaty years. If the same recognition-based approach was used for TY 2019, TY 2020, and TY 2021 without modification or objection, this establishes a course of dealing that reinforces QC&S's interpretation and undercuts Pinnacle Atlantic's belated challenge.

(iii) The allocation was not made in bad faith or to manipulate cession calculations. If the IBNR allocation was driven by legitimate actuarial analysis and not by a desire to shift losses from TY 2021 (below retention) to TY 2022 (above retention) for the purpose of generating a cession, the allocation will be upheld under the cedent deference standard. Dr. Tanaka's credentialed professional opinion is strong evidence of good faith.

Pinnacle Atlantic bears the burden of demonstrating that the allocation was improper. That burden is heightened by two significant factors. First, Pinnacle Atlantic paid the full $78,400,000 across seven billing installments between October 15, 2022 and March 29, 2024, without reserving rights as to the IBNR allocation methodology. Each billing statement was accompanied by Velthorn's actuarial certificate, which reflected the recognition-based allocation methodology. Pinnacle Atlantic's failure to object at the time of payment — or at any point during the approximately two to three years between the initial billings and the December 12, 2025 reservation-of-rights letter — weakens its position and may give rise to waiver or estoppel arguments, as discussed in Section VII. Second, Pinnacle Atlantic's own demand letter does not address the logical consequences of its reallocation theory — specifically, the $6,900,000 cession that would attach for TY 2021 — which undermines the completeness and credibility of its position.

VII. Course of Dealing and Waiver / Estoppel Considerations

A. Pinnacle Atlantic's Payment History

Pinnacle Atlantic paid the full $78,400,000 in ceded losses across seven billing installments between October 15, 2022 and March 29, 2024. Each billing statement submitted by QC&S through Broadreach was accompanied by Velthorn's actuarial certificate, which set forth the aggregate net incurred loss figures and the IBNR component. Based on the information currently available, Pinnacle Atlantic paid each installment without reserving rights as to the IBNR allocation methodology. Pinnacle Atlantic did not request additional information about the allocation methodology, did not object to the inclusion of the $22,300,000 in IBNR in the TY 2022 aggregate net incurred loss, and did not indicate any disagreement with the actuarial certificates.

The first formal expression of concern regarding the IBNR allocation came in the December 12, 2025 reservation-of-rights letter from Hargrove, Fitch & Calloway LLP — more than twenty months after the final payment on March 29, 2024, and more than three years after the initial billing in October 2022. The formal demand for repayment followed on February 14, 2026.

This extended period of silence and unqualified payment raises significant waiver and estoppel issues under New York law.

Waiver. Under New York law, waiver is the vo

PRIVILEGED AND CONFIDENTIAL — ATTORNEY-CLIENT COMMUNICATION / ATTORNEY WORK PRODUCT This memorandum is protected by the attorney-client privilege and the attorney work product doctrine. It is intended solely for the use of the attorneys and client representatives identified herein and may not be disclosed to any third party without the prior written consent of Calderwood & Harkness LLP.
CALDERWOOD & HARKNESS LLP One Liberty Plaza, 55th Floor New York, NY 10006
MEMORANDUM
To: Douglas Carmichael, Responsible Partner; Jeffrey Park, Lead Associate
From: Yuki Nakamura, Associate
Date: March 12, 2026
Re: IBNR Allocation Analysis — Article 5 "Losses Incurred" Definition, Treaty No. PA-QCS-2019-EXL-0044
Matter No.: 1025-00005
Client: Quorum Insurance Holdings Corp. / Quorum Casualty & Surety Company
Right-click to update Table of Contents
I. Executive Summary
This memorandum analyzes Pinnacle Atlantic Reinsurance Ltd.'s ("Pinnacle Atlantic") second ground for dispute in connection with the ongoing reinsurance recovery dispute with Quorum Casualty & Surety Company ("QC&S"): the allegation that QC&S improperly allocated $22,300,000 in incurred but not reported ("IBNR") reserve strengthening to Treaty Year 2022 (July 1, 2022 – June 30, 2023) ("TY 2022") that Pinnacle Atlantic contends should have been attributed to Treaty Year 2021 (July 1, 2021 – June 30, 2022) ("TY 2021"). The dispute arises under Treaty No. PA-QCS-2019-EXL-0044 (the "Treaty"), an aggregate excess-of-loss reinsurance treaty between QC&S, as cedent, and Pinnacle Atlantic, as reinsurer.
If Pinnacle Atlantic's reallocation theory prevails, the financial impact would be a reduction in TY 2022 ceded losses from $69,700,000 to $47,400,000 — a decrease of $22,300,000 in the total disputed reinsurance recoverable. In its pre-suit demand letter dated February 14, 2026, Pinnacle Atlantic does not acknowledge that the reallocation of $22,300,000 to TY 2021 would increase TY 2021 aggregate net incurred losses above the $120,000,000 retention, thereby creating a potential $6,900,000 cession for TY 2021 that QC&S could assert as an offset. The net exposure on this allocation issue is therefore $15,400,000, not $22,300,000.
Based on the analysis set forth below, this memorandum concludes that QC&S has a strong but not dispositive position that the IBNR allocation to TY 2022 is consistent with Article 5.2 of the Treaty, the actuarial methodology employed by Velthorn Actuarial Advisors LLC ("Velthorn"), applicable actuarial standards of practice, and New York reinsurance case law. However, certain ambiguities in Article 5.2 and the factual record regarding the timing of loss development create a moderate risk that an arbitration panel or court could adopt Pinnacle Atlantic's interpretation. The risk assessment is sensitive to the precise treaty language — in particular, whether Article 5.2 anchors the "losses incurred" definition to the accident year of underlying claims or to the reporting period in which reserves are recognized — and to the factual question of whether any portion of the $22,300,000 can be traced to identifiable accident-year 2021 occurrences.
This memorandum should be read in conjunction with the companion memoranda on the late notice issue (Meridian Docks claims) and the follow-the-fortunes doctrine under Article 11 of the Treaty, as these issues intersect with the IBNR allocation question in several important respects.
II. Background and Factual Summary
A. The 2019 Treaty Structure
Treaty No. PA-QCS-2019-EXL-0044, effective July 1, 2019, is an aggregate excess-of-loss reinsurance treaty between QC&S, as cedent, and Pinnacle Atlantic, as reinsurer. The Treaty was placed through Broadreach Intermediaries Inc. ("Broadreach"), with Jamie Castellano serving as the Account Executive responsible for the placement. On behalf of QC&S, the Treaty was negotiated primarily by Tobias Engström, Senior Vice President, Reinsurance & Capital Markets, with actuarial support from Velthorn and claims input from Diana Kessler-Hoff, Vice President, Claims & Reinsurance Recoveries.
The Treaty provides aggregate excess-of-loss reinsurance coverage for three lines of business written by QC&S: (i) commercial general liability, (ii) excess and umbrella liability, and (iii) professional liability. These lines constitute a significant portion of QC&S's casualty book and are characterized by long-tail loss development patterns, particularly the professional liability and excess/umbrella lines. The long-tail nature of these coverages is central to the IBNR allocation dispute, as it means that losses arising from insured events occurring in a given accident year may not be fully reported or reserved for several years following the accident year.
Under the Treaty, QC&S retains the first $120,000,000 in aggregate net incurred losses per treaty year. Pinnacle Atlantic's coverage applies to aggregate net incurred losses in excess of the $120,000,000 retention, up to a cession limit of $200,000,000 per treaty year. The maximum potential recovery from Pinnacle Atlantic for any single treaty year is therefore $200,000,000, representing losses from $120,000,000 to $320,000,000 in aggregate net incurred losses.
Each treaty year runs from July 1 through June 30 of the following calendar year. The Treaty is governed by New York law, as specified in Article 21, and contains an arbitration clause in Article 22 providing for disputes to be resolved through arbitration conducted under the rules of ARIAS-U.S. (the AIDA Reinsurance and Insurance Arbitration Society — U.S.).
The Treaty does not have a fixed termination date but is subject to annual renewal by mutual agreement. The Treaty has been renewed for each treaty year from TY 2019 (July 1, 2019 – June 30, 2020) through TY 2024 (July 1, 2024 – June 30, 2025), although the parties have been in a standstill regarding TY 2025 renewal pending resolution of the current dispute.
B. Treaty Years at Issue
The instant dispute involves cessions under two treaty years, with the IBNR allocation issue specifically implicating TY 2022 and potentially TY 2021.
Treaty Year 2022 (July 1, 2022 – June 30, 2023). QC&S reported aggregate net incurred losses for TY 2022 of $189,700,000. After application of the $120,000,000 per-treaty-year retention, QC&S claimed ceded losses of $69,700,000 ($189,700,000 minus $120,000,000). This represents the largest cession in the history of the Treaty. The aggregate net incurred loss figure of $189,700,000 includes the disputed $22,300,000 in IBNR reserve strengthening that is the subject of this memorandum. It also includes case reserves on reported claims, paid losses, and other IBNR components that are not currently disputed by Pinnacle Atlantic.
Treaty Year 2023 (July 1, 2023 – June 30, 2024). QC&S reported aggregate net incurred losses for TY 2023 of $128,700,000, resulting in a ceded loss of $8,700,000 ($128,700,000 minus $120,000,000). While the TY 2023 cession is part of the broader dispute (Pinnacle Atlantic challenges the entire $78,400,000 recoverable), the IBNR allocation issue addressed in this memorandum concerns only TY 2022. The TY 2023 cession is addressed primarily in the companion memoranda on late notice and follow the fortunes.
Treaty Year 2021 (July 1, 2021 – June 30, 2022). QC&S reported aggregate net incurred losses for TY 2021 of $104,600,000. Because this figure is below the $120,000,000 retention, no cession attached for TY 2021, and no billings were submitted to Pinnacle Atlantic for this treaty year. This is significant because Pinnacle Atlantic's reallocation theory would, by its own logic, increase TY 2021 aggregate net incurred losses to $126,900,000 — above the retention — and thereby create a new cession obligation for TY 2021, a consequence that Pinnacle Atlantic's demand letter fails to address.
The total disputed reinsurance recoverable across the two treaty years for which cessions were billed is $78,400,000, consisting of $69,700,000 for TY 2022 and $8,700,000 for TY 2023. This entire amount was paid by Pinnacle Atlantic across seven billing installments submitted between October 15, 2022 and March 29, 2024. Pinnacle Atlantic's February 14, 2026 demand letter seeks recovery of $22,300,000 on the IBNR allocation ground alone, plus additional amounts on the late notice and follow-the-fortunes grounds, potentially up to the full $78,400,000.
C. The IBNR Dispute
The IBNR allocation dispute was first formally raised by Pinnacle Atlantic in a reservation-of-rights letter dated December 12, 2025, sent by its external counsel, Hargrove, Fitch & Calloway LLP (lead partner Nathaniel R. Calloway III, with senior associate Priya Subramanian). That letter reserved Pinnacle Atlantic's right to challenge the IBNR allocation methodology used by QC&S and Velthorn in computing the TY 2022 aggregate net incurred loss. The reservation-of-rights letter was followed by a formal pre-suit demand letter dated February 14, 2026, also from Hargrove, Fitch & Calloway, setting forth three grounds for Pinnacle Atlantic's claim for recovery of ceded loss payments: (1) late notice under Article 8 (the Meridian Docks claims), (2) improper IBNR allocation under Article 5.2 (the subject of this memorandum), and (3) breach of the follow-the-fortunes/follow-the-settlements obligations under Article 11.
Pinnacle Atlantic's demand letter alleges, in substance, that $22,300,000 in IBNR reserve strengthening included by QC&S in the TY 2022 aggregate net incurred loss figure improperly relates to accident-year 2021 underlying exposures — that is, to insured events that occurred during the period July 1, 2021 through June 30, 2022 — and should have been allocated to TY 2021 under the terms of Article 5.2. Pinnacle Atlantic's theory, as articulated in the demand letter, is that Article 5.2 requires losses to be allocated to the treaty year corresponding to the accident year in which the underlying insured events occurred, and that IBNR reserve development is merely a lagging recognition of losses that were already "incurred" in the accident year, not new losses "incurred" in the later treaty year when the reserves were booked or strengthened.
QC&S's position, supported by Velthorn's actuarial certificates and methodology, is that the IBNR development relates to accident-year 2022 claims whose development factors crystallized during TY 2022 reporting periods, and that even to the extent some of the underlying insured events may have occurred during accident-year 2021, Article 5.2 allocates IBNR to the treaty year during which the reserve development is recognized on the cedent's books in accordance with the cedent's actuarial methodology. This recognition-based approach, QC&S contends, is the methodology contemplated by Article 5.2 and is consistent with the actuarial methodology applied by Velthorn across all treaty years without prior objection from Pinnacle Atlantic.
The IBNR analysis and actuarial certificates underlying the disputed cessions were prepared by Velthorn Actuarial Advisors LLC, an independent actuarial consulting firm located at 75 Ashbury Asset Advisors, Suite 1400, Boston, Massachusetts 02109. The principal actuary responsible for the QC&S engagement is Dr. Stephen Tanaka, FCAS, MAAA. Dr. Tanaka has served as the appointed actuary for QC&S's reinsurance reserve analyses since the inception of the Treaty in 2019 and has provided actuarial certificates supporting each of the billing statements submitted by QC&S to Pinnacle Atlantic through Broadreach.
D. Financial Impact of Reallocation
The financial consequences of Pinnacle Atlantic's reallocation theory are significant and merit detailed analysis. The following numerical modeling illustrates the impact under various scenarios.
Current QC&S-Reported Position (Status Quo):
Pinnacle Atlantic's Reallocation Scenario (Without TY 2021 Offset):
If $22,300,000 is reallocated from TY 2022 to TY 2021:
Under Pinnacle Atlantic's theory as stated in its demand letter, the net reduction in the disputed recoverable would be $22,300,000 ($78,400,000 current total minus $56,100,000 revised total). However, this figure does not account for the TY 2021 cession that would result from the reallocation.
Reallocation Scenario With TY 2021 Offset:
A critical and apparently overlooked consequence of Pinnacle Atlantic's reallocation theory is that the reallocation of $22,300,000 to TY 2021 would increase TY 2021 aggregate net incurred losses from $104,600,000 to $126,900,000 — an amount that exceeds the $120,000,000 retention by $6,900,000. This would create a $6,900,000 cessed loss for TY 2021 that QC&S never billed because, under QC&S's allocation methodology, TY 2021 aggregate net incurred losses remained below the retention.
If QC&S asserts this offset (either as a counterclaim or as a setoff against Pinnacle Atlantic's recovery):
The net reduction from the current total of $78,400,000 would be $15,400,000 ($78,400,000 minus $63,000,000), not the $22,300,000 that Pinnacle Atlantic claims on this ground. Pinnacle Atlantic's demand letter does not acknowledge this arithmetic consequence of its own reallocation theory. This is a significant gap in Pinnacle Atlantic's position and should be preserved as a defensive and counter-recovery measure in any response, arbitration demand, or counterclaim.
Impact on QC&S Statutory Surplus:
As of December 31, 2025, QC&S's statutory surplus stood at approximately $1,420,000,000. The Cartwright National Bank credit agreement contains a minimum surplus covenant of $1,100,000,000, providing current headroom of approximately $320,000,000. If QC&S were required to write down $22,300,000 in reinsurance recoverables (the gross exposure on the IBNR allocation issue), QC&S's statutory surplus would be reduced to approximately $1,397,700,000, with covenant headroom of approximately $297,700,000. While this reduction is material from a financial reporting perspective, it would not trigger a covenant breach under the credit agreement. If the net exposure (accounting for the TY 2021 offset) of $15,400,000 is used, the surplus impact is correspondingly reduced. Neither scenario presents an existential financial risk to QC&S, but the reputational and precedential implications of an adverse ruling on this issue are significant, particularly for future treaty year renewals and other reinsurance relationships.
III. Article 5.2 — "Losses Incurred" Definition
A. Treaty Language
Article 5 of the Treaty is titled "Losses Incurred" and provides the framework for calculating the aggregate net incurred loss for each treaty year — the figure that determines whether the $120,000,000 retention has been breached and the amount, if any, of ceded losses owed by Pinnacle Atlantic. The critical operative provision for purposes of this dispute is Article 5.2, which defines the components of "losses incurred" and specifies the methodology for calculating the aggregate net incurred loss.
Article 5.2 provides, in relevant part:
"For purposes of this Treaty, the term 'Losses Incurred' shall mean, with respect to each Treaty Year, the sum of: (a) all loss and loss adjustment expense payments made during such Treaty Year on claims arising under Covered Policies, net of salvage, subrogation, and other recoveries; (b) case reserves established or maintained as of the end of such Treaty Year on reported claims arising under Covered Policies; and (c) reserves for losses incurred but not reported ('IBNR') as determined by the Cedent's appointed actuary in accordance with generally accepted actuarial principles and applicable actuarial standards of practice, as recognized and reported by the Cedent during such Treaty Year."
Several features of this provision are critical to the analysis:
First, Article 5.2(a) addresses paid losses. The allocation of paid losses to treaty years is straightforward: they are allocated to the treaty year during which the payment was made. This is a cash-basis allocation that creates no ambiguity with respect to treaty year assignment.
Second, Article 5.2(b) addresses case reserves on reported claims. These are allocated to the treaty year during which the reserves are "established or maintained as of the end of such Treaty Year." Again, the temporal anchor is the treaty year in which the reserve exists, not the accident year of the underlying claim. This is consistent with a recognition-based or reporting-period allocation methodology.
Third, and most critically, Article 5.2(c) addresses IBNR reserves. The provision states that IBNR is to be "determined by the Cedent's appointed actuary in accordance with generally accepted actuarial principles and applicable actuarial standards of practice." This language vests significant discretion in the cedent's actuary (here, Velthorn) and anchors the determination to actuarial standards. The trailing clause — "as recognized and reported by the Cedent during such Treaty Year" — is the key phrase in dispute. QC&S reads this clause as allocating IBNR to the treaty year in which the IBNR is "recognized and reported" by QC&S on its books, regardless of the accident year of the underlying claims. Pinnacle Atlantic reads the clause differently, arguing that it merely specifies the timing of reporting and does not override what it contends is an implicit accident-year allocation framework embedded in the concept of "losses incurred."
The Treaty does not contain a separate definition of "accident year" or an explicit provision requiring the mapping of IBNR reserves to specific accident years for purposes of treaty year allocation. Article 5.1 defines "Treaty Year" as the twelve-month period from July 1 through June 30, and Article 5.3 defines "Covered Policies" as the policies incepting or renewing during the Treaty Year. Article 5.4 addresses "allocated loss adjustment expenses" and "unallocated loss adjustment expenses," neither of which is relevant to the IBNR allocation issue.
B. Ambiguity Analysis
The central interpretive question is whether Article 5.2(c) allocates IBNR to treaty years on a recognition basis (the treaty year in which the IBNR reserve development is recognized and reported by the cedent) or on an accident-year basis (the treaty year corresponding to the accident year in which the underlying insured events occurred). The language of Article 5.2(c) is susceptible to both readings, creating a genuine ambiguity that may require extrinsic evidence to resolve.
Arguments supporting QC&S's recognition-based reading. The phrase "as recognized and reported by the Cedent during such Treaty Year" most naturally reads as a temporal allocation mechanism: IBNR is assigned to the treaty year during which the cedent recognizes and reports the reserve. This reading is consistent with the structure of Article 5.2 as a whole, which allocates paid losses to the treaty year of payment (Article 5.2(a)) and case reserves to the treaty year of establishment or maintenance (Article 5.2(b)). All three components — paid losses, case reserves, and IBNR — are allocated on a consistent reporting-period basis, creating a coherent framework. Under this reading, the aggregate net incurred loss for any treaty year is a snapshot of the cedent's financial position as of the end of that treaty year, reflecting all losses recognized and reported during the period. This is a practical and administratively workable approach that is commonly used in reinsurance accounting.
Arguments supporting Pinnacle Atlantic's accident-year reading. Pinnacle Atlantic may argue that the phrase "losses incurred" inherently refers to losses arising from insured events that have occurred (been "incurred") and that the concept of "incurred" is tied to the occurrence of the insured event, not to the timing of reserve recognition. Under this reading, "as recognized and reported by the Cedent during such Treaty Year" merely describes the process by which the cedent communicates its IBNR estimates to the reinsurer — it is a reporting requirement, not an allocation mechanism. Pinnacle Atlantic may further argue that the reference to "generally accepted actuarial principles" in Article 5.2(c) incorporates the actuarial convention of attributing IBNR to the accident year of the underlying claims, because that is how IBNR is typically analyzed in actuarial loss reserving (i.e., by constructing development triangles organized by accident year). This reading finds some support in the fact that Article 5.3's definition of "Covered Policies" is organized by inception or renewal date, suggesting that the Treaty contemplates a temporal mapping of losses to the periods in which the underlying risk was assumed.
Contra proferentem analysis. Under New York law, the principle of contra proferentem provides that ambiguous contractual provisions are construed against the drafter. The identity of the drafter of Article 5.2 is a factual question that requires further investigation. The Treaty was placed through Broadreach Intermediaries Inc., and it is typical in reinsurance placements for the intermediary to prepare the treaty wording based on a combination of cedent-provided terms, market-standard language, and reinsurer-negotiated modifications. If the treaty language was principally drafted by Broadreach at QC&S's direction, contra proferentem would favor Pinnacle Atlantic's reading. Conversely, if the language was derived from Pinnacle Atlantic's standard terms or was the product of mutual negotiation, the principle may carry less weight. We should review the slip history, correspondence files, and any mark-ups exchanged during the placement process to determine the drafting history of Article 5.2. Tobias Engström and Jamie Castellano should be able to provide context on this point.
Express reference to actuarial methodology. Article 5.2(c) contains a general reference to "generally accepted actuarial principles and applicable actuarial standards of practice" but does not specify a particular actuarial methodology for IBNR allocation (e.g., accident-year allocation versus recognition-based allocation). This silence cuts both ways. It supports QC&S's position that the actuary has discretion in selecting the allocation methodology. But it also supports Pinnacle Atlantic's position that the Treaty did not intend to create a novel or idiosyncratic allocation method and instead intended to rely on what Pinnacle Atlantic may characterize as the "default" actuarial practice of accident-year allocation.
C. QC&S's Interpretation
QC&S reads Article 5.2 as allocating IBNR to the treaty year in which the reserve development is recognized in the cedent's books — that is, the treaty year during which actuarial development factors indicate the need for increased IBNR. This is a "recognition-based" or "reporting-period" allocation that treats IBNR as a component of the cedent's loss position as of the end of each treaty year, rather than as a historical attribution of losses to prior accident years.
Under this reading, the $22,300,000 in IBNR development was properly booked during TY 2022 because Velthorn's actuarial analysis determined that the development factors crystallized in the reporting periods falling within TY 2022 (July 1, 2022 through June 30, 2023). The development factors used by Velthorn are derived from QC&S's historical loss triangles and reflect the pattern of loss emergence over time. As development data accumulated during TY 2022, Velthorn's analysis indicated that additional IBNR reserves were warranted, and these reserves were recognized and reported by QC&S during TY 2022. Under QC&S's interpretation of Article 5.2(c), this recognition event is the trigger for allocating the IBNR to TY 2022.
QC&S's interpretation is supported by several considerations:
First, the plain language of Article 5.2(c) — "as recognized and reported by the Cedent during such Treaty Year" — expressly ties IBNR allocation to the act of recognition and reporting during a specific treaty year.
Second, the structural consistency of Article 5.2 supports a reporting-period allocation: paid losses are allocated to the treaty year of payment, case reserves to the treaty year of establishment, and IBNR to the treaty year of recognition. Each component is allocated based on when it appears on the cedent's books during the treaty year, not based on the accident year of the underlying claims.
Third, a recognition-based approach is the most practical and administrable methodology for an aggregate excess-of-loss treaty. IBNR reserves, by their very nature, relate to claims that have not yet been reported. Until the actuarial analysis indicates the need for reserves, the losses cannot be identified at the individual-claim level, making accident-year attribution impractical or speculative. The recognition-based approach avoids the need for retroactive reallocation of IBNR as additional claim information emerges.
Fourth, QC&S's interpretation is consistent with the economic structure of the Treaty. The retention of $120,000,000 per treaty year is designed to capture the cedent's expected loss level for each period. If IBNR development is allocated to the treaty year in which it is recognized, the aggregate net incurred loss for each treaty year reflects the cedent's actual financial experience during that period, including the impact of emerging development. This alignment between financial experience and cession calculations is consistent with the commercial purpose of the Treaty.
D. Pinnacle Atlantic's Interpretation
Pinnacle Atlantic reads Article 5.2 as requiring allocation of IBNR to the accident year of the underlying claims to which the reserves relate. Under this "accident-year" allocation, the $22,300,000 relates to accident-year 2021 exposures — losses arising from insured events that occurred during the period July 1, 2021 through June 30, 2022 — and must be attributed to TY 2021 regardless of when the IBNR development was recognized on QC&S's books.
Pinnacle Atlantic's theory, as articulated in the February 14, 2026 demand letter, rests on the following propositions:
First, Pinnacle Atlantic contends that "losses incurred" is an industry term of art that refers to losses arising from insured events that have actually occurred. The word "incurred," in Pinnacle Atlantic's view, refers to the occurrence of the underlying loss — not the recognition of a reserve for that loss. Under this reading, a loss is "incurred" when the insured event happens, not when the cedent's actuary determines that a reserve is warranted.
Second, Pinnacle Atlantic argues that IBNR reserve development merely reflects the maturation of already-existing losses that were "incurred" in a prior treaty year. The fact that the losses were not reported or reserved until a later period does not change when they were "incurred." The reserve strengthening during TY 2022 was simply the belated recognition of losses that were already incurred during TY 2021, and those losses should be attributed to TY 2021 for purposes of the Treaty.
Third, Pinnacle Atlantic contends that Velthorn's own actuarial workpapers likely organize the IBNR analysis by accident year, using loss development triangles that attribute emerging development to the accident year in which the underlying claims arose. If Velthorn's internal methodology attributes the $22,300,000 to accident-year 2021 claims, Pinnacle Atlantic argues, the Treaty requires the same attribution for cession calculation purposes.
Fourth, Pinnacle Atlantic argues that the phrase "as recognized and reported by the Cedent during such Treaty Year" in Article 5.2(c) is a procedural reporting requirement — it specifies when the cedent must report its IBNR estimates to the reinsurer, not how the IBNR should be allocated among treaty years. The allocation question, Pinnacle Atlantic contends, is governed by the concept of "losses incurred," which inherently ties to accident-year attribution.
Pinnacle Atlantic's interpretation, while not without textual support, faces significant challenges. As noted above, the structural consistency of Article 5.2 — allocating all three components (paid losses, case reserves, and IBNR) on a reporting-period basis — undermines the argument that IBNR alone should be subject to accident-year allocation. Moreover, Pinnacle Atlantic's interpretation would require retroactive reallocation of IBNR across treaty years whenever emerging development data allows for more precise accident-year attribution, creating administrative complexity and uncertainty that is inconsistent with the practical operation of a treaty that uses rolling twelve-month reporting periods.
IV. Velthorn Actuarial Advisors' Methodology
A. Overview of Velthorn's Analysis
Dr. Stephen Tanaka, FCAS, MAAA, is the Principal Actuary at Velthorn Actuarial Advisors LLC who performed the loss development and reserve analysis underlying the TY 2022 cession. Dr. Tanaka holds the Fellow of the Casualty Actuarial Society (FCAS) designation and is a Member of the American Academy of Actuaries (MAAA). He has over twenty years of experience in property and casualty actuarial consulting, with particular expertise in reinsurance reserving, loss development analysis, and actuarial opinions for regulatory and contractual purposes. Dr. Tanaka has served as the appointed actuary for QC&S's reinsurance reserve analyses since the inception of the Treaty in 2019.
For each treaty year, Velthorn prepares an actuarial analysis of QC&S's aggregate net incurred losses across the three covered lines of business. This analysis includes a review of paid loss data, case reserve data, and IBNR reserve estimates. Based on this analysis, Velthorn prepares actuarial certificates that are attached to the billing statements submitted by QC&S to Pinnacle Atlantic through Broadreach. The actuarial certificates represent Dr. Tanaka's professional opinion that the aggregate net incurred loss figures are reasonable and have been determined in accordance with generally accepted actuarial principles and applicable actuarial standards of practice, as required by Article 5.2(c) of the Treaty.
Velthorn's analysis employs a blend of three recognized actuarial methodologies for estimating ultimate losses and IBNR:
1.  Paid Loss Development Method: This method projects ultimate losses by applying development factors to paid loss data organized by accident year. The development factors are derived from QC&S's historical loss triangles, which track the emergence of paid losses over successive reporting periods. This method is most reliable for mature accident years with stable payment patterns but may be less reliable for recent accident years with limited payment history.
2.  Incurred Loss Development Method: This method is similar to the paid loss development method but uses incurred loss data (paid losses plus case reserves) as the starting point. Incurred losses typically exhibit less volatility than paid losses in the early development periods because case reserves provide an early estimate of ultimate loss values. Velthorn applies development factors to the incurred loss data to project ultimate losses by accident year.
3.  Bornhuetter-Ferguson Method: This method is a credibility-weighted blend of the development method and an a priori expected loss estimate. It is particularly useful for recent accident years where the loss development data is immature and the development method alone may produce volatile or unreliable estimates. The Bornhuetter-Ferguson method tempers the development method projections with a prior expectation of loss levels, providing a more stable estimate in the early stages of development.
Velthorn applies these three methods to each covered line of business separately and then selects a point estimate of ultimate losses and IBNR for each accident year based on professional judgment, considering the relative reliability of each method given the maturity of the data and the characteristics of the line of business.
B. IBNR Development Allocation Approach
The IBNR allocation methodology employed by Velthorn is central to this dispute. Velthorn allocated IBNR development to treaty years based on the reporting period in which development factors indicated that reserve changes were appropriate — consistent with QC&S's interpretation of Article 5.2(c). Under Velthorn's approach, when the actuarial analysis performed during a given treaty year indicates that IBNR reserves should be increased (or decreased), the resulting change in IBNR is allocated to the treaty year in which the analysis is performed and the reserve change is recognized on QC&S's books.
The $22,300,000 in disputed IBNR represents reserve strengthening that was recognized during the period July 2022 through June 2023 (TY 2022). This reserve strengthening was driven by emerging claim frequency and severity data received by QC&S and analyzed by Velthorn during that period. Specifically, as QC&S's claims department received new claim reports and updated information on existing claims during TY 2022, Velthorn incorporated this data into its actuarial models and determined that the previously estimated IBNR was insufficient. The resulting $22,300,000 increase in IBNR was reflected in the actuarial certificates supporting the TY 2022 billing statements.
Velthorn's position, as communicated by Dr. Tanaka in his written actuarial analyses and supporting documentation, is that while some of the underlying insured events contributing to the $22,300,000 in IBNR development may have occurred during accident-year 2021 (i.e., events between July 1, 2021 and June 30, 2022), the loss development patterns did not support IBNR recognition until the actuarial evaluations performed during TY 2022. In other words, the data available as of June 30, 2022 (the end of TY 2021) did not indicate the need for the additional $22,300,000 in reserves; it was only as additional loss data emerged during TY 2022 that the actuarial analysis supported the reserve increase.
This temporal distinction is critical. In long-tail casualty lines such as commercial general liability, excess/umbrella liability, and professional liability, it is common for significant loss development to occur well after the accident year. Claims may not be reported for months or years after the insured event, and even reported claims may undergo substantial reserve changes as litigation or investigation progresses. The loss development factors used by Velthorn are based on QC&S's historical loss triangles for each covered line and reflect the typical pattern of loss emergence over time. These development factors are not static; they are updated periodically as additional data becomes available. The $22,300,000 in IBNR development reflects the application of these updated development factors to the loss data available during TY 2022.
C. Consistency with Prior Practice
A significant element of QC&S's defense is the consistency of Velthorn's allocation methodology across treaty years. If the same recognition-based allocation methodology was used in TY 2019, TY 2020, and TY 2021 without objection from Pinnacle Atlantic, this establishes a course of dealing that supports QC&S's interpretation of Article 5.2 and undercuts Pinnacle Atlantic's claim that the methodology is improper.
We need to verify with Velthorn and with QC&S's internal records that the same approach was consistently applied. Based on preliminary conversations with Diana Kessler-Hoff, the recognition-based approach has been used since the inception of the Treaty. In each prior treaty year, Velthorn allocated IBNR to the treaty year in which the reserve development was recognized, and the resulting aggregate net incurred loss figures were reported to Pinnacle Atlantic through Broadreach in the same manner as the TY 2022 figures. Pinnacle Atlantic paid the TY 2020 cessions (TY 2020 was the first treaty year in which aggregate net incurred losses exceeded the $120,000,000 retention, with a cession of approximately $14,200,000) without raising any concerns about the IBNR allocation methodology. Similarly, no IBNR allocation issues were raised during TY 2021, a year in which aggregate losses fell below the retention.
If this consistency can be confirmed through documentary evidence — specifically, through comparison of Velthorn's actuarial certificates and supporting workpapers across treaty years — it will provide a strong foundation for a course-of-dealing argument. The argument would be that Pinnacle Atlantic acquiesced in the recognition-based allocation methodology for several years, received and reviewed the actuarial certificates that reflected this methodology, paid the cessions without objection, and cannot now retroactively challenge the methodology for TY 2022.
However, we should also consider whether Pinnacle Atlantic might argue that the IBNR allocation methodology was not transparent in the actuarial certificates — that the certificates simply reported aggregate IBNR figures without disclosing the allocation methodology in sufficient detail for Pinnacle Atlantic to identify and object to the approach. This would require review of the certificates themselves to assess how much detail was provided about the allocation methodology.
D. Potential Vulnerabilities in Velthorn's Approach
While Velthorn's recognition-based allocation methodology is a legitimate actuarial approach, several potential vulnerabilities should be noted and investigated:
Traceability of IBNR to accident-year 2021 claims. If any of the $22,300,000 in IBNR development can be traced to specific underlying claims that are clearly accident-year 2021 occurrences — for example, if Velthorn's actuarial workpapers organize the IBNR by accident year and show that a substantial portion of the $22,300,000 relates to accident-year 2021 exposures — this could undermine the recognition-based allocation theory. The distinction between "the development factors crystallized during TY 2022" and "the underlying losses occurred during accident-year 2021" may be difficult to maintain if Velthorn's own internal analysis attributes the IBNR to accident-year 2021 claims. We need to review Velthorn's complete workpapers, particularly the accident-year triangles and development factor selections, to assess this vulnerability. If the workpapers show that the $22,300,000 is principally attributable to development on accident-year 2022 claims (i.e., claims arising from events occurring after July 1, 2022), this vulnerability is significantly diminished. If the workpapers show a mix of accident-year 2021 and accident-year 2022 claims, the analysis becomes more nuanced.
Timing concerns and proximity to retention threshold. Pinnacle Atlantic may argue that the timing of the IBNR recognition during TY 2022 — a year in which aggregate losses were approaching and ultimately exceeded the $120,000,000 retention — is suspicious and suggests that QC&S or Velthorn engaged in strategic reserve manipulation. This argument would be that QC&S had an economic incentive to allocate IBNR to TY 2022 (where it would contribute to a cession above the retention) rather than to TY 2021 (where aggregate losses were well below the retention and the IBNR would be absorbed entirely by QC&S). While this is a serious allegation, it would require Pinnacle Atlantic to present evidence of bad faith or manipulation, which is a heavy burden. Dr. Tanaka's credentialed professional opinion carries significant weight, and absent evidence that the actuarial analysis was influenced by non-actuarial considerations, the timing argument is speculative. Nevertheless, we should investigate whether there were any communications between QC&S's management and Velthorn regarding the allocation of IBNR to specific treaty years, any discussions about the impact of the allocation on cession calculations, or any changes in reserving philosophy that coincided with the TY 2022 reporting.
Claims department reclassifications. We should also assess whether any of the $22,300,000 in IBNR development was influenced by claims department reclassifications or changes in coding practices during TY 2022. If claims that were originally coded to accident-year 2022 were reclassified to accident-year 2021 (or vice versa) during the period, this could complicate the allocation analysis and provide fodder for Pinnacle Atlantic's argument.
Completeness of workpapers. The importance of obtaining Velthorn's complete workpapers and data exhibits cannot be overstated. The actuarial certificates provided to Pinnacle Atlantic are summary documents that do not contain the level of detail necessary to evaluate the IBNR allocation. The workpapers — including loss development triangles, development factor selections, expected loss ratio assumptions, claim count data, and any memoranda documenting Velthorn's professional judgment — are essential to understanding the actuarial basis for the allocation and to assessing whether the allocation can withstand scrutiny from an opposing actuarial expert.
V. Actuarial Standards of Practice
A. Applicable Actuarial Standards
The actuarial analysis performed by Velthorn is governed by several Actuarial Standards of Practice ("ASOPs") promulgated by the Actuarial Standards Board (the "ASB"), which is the body responsible for establishing binding standards for actuaries in the United States. Article 5.2(c) of the Treaty expressly requires that IBNR reserves be determined "in accordance with generally accepted actuarial principles and applicable actuarial standards of practice." The following ASOPs are directly relevant to the IBNR allocation analysis:
Actuarial Standard of Practice No. 43 — "Unpaid Claim Estimates." ASOP No. 43 sets forth the standards for actuaries when estimating unpaid claims, including IBNR. It applies to actuaries who issue statements of actuarial opinion on unpaid claim estimates for property and casualty insurance and reinsurance. ASOP No. 43 addresses the selection of methods, the use of data, the evaluation of assumptions, and the communication of results. Importantly, ASOP No. 43 recognizes that the actuary may use multiple methods and should exercise professional judgment in selecting a point estimate of unpaid claims. The standard does not prescribe a single allocation methodology for attributing IBNR to accident years or reporting periods; rather, it provides a framework within which the actuary exercises judgment based on the characteristics of the data and the purpose of the analysis.
Actuarial Standard of Practice No. 36 — "Statements of Actuarial Opinion Regarding Property/Casualty Loss and Loss Adjustment Expense Reserves." ASOP No. 36 is relevant to the actuarial opinions that Velthorn provided in connection with QC&S's loss reserve certifications. This standard addresses the actuary's responsibilities when preparing a statement of actuarial opinion regarding loss reserves, including the scope of the opinion, the data reviewed, the methods used, and the disclosures required. ASOP No. 36 requires the actuary to disclose any limitations on the scope of the analysis and any significant assumptions that materially affect the results.
Actuarial Standard of Practice No. 23 — "Data Quality." ASOP No. 23 sets forth standards for the data relied upon by the actuary in performing actuarial analyses. It requires the actuary to evaluate the data for reasonableness, completeness, and consistency, and to disclose any limitations in the data that may affect the results. In the context of the IBNR allocation dispute, ASOP No. 23 is relevant to the question of whether the data used by Velthorn to allocate IBNR to treaty years was sufficient and appropriate for that purpose. If the data does not support a reliable attribution of IBNR to specific accident years, this supports the use of a recognition-based allocation that relies on aggregate development factors rather than accident-year-level data.
B. Treatment of IBNR Allocation in Actuarial Practice
A fundamental point in the actuarial analysis is that the applicable ASOPs do not mandate a single method for allocating IBNR to specific accident years or reporting periods. The selection of methodology is within the actuary's professional judgment, subject to the constraints of reasonableness, consistency, and disclosure. Both accident-year allocation and recognition-based allocation methods are recognized in actuarial practice, and neither is inherently superior to the other in all circumstances.
In general actuarial practice, IBNR is most commonly estimated at the accident-year level — that is, the actuary constructs development triangles organized by accident year and estimates ultimate losses and IBNR for each accident year separately. This is the standard approach to loss reserving for purposes of financial reporting, regulatory filings, and internal management of the insurance enterprise. However, this accident-year organization of the actuarial analysis does not necessarily dictate how the resulting IBNR estimates should be allocated among contractual reporting periods (such as treaty years in a reinsurance treaty) for purposes of calculating ceded losses. The allocation to contractual periods is a distinct question that depends on the terms of the contract.
In the context of reinsurance treaties, the allocation method is typically governed by the treaty language, and the actuary's role is to apply the method consistent with the contractual framework. Where the treaty language is ambiguous or silent on the allocation method, the actuary may exercise professional judgment in selecting a methodology, considering the purpose of the analysis (cession calculation under an aggregate excess-of-loss treaty), the characteristics of the data, and any applicable industry conventions.
Both the recognition-based approach (allocating IBNR to the treaty year in which the reserve development is recognized) and the accident-year approach (allocating IBNR to the treaty year corresponding to the accident year of the underlying claims) are legitimate methodologies that find support in actuarial practice. The recognition-based approach is more commonly used in reinsurance contexts where the treaty language defines "losses incurred" by reference to the cedent's reported loss position, as it avoids the need for retroactive reallocation of IBNR among treaty years as development data matures. The accident-year approach is more commonly used in regulatory and financial reporting contexts, where the goal is to attribute losses to the periods in which the underlying risk was assumed.
C. Implications for QC&S's Defense
The actuarial standards landscape has several important implications for QC&S's defense:
If Article 5.2 is determined to be ambiguous, actuarial custom and practice may inform the interpretation. The fact that Velthorn's recognition-based approach is a legitimate actuarial methodology — not a novel or idiosyncratic method — supports QC&S's position. An arbitration panel or court considering extrinsic evidence of actuarial practice would likely find that both allocation methods are reasonable, which may lead to a finding that Article 5.2 is genuinely ambiguous and must be resolved based on other interpretive tools (contra proferentem, course of dealing, etc.).
However, if Pinnacle Atlantic retains its own actuarial expert who opines that accident-year allocation is the industry-standard method for aggregate excess-of-loss treaties and that recognition-based allocation is disfavored or unusual in this context, this could create a battle of experts. An opposing expert might argue that the recognition-based approach, while technically permissible under the ASOPs, is not the approach contemplated by the Treaty or customarily used in the aggregate excess-of-loss treaty market. The credibility and qualifications of the competing experts would be critical, and QC&S should be prepared to retain an independent actuarial expert (separate from Velthorn, which is a fact witness) to provide an opinion supporting the reasonableness of the recognition-based approach.
It is also worth noting that the ASOPs provide procedural protections for the actuary's professional judgment. ASOP No. 43 requires the actuary to document the rationale for the selected methodology and to disclose any significant assumptions. If Dr. Tanaka's workpapers and documentation demonstrate that he considered the accident-year allocation approach and concluded that the recognition-based approach was more appropriate given the Treaty language and the characteristics of the data, this would be a significant point in QC&S's favor.
VI. New York Reinsurance Case Law on IBNR Allocation
A. General Principles of Treaty Interpretation Under New York Law
New York courts interpret reinsurance treaties as contracts, applying the standard principles of contract construction that govern commercial agreements generally. These principles are well-established and may be summarized as follows:
The primary objective of contract interpretation is to give effect to the parties' intent as expressed in the language of the agreement. Where the contract language is clear and unambiguous, the court will enforce the provision according to its plain meaning, without resort to extrinsic evidence. See W.W.W. Assocs., Inc. v. Giancontieri, 77 N.Y.2d 157, 162 (1990) ("A familiar and eminently sensible proposition of law is that, when parties set down their agreement in a clear, complete document, their writing should as a rule be enforced according to its terms."). The question of whether a contractual provision is ambiguous is a question of law for the court to determine. A provision is ambiguous if it is reasonably susceptible of more than one interpretation. Id. at 162–63.
When a reinsurance treaty provision is found to be ambiguous, New York courts may consider extrinsic evidence to resolve the ambiguity, including the parties' course of dealing, industry custom and usage, the circumstances surrounding the formation of the contract, and the reasonable expectations of the parties. See Greenfield v. Philles Records, Inc., 98 N.Y.2d 562, 569 (2002). The court may also apply the doctrine of contra proferentem, construing ambiguous provisions against the party that drafted them. See Ace Fire Underwriters Ins. Co. v. ITT Indus., Inc., 15 A.D.3d 263, 264 (1st Dep't 2005). This doctrine has particular application in reinsurance disputes where one party prepared the treaty wording.
In the reinsurance context, New York courts have recognized that reinsurance treaties are specialized commercial contracts that should be interpreted in light of the customs and practices of the reinsurance industry. See Unigard Sec. Ins. Co. v. Greaves Specialty Insurance, 79 N.Y.2d 576, 581 (1992). Courts are generally reluctant to construe reinsurance treaties in a manner that produces commercially unreasonable results or that is inconsistent with the economic structure of the transaction.
The "losses incurred" definition is a critical term in aggregate excess-of-loss treaties because it directly determines the amount of ceded losses and, consequently, the reinsurer's financial obligation. As such, it has been the subject of significant reinsurance litigation, both in court and in arbitration.
B. Relevant New York Precedent on IBNR Allocation in Reinsurance
New York case law addressing the specific question of IBNR allocation between treaty years in aggregate excess-of-loss reinsurance treaties is limited, reflecting the fact that many such disputes are resolved through arbitration under ARIAS-U.S. or similar rules. However, several categories of precedent provide useful guidance:
Cedent deference in loss allocation. New York courts have generally afforded deference to the cedent's reasonable loss allocation methodology in reinsurance disputes. The principle underlying this deference is that the cedent, as the party with primary responsibility for administering the underlying insurance claims, is in the best position to determine how losses should be allocated and reported. Courts have held that a cedent's allocation methodology will be upheld unless it is shown to be unreasonable, arbitrary, or made in bad faith. See, e.g., Greaves Specialty Insurance v. ACE Am. Reinsurance Co., 361 F.3d 134, 140 (2d Cir. 2004) (applying New York law and affirming the principle that a reinsurer is generally bound by the cedent's reasonable claims-handling decisions). This principle of cedent deference is relevant to the IBNR allocation dispute because it places the burden on Pinnacle Atlantic to demonstrate that QC&S's allocation methodology is unreasonable, rather than requiring QC&S to prove that its methodology is the only correct approach.
IBNR allocation and the "losses incurred" definition. Courts and arbitration panels have addressed the treatment of IBNR in the "losses incurred" calculation under various reinsurance treaty formulations. In Employers Reinsurance Corp. v. Mid-Continent Cas. Co., No. 07 Civ. 6009 (S.D.N.Y. 2009), the court considered a dispute over the inclusion of IBNR reserves in the calculation of losses under an aggregate excess-of-loss treaty. While the specific allocation issue was not identical to the one presented here, the court emphasized the importance of the treaty language in defining "losses incurred" and held that the cedent's approach to including IBNR in the loss calculation was proper where it was consistent with the treaty's definition. Similarly, in various ARIAS-U.S. arbitrations that have been reported or discussed in secondary sources, arbitration panels have generally upheld the cedent's IBNR methodology where it was consistent with the treaty language, was applied consistently across reporting periods, and was supported by a credentialed actuary's opinion.
Follow-the-fortunes and IBNR. The "follow the fortunes" doctrine — which requires a reinsurer to follow the cedent's reasonable claims-handling and underwriting decisions — has been applied in some cases to IBNR allocation decisions. Under this doctrine, a reinsurer is bound by the cedent's good-faith determination of losses, including the allocation of IBNR, absent evidence of bad faith, fraud, or manifest error. See Vireo Life Insurance v. Certain Underwriters at Sundial Indemnity Co., 96 N.Y.2d 583, 596 (2001) (discussing the follow-the-fortunes doctrine in the reinsurance context). The application of this doctrine to the IBNR allocation issue is analyzed in greater detail in the companion memorandum on Article 11 (follow the fortunes / follow the settlements). For present purposes, it is sufficient to note that the follow-the-fortunes doctrine may provide an additional layer of protection for QC&S's IBNR allocation, even if Article 5.2 is found to be ambiguous, by requiring Pinnacle Atlantic to accept the cedent's reasonable allocation absent bad faith.
Arbitration forum considerations. Given that the Treaty contains an arbitration clause providing for resolution under ARIAS-U.S. Rules, it is important to recognize that the ultimate forum for this dispute is likely to be a reinsurance arbitration panel rather than a court. Arbitration panels in reinsurance disputes are composed of experienced industry professionals who are familiar with actuarial practices, treaty interpretation conventions, and the commercial realities of the reinsurance market. These panels tend to give significant weight to industry custom and actuarial practice in resolving treaty interpretation disputes. An arbitration panel is likely to be receptive to testimony from actuarial experts and to evaluate the reasonableness of the cedent's allocation methodology in light of industry norms, rather than engaging in a purely textual analysis of the treaty language. This forum consideration favors QC&S, as its recognition-based allocation methodology is supported by actuarial practice and was applied consistently without prior objection.
C. Application to the Treaty Dispute
Applying the foregoing legal principles to the facts of this dispute, we conclude that QC&S's allocation of the $22,300,000 in IBNR to TY 2022 based on Velthorn's recognition-based methodology is defensible under New York law, provided that three conditions are satisfied:
(i) The methodology is consistent with Article 5.2. As discussed in Section III, QC&S's interpretation of Article 5.2(c) — that IBNR is allocated to the treaty year in which it is "recognized and reported by the Cedent" — is a reasonable reading of the treaty language and may well be the best reading when considered in the context of Article 5.2 as a whole.
(ii) The methodology was applied consistently across treaty years. If the same recognition-based approach was used for TY 2019, TY 2020, and TY 2021 without modification or objection, this establishes a course of dealing that reinforces QC&S's interpretation and undercuts Pinnacle Atlantic's belated challenge.
(iii) The allocation was not made in bad faith or to manipulate cession calculations. If the IBNR allocation was driven by legitimate actuarial analysis and not by a desire to shift losses from TY 2021 (below retention) to TY 2022 (above retention) for the purpose of generating a cession, the allocation will be upheld under the cedent deference standard. Dr. Tanaka's credentialed professional opinion is strong evidence of good faith.
Pinnacle Atlantic bears the burden of demonstrating that the allocation was improper. That burden is heightened by two significant factors. First, Pinnacle Atlantic paid the full $78,400,000 across seven billing installments between October 15, 2022 and March 29, 2024, without reserving rights as to the IBNR allocation methodology. Each billing statement was accompanied by Velthorn's actuarial certificate, which reflected the recognition-based allocation methodology. Pinnacle Atlantic's failure to object at the time of payment — or at any point during the approximately two to three years between the initial billings and the December 12, 2025 reservation-of-rights letter — weakens its position and may give rise to waiver or estoppel arguments, as discussed in Section VII. Second, Pinnacle Atlantic's own demand letter does not address the logical consequences of its reallocation theory — specifically, the $6,900,000 cession that would attach for TY 2021 — which undermines the completeness and credibility of its position.
VII. Course of Dealing and Waiver / Estoppel Considerations
A. Pinnacle Atlantic's Payment History
Pinnacle Atlantic paid the full $78,400,000 in ceded losses across seven billing installments between October 15, 2022 and March 29, 2024. Each billing statement submitted by QC&S through Broadreach was accompanied by Velthorn's actuarial certificate, which set forth the aggregate net incurred loss figures and the IBNR component. Based on the information currently available, Pinnacle Atlantic paid each installment without reserving rights as to the IBNR allocation methodology. Pinnacle Atlantic did not request additional information about the allocation methodology, did not object to the inclusion of the $22,300,000 in IBNR in the TY 2022 aggregate net incurred loss, and did not indicate any disagreement with the actuarial certificates.
The first formal expression of concern regarding the IBNR allocation came in the December 12, 2025 reservation-of-rights letter from Hargrove, Fitch & Calloway LLP — more than twenty months after the final payment on March 29, 2024, and more than three years after the initial billing in October 2022. The formal demand for repayment followed on February 14, 2026.
This extended period of silence and unqualified payment raises significant waiver and estoppel issues under New York law.
Waiver. Under New York law, waiver is the vo

MEMORANDUM

Record

Parties

Dates

Amounts