A cell captive is a captive insurance structure in which one legal entity, the protected cell company (PCC) or sponsored captive, hosts multiple independent cells, each ring-fenced from the others by statute. The PCC owns the legal license; each cell operates under the license and writes its own book of business with its own assets, its own liabilities, and its own reserves. Insolvency in one cell, in principle, does not reach the assets of another cell or of the PCC’s core.
The structure has grown from a niche Bermuda and Guernsey concept in the 1990s to a mainstream alternative across most major U.S. captive domiciles and offshore jurisdictions. Florida, Vermont, Tennessee, the District of Columbia, Bermuda, Guernsey, the Cayman Islands, Malta, and Mauritius all have established PCC legislation and an active cell market. The structural variants matter, the ring-fencing has limits worth knowing, and the reserving problem inside a cell is the same as the reserving problem in a single-parent captive with thinner data and structural assumptions worth scrutinizing.
This article explains what a cell captive actually is, how the ring-fencing affects reserving and capital sizing, what the cross-cell contamination risk looks like, and what the buyer should require from the actuary’s report on a cell’s reserve. The starting points are self-insurance and captive structures for the broader risk-financing landscape and single-parent captive reserving for the standard reserving framework that cells inherit.
What a cell captive actually is
Three structural variants matter. The terminology varies by jurisdiction; the underlying mechanics are similar.
Protected cell company (PCC). A single legal entity authorized by statute to create multiple cells. Each cell has its own assets, its own liabilities, and its own statutory ring-fencing protecting the cell’s assets from claims against other cells or against the PCC’s core. The cells do not have separate legal personality; the PCC is the legal entity. Bermuda, Guernsey, and most U.S. PCC statutes use this model.
Incorporated cell company (ICC). A variant in which each cell is a separately incorporated entity, owned by the ICC. The cell’s separate legal personality strengthens the ring-fencing against legal challenge, at the cost of additional incorporation expense per cell. Guernsey, Jersey, and Malta authorize ICCs alongside their PCC legislation.
Sponsored captive. Used in some U.S. domiciles (Vermont, South Carolina, and others), the sponsored captive is functionally similar to a PCC: the sponsoring organization holds the license, and participants subscribe to cells that operate under it. Statutory ring-fencing language varies; the underwriting and reserving mechanics resemble the PCC model.
For reserving purposes, the structural label matters less than four operational features the actuary should confirm. Whether the cell has its own balance sheet within the host entity’s accounts. Whether the cell’s liabilities are statutorily ring-fenced from the other cells and from the host’s core. Whether the cell shares any administrative or service-provider arrangements with the host (most do; this is the cost advantage that drives cell formation). Whether the cell is contractually required to participate in cross-cell guarantees, support, or contingent assessments (most are not; some are).
The four features together describe the cell’s reserving posture more precisely than the host jurisdiction’s statute title. The actuary should disclose all four in the report’s structural description.
Why companies use cells
Three operational motivations dominate.
Lower capital threshold. A single-parent captive in a U.S. domicile typically requires minimum capital of $250,000 to $1 million, depending on jurisdiction and lines written. A cell in a PCC can operate at a much lower capital threshold, sometimes as low as $50,000 or $100,000 per cell, because the host PCC carries the core capital and regulatory infrastructure. The lower threshold opens captive participation to mid-market organizations that cannot economically support a single-parent structure.
Speed to market. Forming a single-parent captive is a multi-month licensing process. Subscribing to a cell in an existing PCC can be completed in weeks once the underwriting package and feasibility analysis are in hand. For a parent organization responding to a market hardening, a sudden coverage gap, or a one-time risk-financing decision, the speed advantage can be material.
Administrative cost sharing. The PCC’s captive manager, auditor, actuary, and domicile counsel serve all cells under the license. Per-cell fees are typically a fraction of what a single-parent captive pays, because the fixed costs are amortized across cells. The cost savings are largest for cells with relatively small premium volume, where standalone administration would be uneconomic.
Cells are not a free option. They share infrastructure with other cells whose risk profiles and management quality the cell owner does not control. Cells are also subject to the host domicile’s regulatory framework, which may differ from the domicile the parent would have chosen for a single-parent captive. The decision to cell rather than to incorporate is a trade-off, and the right answer depends on the specific risk profile, the expected premium volume, and the long-term strategy.
The feasibility analysis for a cell looks similar to the feasibility analysis for a single-parent captive, with two additions. The cell’s structural ring-fencing must be diligenced (more on this below). The host PCC’s overall financial health must be diligenced; even with ring-fencing, the cell’s day-to-day operation depends on the host’s continued solvency and operational continuity.
Ring-fencing and what it actually means
The legal doctrine of ring-fencing is what makes the cell structure work. Statutory ring-fencing provides that the assets attributable to one cell are available only to satisfy the liabilities of that cell, and the assets attributable to the PCC’s core are available only to satisfy liabilities of the core. A creditor of one cell cannot reach the assets of another cell or of the core; a creditor of the core cannot reach the assets of any cell.
In well-established PCC jurisdictions with mature case law, ring-fencing has held up under stress. Bermuda’s PCC framework, dating to 1997 and refined through subsequent amendments, has been tested in restructurings and has performed. Guernsey’s framework has a similar history. U.S. PCC statutes are newer and less tested in litigation; the most-litigated U.S. domiciles (Vermont, Hawaii, the District of Columbia) have built up some experience, but the body of cross-cell contamination case law is still thin compared to the offshore precedents.
For the actuary’s reserving analysis, ring-fencing has two practical implications.
The cell’s reserve liability is its own. The actuary estimating the cell’s reserve is not estimating the host PCC’s reserve. The cell’s loss data, exposure base, and reinsurance structure are the inputs; the cell’s own balance sheet is the output. The actuary should not aggregate cell reserves into the host’s reserves except for combined financial-statement purposes the auditor governs separately.
The cell’s capital adequacy is its own. A cell that is undercapitalized for its reserve liability is a problem the cell’s participant must address; the host PCC’s core capital does not stand behind the cell as a matter of statutory right. Cells operating near the capital threshold need the same confidence-level funding analysis that a single-parent captive needs, scaled to the cell’s risk profile.
What ring-fencing does not protect against: reputational contagion across cells, regulatory action that affects the entire PCC (a license revocation or domicile suspension would reach every cell), and the practical operational dependency cells have on the host’s service providers. A cell whose host PCC’s actuary, auditor, or captive manager fails is a cell whose continuity is at risk regardless of statutory ring-fencing.
How reserving differs for a cell
The reserving methodology for a cell is the methodology for any captive: expected claims method when the cell is new and has no own data; Bornhuetter-Ferguson as the cell accumulates one to three years of experience; chain ladder on mature accident years once the cell has the data to support it. Friedland’s framework for the five core methods applies without modification. The thin-data problem (Friedland, p. 89) and the leveraged-factor concern (Friedland, p. 134) apply with full force, often more so than at a single-parent captive, because the cell’s expected premium and claim volumes are typically smaller.
Four operational differences shape the reserving analysis.
Thinner data per cell. A typical cell writes one line of business at one parent organization with annual premium below what a single-parent captive would typically retain. The cell may have only a few dozen claims a year, sometimes fewer. Credibility is correspondingly lower. The actuary’s reliance on industry benchmarks and the parent’s prior loss experience is higher. Friedland’s benchmark caution (Friedland, p. 88) applies more strongly to cells than to larger captive structures.
Reinsurance dependency. Cells routinely retain only a working layer and cede everything else through reinsurance. The cell’s gross-to-net bridge is materially more important to the cell’s reserve than to a captive that retains larger layers. The frame in fronting and the gross-to-net bridge applies; the cell’s actuary should produce the same three-view reserve presentation the captive’s actuary produces.
Sponsor and program-wide reinsurance. Some PCC structures provide program-wide reinsurance or aggregate stop-loss at the host level, with attachments and limits applying across multiple cells. The cell’s actuary must understand how the host-level reinsurance interacts with the cell’s retained reserves, and whether the cell’s recoverable from the host-level program would actually pay before another cell’s claims exhausted the limit. This is one of the few places where cross-cell interaction directly affects the individual cell’s reserve.
Shared service-provider methodology. When the same actuary, captive manager, or auditor serves every cell under a PCC, the methodology applied to each cell tends to be similar. Consistency has value, but it can also obscure cell-specific reserving needs. A cell whose exposure profile differs materially from the rest of the PCC’s cell book may need a different reserving methodology, and the buyer should verify that the actuary tailored the analysis rather than applying the PCC’s template.
Capital sizing inside a cell
A cell’s capital requirement is the cell’s own. Domicile regulators set minimum capital thresholds, but the more economically relevant number is the capital the cell needs to fund its reserves at a chosen confidence level. The percentile-funding analysis Friedland describes for captive structures generally (Friedland, p. 10 on the actuarial central estimate) applies to cells with no modification, with one operational caution: the cell’s smaller loss volume produces a more variable aggregate distribution. Funding at the 90th percentile on a cell with twenty expected claims a year requires a larger multiple of expected losses than funding at the same percentile on a single-parent captive with two hundred expected claims a year.
The variance-of-the-mean intuition is direct. With smaller expected claim counts, the coefficient of variation of the aggregate distribution is larger. Larger coefficient of variation produces a fatter right tail at a given expected loss level, and the gap between the central estimate and the 90th-percentile funded amount grows. A cell that is capitalized at one-and-a-half times the central estimate may be funded to the 75th percentile rather than to the 90th, simply because the loss distribution is wider.
The buyer’s takeaway is straightforward. Cells need confidence-level funding analysis at least as rigorous as the analysis a single-parent captive needs, and the analysis should be tailored to the cell’s specific loss volume rather than scaled down from a larger captive’s template. A cell that adopts the host PCC’s funding template without revisiting the variance assumption may be carrying capital adequacy that looks acceptable on paper and is insufficient in practice.
Cross-cell contamination concerns
Statutory ring-fencing is the first defense against cross-cell contamination. It is not the only defense, and it is not absolute.
Three categories of cross-cell risk are worth understanding.
Regulatory action against the host. A domicile regulator suspending or revoking the PCC’s license affects every cell, regardless of ring-fencing. The cells lose access to the regulatory framework that authorized them, and the wind-down or transfer of the cells becomes a domicile-specific exercise. The diligence in the captive domicile survey on regulatory posture and stability applies at the PCC level for any cell participant.
Litigation challenging ring-fencing. A creditor of one cell whose recovery is limited by the cell’s assets may sue to reach assets of other cells or of the host’s core, arguing that ring-fencing should not apply because of fraudulent transfer, inadequate capitalization, or some other equitable theory. Mature PCC jurisdictions have case law that supports ring-fencing in most circumstances. Newer jurisdictions have less. The cell participant should understand the host domicile’s case law before subscribing, because contestable ring-fencing is not the same as enforceable ring-fencing.
Operational failure at the host. A PCC whose captive manager, actuary, or auditor fails affects every cell operationally even if the cells’ assets remain statutorily protected. The cells need a continuity plan that does not depend on the host’s specific service-provider lineup. For a small cell, the practical reality is that operational failure at the host can be more disruptive than direct liability exposure to another cell, because the cell’s own management depth is limited.
The Lloyd’s of London structural analogy is imperfect but useful. Lloyd’s syndicates are individually liable in much the way PCC cells are, with Lloyd’s central fund providing a layer of cross-syndicate support that PCCs typically do not provide. PCC participants generally bear their own losses without a cross-cell backstop; the cell’s continuity is its own to engineer. The captive cluster of which PCCs are part operates under tighter capital constraints and looser inter-participant support than the Lloyd’s model, and the cell participant should plan accordingly.
Tax and regulatory treatment
The U.S. tax treatment of cells is a meaningful diligence item. The IRS’s position, developed through revenue rulings and procedural guidance over more than two decades, is that a properly structured cell can be treated as a separate insurance company for federal tax purposes, qualifying for §831(b) micro-captive election if eligible and for §832 deduction treatment for the parent’s premium payments. Each cell is evaluated on its own merits; a cell that fails the underwriting, risk-shifting, or risk-distribution tests does not qualify, regardless of the PCC’s overall posture.
The §832 captive deduction analysis applies at the cell level, not at the host PCC level. The cell’s premium volume, risk distribution across insureds, and underwriting independence each matter. A cell whose underwriting is dictated by the parent organization or whose risk distribution is insufficient may fail tax recognition while a sister cell in the same PCC passes.
State and domicile regulatory treatment varies. Most U.S. PCC domiciles treat each cell as a separate subject of regulatory examination, with cell-level financial statements, cell-level actuarial opinions, and cell-level supervisory attention. Offshore domiciles’ treatment ranges from cell-by-cell examination to PCC-aggregated examination depending on jurisdiction. The cell’s actuary should know which regime applies and structure the actuarial deliverables accordingly. A cell required to file a standalone Statement of Actuarial Opinion is a different deliverable from a cell whose actuarial work is folded into a PCC-aggregate filing.
Cell exit and dissolution
A cell that no longer serves its participant’s purpose can be exited through three routes. Each has actuarial implications.
Dissolution. The cell stops writing new business, pays out its existing reserves over the runoff horizon, and is formally dissolved by the host PCC once the last claim closes. For a long-tail line, the runoff horizon can be ten to twenty years, and the cell continues to require actuarial opinions and capital adequacy until dissolution.
Loss portfolio transfer. The cell transfers its existing reserves to a third-party reinsurer through an LPT, releasing the cell’s capital and closing the cell on the host’s books. The mechanics are the same as the LPT mechanics for a single-parent captive, with the additional wrinkle that the host PCC’s approval and the domicile regulator’s approval are both required.
Migration to a single-parent structure. A cell that has grown beyond its original economic justification may migrate to a single-parent captive in the same or a different domicile, with the existing reserves novated to the new entity. This is less common than LPT or dissolution, because the migration involves more legal and regulatory friction than either alternative, but it is the right answer when the cell’s expected long-term volume justifies a standalone structure.
For each route, the cell’s actuary produces a reserve estimate that the buyer, the host PCC, and the assuming party (if any) negotiate around. The frame in LPT and ADC transactions applies in miniature; the cell’s smaller scale does not reduce the diligence required, and the cell participant should bring the same rigor to a cell exit that a single-parent captive owner would bring to a single-parent exit.
What to require from the actuary’s report on a cell
The cell-specific items extend the standard captive reserve study checklist. Each item below is reasonable to expect.
A description of the cell’s structural posture: which PCC or sponsored captive hosts the cell, the cell’s ring-fencing status under the host domicile’s statute, the cell’s reinsurance program (cell-specific and any host-level), and the cell’s service-provider arrangement with the host.
A separately presented reserve estimate for the cell, not aggregated with the PCC’s other cells. The estimate should be gross, ceded, and net, in the three-view structure fronting and the gross-to-net bridge describes.
A confidence-level funding analysis specific to the cell, with the variance assumption tailored to the cell’s loss volume rather than copied from a PCC template. The confidence-level funding piece frames the choices.
A reconciliation against the cell’s previous-period reserve, with explanations for material changes that are not driven by paid claims or loss emergence in the period.
For cells participating in host-level reinsurance or aggregate stop-loss, an explanation of how the host-level program would respond to the cell’s losses and whether the cell’s recoverable could be impaired by simultaneous claims at sister cells.
For cells operating in a domicile with mature ring-fencing case law, a citation to the relevant authority that confirms the cell’s protection. For cells operating in newer jurisdictions, an explicit acknowledgment of the case-law gap.
A signed actuarial opinion on the cell’s reserves under the standard that applies to the host domicile’s filing requirements.
A report that omits any of these items has a documentation gap, and the gap is the cell participant’s problem to fix. The diligence frames in reading the actuarial proposal and the RFP for reserve review apply at the cell scale as well as the single-parent scale, and the cell participant should bring the same rigor to a cell engagement that a single-parent captive owner would bring to a much larger engagement.
When a cell is the right answer
Cells are not a universal substitute for single-parent captives. A parent organization with sufficient premium volume to support a standalone captive often does better with the standalone structure, because the parent gains full control of the regulatory posture, the service-provider relationships, and the long-term strategic direction. The cell structure trades control for cost.
Cells make sense when premium volume is too small to economically support a single-parent captive, when speed to market matters more than long-term control, when the parent wants to test a captive structure before committing to a standalone, or when the parent participates in an industry consortium PCC whose member-shared infrastructure is part of the value proposition. In each case, the trade-offs are real and the analysis should be specific to the parent’s risk profile, premium volume, and strategic intent.
The decision is the parent’s, not the actuary’s. The actuary’s role is to size the reserves and the capital at whatever structural choice the parent makes. For a cell, that role is the same role the actuary plays for a single-parent captive, with thinner data, structural caveats, and a closer eye on cross-cell and host-level dependencies. The reserving methodology is the same. The judgment around the methodology requires a tighter focus on the cell’s specific posture.