RESEARCH DOSSIER
Insurance Sector Market Liquidity Support during Systemic Shocks
Abstract. Systemic liquidity shocks challenge the institutional capacity of insurers to convert balance-sheet resources into timely payments while preserving the operation of the markets in which they invest and hedge. This monograph examines funding shortages, public liquidity facilities and their terms through a comparative documentary investigation of the United States, the United Kingdom, Canada, European supervisory exercises and international standard setting. The analysis reconstructs American International Group's 2008 emergency assistance, the market-wide interventions of March 2020, the British gilt market disturbance of 2022, and contingent non-bank lending arrangements. It distinguishes legal-entity liquidity from consolidated resources, cash-flow adequacy from economic solvency, and liquidity provision from loss absorption. Official transaction descriptions, legislation, central bank notices, supervisory exercises, monitoring reports and bounded findings from financial economics provide the evidentiary basis. Explicit hypothetical cash-flow and collateral calculations test the consequences of timing, asset encumbrance, haircuts and incomplete access. The principal conclusion is that market stability interventions can interrupt forced-sale transmission and preserve insurance payment capacity when the instrument matches the source of distress and the recipient can remain viable. Their design requires credible private resilience, conditional access, enforceable collateral rights, fiscal accountability and a workable exit. The evidence supports institutional comparison and mechanism-based interpretation; it does not yield a universal ranking of facilities or a causal estimate of avoided insurance failures.
Keywords: Insurance liquidity; systemic shocks; funding liquidity; market liquidity; non-bank financial intermediation; lender of last resort; central bank facilities; collateral; margin calls; cash-flow stress testing; fire sales; insurance supervision; financial stability; AIG; gilt market; contingent repo; moral hazard; resolution; policyholder protection.
Introduction
Insurance transforms uncertain future losses into contractually organized financial obligations. Its economic contribution depends on the continuity of those obligations through time: a policy promising payment at a specified contingency remains socially effective only where the institution responsible for it has the capacity to deliver the payment when due. The same insurer commonly participates in markets for government and corporate debt, investment fund shares, derivatives, securities financing and reinsurance. A disturbance can therefore reach insurance from either direction. Claims and withdrawals may create the need to raise money from markets; market events may create demands for funds independently of insured losses. The resulting interactions make liquidity support a question about the institutional architecture of the financial system.
A systemic shock is a disturbance capable of materially disrupting interconnected financial functions across institutions or markets. Within this investigation, the relevant function is the reliable transfer of purchasing power at the times required by contracts and settlement arrangements. Market support denotes operations directed toward trading, financing or settlement capacity; firm support denotes an intervention addressed to an identified economic or legal entity. Both may affect the same insurer, but the allocation of credit risk, discretion and possible loss differs. An insurer can benefit indirectly from the restoration of corporate bond trading even when it never borrows from a central bank.
The study asks three questions. How are funding shortages generated and observed at insurer and sector level? Through which contractual and institutional channels do public facilities mitigate shortages and market dysfunction? Which conditions allow public support to reduce systemic transmission while maintaining private responsibility and preserving legitimate claims? These questions correspond to the independent research object defined within Global Insurance Markets and Institutions and Systemic and Emerging Risks. They are addressed through six substantive chapters, historical reconstruction, comparative mechanism analysis and explicitly bounded calculations.
The investigation prioritizes evidence about operations and legal terms over general descriptions of stability. It distinguishes prospective policy design, official reports of implemented operations, simulated outcomes and empirical scholarship. Public records provide relatively strong evidence about facility dates, eligibility and balances, while estimates of systemic losses avoided are more uncertain. This distinction is essential for explaining what an intervention accomplished and what remains a plausible counterfactual.
Conceptual and Historical Foundations
The temporal structure of an insurance balance sheet
The economic value of insurance assets and liabilities is measured across many future dates. A liquidity assessment concerns the availability of cash or acceptable noncash settlement resources at specific times, within specific entities and currencies. Long-dated liabilities may make an insurer a patient investor, creating an economic capacity to hold long-duration assets. That capacity changes when a contract allows surrender, borrowing against policy value, early termination or collateral calls. At the same time, a non-life insurer may face a sudden concentration of claims following a catastrophe even though its ultimate underwriting loss is limited by reinsurance.
Funding liquidity is the capacity to obtain resources for scheduled or contingent payments. Market liquidity is the capacity to execute transactions in assets promptly, in sufficient volume and at tolerable execution cost. Both are conditional on time, size, instrument and the availability of counterparties. The market price quoted for a modest order is insufficient evidence of the price at which an insurer could sell a substantial portfolio under stress. An asset eligible for a private repo might cease to provide the expected proceeds if its valuation falls or the counterparty reduces its lending limit.
The IAIS (2020) liquidity risk management application paper explains why liquidity and capital require separate attention and identifies activities that can intensify shortages. A maturity-matched balance sheet may still have mismatched payment deadlines. Consider an insurer whose assets and economic liabilities respond similarly to an increase in interest rates. Where a derivative hedge loses market value, margin may need to be posted today; the reduction in the economic value of distant contractual payments creates no immediate bank deposit. Economic valuation may consequently provide a qualified picture of resilience while the institution experiences immediate payment difficulty.
Liquidity adequacy must be evaluated against the payment form. An obligation settled in cash is distinct from one that permits delivery of eligible securities. The latter can be met with assets that remain saleable or repo-eligible for other purposes only until they become encumbered. The delivery process also consumes time and requires functioning settlement and custody arrangements. The concept of available resources therefore depends on the legal power to dispose of a security, its operational location, prevailing market terms and the settlement deadline.
Market frictions and amplification
The interaction between trading capacity and financing capacity has an established theoretical foundation. Brunnermeier and Pedersen's (2009) market and funding liquidity study describes a model in which margin and funding constraints can reinforce deteriorating market liquidity. Its published abstract provides the conceptual mechanism used here; the monograph does not recalibrate the authors' model or claim that every market decline displays a liquidity spiral.
An insurer selling an asset to meet a payment accepts the execution price available at the relevant time. The trade may reduce the observable price of similar assets. Other institutions that own related securities experience a deterioration in collateral values, risk limits or measured capital. Where those institutions respond with additional sales, the original shock travels through prices. Dealer balance sheets and alternative buyers determine whether the transactions can be absorbed without exceptional price movement. If many institutions choose the same instruments as liquid reserves, their individual preferences organize a common vulnerability.
The corporate bond evidence summarized by Ellul, Jotikasthira and Lundblad (2011) in Regulatory Pressure and Fire Sales associates greater regulatory constraint with a higher propensity among sampled insurers to sell downgraded bonds. This is a bounded empirical result from the published abstract. It supports examination of regulatory incentives in stressed trading, rather than a prediction that all solvency rules induce procyclicality. Changes in credit quality can justify asset sales independently of regulatory constraints. Exposure, portfolio mandate and the timing of the downgrade must therefore be separated from the mechanism attributed to the rule.
A systemic liquidity event arises when mutually interacting responses impair functions important beyond the initial distressed institution. An insurer with a temporary funding shortage can often arrange private financing, sell assets gradually or receive an intragroup transfer. Those options might contract together when shocks affect many counterparties. The externality emerges because the selling institution considers the cost of its own trade, while its effect on collateral and financing conditions is distributed across other holders. This supplies an economic rationale for collective resilience measures and, under exceptional conditions, market-wide intervention.
Historical transformation of the liquidity problem
The crisis involving American International Group (AIG) illustrates insurance-affiliated financial activities with rapid cash-payment demands. The New York Fed's AIG facility record documents the sequence of public lending and restructuring beginning in September 2008. Market turbulence in 2020 and the British gilt disturbance in 2022 added evidence about distributed demands for cash and collateral across non-bank institutions. The relevant historical development is an expansion of diagnostic attention from one highly interconnected conglomerate toward the market functions that multiple distinct institutional forms depend upon.
The historical cases also distinguish preventive liquidity resilience from contingent crisis capacity. Insurance institutions can alter product terms, asset liquidity, hedge collateral arrangements and funding diversification within their commercial and legal constraints. Central banks can create domestic-currency settlement resources and may provide exceptional finance within their mandates. Fiscal authorities can authorize guarantees or capital support where risks of permanent loss justify a different decision process. Resolution authorities work with claims hierarchy, transfer of liabilities and continuation of critical functions when an institution's viability has failed. The institutional separation matters because cash creation alone does not restore a balance sheet with a permanent economic deficit.
A claim on the state can itself change the structure of private incentives. The expectation of generous assistance may encourage firms or their creditors to accept more funding risk. Conversely, complete reliance on each institution's private contingency arrangements may be insufficient when the whole market loses trading capacity. The appropriate object of analysis is thus the combination of private preparation, supervisory intervention, market backstop and a credible method of allocating losses. Crisis assistance is one component of the institutional system, not a substitute for insurance prudence.
Primary Institutional Documents and Comparative Research Design
Units of analysis and evidence hierarchy
The comparative design uses the legal entity as the basic unit for payment and facility eligibility, the group as the unit for consolidated economic exposure, and the market as the unit for trading capacity and external effects. These units are related but not interchangeable. An insurance group's reporting may consolidate a solvent regulated subsidiary with a parent suffering a shortfall. An aggregated insurance-sector position may combine cash-rich institutions with entities whose obligations cannot be financed at the required time. A securities market can experience dysfunction even where the institutions trading within it remain individually viable.
Each event is classified by jurisdiction, period, primary shock, affected participant, channel of liquidity transmission, public instrument and recorded termination or contingent status. The study includes the following purposively selected observations:
| Observation | Jurisdiction and unit | Evidence and intervention |
|---|---|---|
| AIG, 2008–2011 revolving facility | United States; identified group and subsidiaries | Firm-specific emergency finance and restructuring |
| Corporate credit market, 2020–2021 | United States; asset market and SPV | Secondary-market purchase facility |
| CTRF, 2020 and later facility design | Canada; eligible market counterparties | Contingent secured Canadian-dollar finance |
| Gilt market, 2022–2023 | United Kingdom; LDI-related selling and market intermediation | Temporary sovereign bond purchases and unwind |
| CNRF, January 2025 notice | United Kingdom; eligible non-banks including insurers | Contingent collateralized lending design |
| EIOPA 2024 and UK SWES 2024 | European and UK supervisory samples | Hypothetical stress and behavioral assessment |
This table does not assign the same causal status to every row. A public facility that disbursed funds produces an implementation record. An application notice establishes the legal-operational design and neither demonstrates the facility's use in a systemic shock nor measures its effectiveness. A stress exercise establishes the results conditional on its specification, while a historical observation establishes events actually recorded. The comparison uses those differences to prevent the inadvertent conversion of simulated shortfalls into observed losses.
Documentary critique gives priority to central bank transaction accounts and statutory materials for legal and operational questions. Supervisory guidance establishes the intended architecture of prudential controls, not proof of compliance by each firm. Independent official audit findings help test the account offered by the implementing authority. Peer-reviewed literature is employed for explicitly identified conceptual and sample-specific results; where only abstracts are accessible, claims are limited to those summaries. There is no independent estimation using confidential insurer transaction files.
Cross-border comparability and measurement
National markets differ in surrender conditions, product guarantees, group organization, supervisory treatment and settlement arrangements. The role of an insurer in one country's government bond market may differ considerably from its role in another. A public support amount is always interpreted in its own currency, valuation date and accounting category. US dollars of authorized assistance, pounds sterling of acquired gilts and euros of projected liquidity deficiency describe fundamentally different observations.
Comparative statistical analysis requires a stable denominator. Asset-based market share measures balance-sheet size; gross written premiums measure contractual business flow; claims paid measure realized cash disbursement; incurred claims include changes in estimates of ultimate liabilities. The denominator for emergency cash needs might be total assets, short-horizon contractual obligations or the stock of assets eligible for a particular facility. A percentage calculated with one denominator cannot be compared directly with a percentage calculated using another. This is especially important when data from large international groups are combined with domestic supervisory populations.
The IAIS (2022) ancillary liquidity metrics announcement characterizes liquidity indicators as complementary macroprudential information and explains why multiple measures are used. The 2025 Global Insurance Market Report documents the one-year insurance liquidity ratio, with year-end 2024 observations from individual insurer monitoring. The measure relates available sources to liquidity needs over its specified horizon; reported ratios were strong across the principal groups, with heterogeneous firm movements. That observation supports a general assessment of the monitored population and horizon. It cannot establish uniform access to cash on the next settlement day.
The IAIS's 2026 midyear monitoring update reports preliminary estimates of a slight increase in aggregate systemic risk scores at year-end 2025, with asset liquidation and interconnectedness among the areas of increase. A preliminary global score is a supervisory signal rather than an estimate of the probability that an insurer will require public assistance. The two reports also refer to different observation dates. Their joint use here illustrates why stable aggregate resources can coexist with growing transmission concerns.
Method of inference and its limits
The study compares mechanisms rather than attempting a pooled regression on a handful of historical events. The AIG case identifies a legally specific response to a conglomerate's obligations. The 2020 and 2022 cases identify market-wide intermediation and funding channels. The Canadian and British contingent lending designs identify a possible route to transmit central bank cash to eligible participants. EIOPA and the UK scenario exercise demonstrate how alternative assumptions affect estimates of private resilience. These observations are complementary because each describes a different link in the adjustment process.
The underlying events were accompanied by other economic and policy changes. A decline in volatility after an announcement can result from changing expectations, adjusted risk positions, fiscal measures, private market responses or concurrent central bank actions. An intervention is also endogenous: policymakers normally choose to act when observed stress is unusually severe. Before-and-after comparisons therefore do not identify the causal effect of a facility without additional assumptions or an appropriate comparison population. The explicit arithmetic used below reconciles recorded amounts or illustrates contractual cash-flow constraints; it does not supply missing counterfactual observations.
Research transparency requires recording absent data as well as accessible data. Individual borrowers' payment schedules, collateral pools and firm liquidity usage are frequently unavailable publicly. The dataset assembled here consists of documentary observations rather than a representative panel of all insurance companies. The evidence ledger records the exact URL, scope of source review and claim supported. No public market intervention is credited with preventing a specified insurer default absent an identified and substantiated counterfactual.
Funding Shortages and the Transmission of Liquidity Stress
Sources of funding shortages
An insurer's payment pressure can be organized into three overlapping classes. Technical flows arise from premiums, claims, surrender payments, commissions, expenses and reinsurance. Financial flows arise from asset maturities, purchases, derivative margins, repo maturities, securities lending and debt service. Operational flows arise from settlement failure, disrupted collection, inaccessible bank accounts and arrangements with service providers. The timing of any source of liquidity depends on the enforceability and performance of the corresponding obligation.
Life insurers can experience concentrated withdrawals when policyholders find alternative returns more attractive, need cash or lose confidence in an institution. The response depends on product design: surrender charges, market-value adjustments, tax effects, guaranteed returns, switching costs and settlement periods all matter. An analysis of surrender pressure should compare similar contract cohorts and distribution channels. A uniform lapse rate imposed on the entire balance sheet is a useful stress assumption only when explicitly defined as such.
Non-life insurance has a different cash-flow mechanism. After a major event, numerous claims may require advances and final settlement before a reinsurer reimburses the primary insurer. Ultimate net loss can be modest relative to gross payment pressure. Claims validation, dispute resolution and reinsurer payment schedules determine this timing. Recoverables may be economically valuable while failing to meet tomorrow's funding needs. Operational outages at loss adjusters, payment systems or claims administrators can further disrupt the cash schedule.
Derivatives connect insurers' long-run risk hedging with short-run liquidity. A margin agreement may require cash or collateral at frequent intervals, and variation margin depends on changes in the market value of positions. Initial margin may increase with volatility or revised risk calculations. The exact direction depends on the position; a rate increase does not create the same call for every insurer. In securities lending, the borrower of a security can return the security and demand the return of cash collateral. Where the insurer has reinvested collateral proceeds in longer-dated or credit-sensitive instruments, a term mismatch emerges.
The ECB's (2020) euro area insurance assessment identifies reduced inflows and potential increases in claims, policy withdrawals and derivative cash demands during the pandemic. Its value for the present investigation lies in the joint direction of different cash flows. A previously adequate funding projection can deteriorate from both sides without any single payment category crossing a historical extreme.
Collateral, encumbrance and cross-entity restrictions
A liquidity inventory begins with assets and their contractual availability. Cash in the relevant currency is the immediate settlement resource. Other assets must be evaluated for their expected realization value, the speed of their conversion, any existing liens, eligible counterparties and the possibility of replacement collateral calls. A Treasury or gilt may possess high market quality while being unavailable because it secures an outstanding repo. The relevant quantity is unencumbered resources within the entity entitled to use them.
Group funding presents a closely related problem. A consolidated balance sheet eliminates intercompany receivables and liabilities. Yet a crisis management team needs to identify which legal entity controls the money and which legal entity owes the payment. Policyholder protection rules, guarantees, contractual conditions, exchange controls, dividend restrictions and local supervisory decisions may constrain transfers. The existence of substantial aggregate group assets cannot alone establish an ability to pay a parent company's unsecured creditors.
Currency adds another constraint. An asset in sterling can finance a dollar obligation only if an exchange or collateral arrangement produces dollars within the required period. Foreign exchange swap capacity and bank payment limits can be reduced during stress. Liquidity coverage expressed in a common reporting currency may therefore conceal a shortage in the actual settlement currency. A realistic liquidity assessment examines material currencies before consolidating them.
The funding gap should also identify the intended use of an asset. An unencumbered bond can be retained as a prudential buffer, posted as collateral, sold or used for borrowing. Each action may be individually possible. Counting the same bond in multiple simultaneous funding paths overstates resilience. A facility commitment is subject to a similar problem when its contract permits conditions, when the lender's capacity is correlated with the shock, or when drawing requires a prior notice period. The contingent resource is the amount that remains executable under the scenario and deadline.
System-wide interaction and the 2020 shock
The FSB (2020) Holistic Review of the March Market Turmoil describes liquidity pressures in multiple financial activities and the limited ability of dealers to absorb selling. From an insurance perspective, these channels combine with the need for a liquidity reserve that may be held through an investment fund rather than directly. If many institutions redeem shares simultaneously, the fund may finance their requests through sales in the same bond markets on which the institutions depend for collateral and borrowing. The institution's nominally diversified liquidity portfolio can thus depend on a narrower set of market intermediaries.
This transmission is conditional rather than automatic. A fund may hold adequate cash or use liquidity management tools; a dealer may have spare balance-sheet capacity; an insurer may post securities instead of selling them. Which action is feasible depends on the contracts and the sequence of shocks. Institutional research should distinguish the demand for cash from the actual volume of securities sold and identify the intermediary that financed the purchase. Otherwise an observed sale might be attributed to liquidity strain even where it reflects portfolio rebalancing or changed credit assessment.
System stability benefits when institutions can defer unnecessary transactions while meeting binding payments. It also benefits when well-capitalized investors with available liquidity can absorb sales at prices reflecting underlying risk. A facility that prevents every asset repricing would suppress price information and encourage exposures based on policy protection. The economic objective is to preserve the functioning of trading and financing mechanisms, allowing market prices to continue revealing changes in credit and interest-rate fundamentals.
Evidence from supervisory scenarios
The 2024 Insurance Stress Test provides a useful measurement example. Its hypothetical three-month liquidity component reports aggregate cash plus net flows of EUR 110.3 billion in the baseline, negative EUR 40.9 billion under the fixed-balance-sheet assumptions, and positive EUR 61.1 billion under the constrained-balance-sheet assumptions with reactive actions. The mass-lapse scenario assumes payout of surrenders within the specified horizon. These results are modeled positions, not evidence of actual borrowing demand.
The separately described scope of the EIOPA exercise identifies 48 undertakings across 20 jurisdictions for the capital component, covering 75 percent of EU-wide assets on the Solvency II measure. That denominator cannot automatically be attached to the liquidity observations. The result of greatest scientific relevance is the distinction between available cash, modeled net flows and other assets that can be mobilized. A positive sustainability measure may include assets whose conversion is necessary to meet the payments that generated a negative cash-flow position.
A second perspective comes from the Bank of England's (2024) system-wide exploratory scenario exercise. Its insurance annex describes simulated margin calls, the use of several forms of collateral, and heterogeneous management actions. The report indicates that around 30 percent of insurer initial and variation margin calls in the exercise are met with corporate bond collateral. Some participants would sell assets to rebuild buffers, while others could consider purchasing credit assets. This is a useful demonstration of institutional heterogeneity; the exercise does not predict the precise behavior of insurers under a future shock.
Sectoral stress modeling should therefore combine the direct obligation schedule with endogenous asset sales and private lending responses. A firm-level scenario may assume that a securities market remains liquid. A market-wide scenario tests the implications of multiple firms attempting to use that market at once. Neither level alone supplies the whole answer. The same process should be repeated under alternative collateral eligibility, weaker bank credit capacity, settlement delays and slower reinsurance recoveries.
Liquidity Facilities and Comparative Institutional Economics
Emergency finance and the AIG case
In September 2008 the Federal Reserve authorized a revolving credit facility for AIG. The initial commitment was USD 85 billion and peak use USD 72 billion, although the commitment changed as authorities restructured the assistance. According to the New York Fed facility table, the revolving credit facility was repaid in January 2011. This record establishes specified balances and dates, alongside the different timing of other AIG-related arrangements.
The GAO review of AIG assistance (2011) reports USD 182.3 billion in aggregate assistance authorized through a series of measures including Treasury support. It documents the USD 40 billion Treasury share purchase and the pressure from AIG's securities lending and credit protection business. The number measures authorized federal assistance, rather than cumulative economic loss or a single draw on the central bank. The firm's legal and financing structure is central to explaining why a simple comparison with the liquidity facilities of later crises is inadequate.
Two special-purpose transactions targeted sources of distress more directly. Maiden Lane II transaction terms show that the New York Fed financed a vehicle with a USD 19.5 billion senior loan to acquire residential mortgage-backed securities from the securities lending portfolios of regulated insurance subsidiaries. The subsidiary sellers deferred USD 1 billion of the purchase price behind the senior financing. The transaction terms distinguish estimated fair value and par value of the securities, and specify a repayment priority. Public finance was thus coupled with asset transfer and allocation of risk between claims on the vehicle.
Maiden Lane III addressed another channel by acquiring assets connected to AIG's credit default swap contracts. The GAO documents USD 24.3 billion in actual New York Fed lending to the vehicle and the termination of covered contracts. Eliminating prospective collateral calls can reduce future financing needs in a way that repeated advances merely accommodate. The price and terms of transactions also influence the distribution of losses and benefits to counterparties; these features must remain visible when judging the effectiveness of support.
AIG demonstrates that firm assistance can acquire the characteristics of restructuring. A lender facing an impaired borrower may change maturities, pricing, collateral and ownership arrangements to make repayment feasible. Such changes create an obligation to identify which part of the transaction addresses temporary cash timing and which part bears or reallocates economic loss. Subsequent repayment does not by itself establish that the original terms represented the best possible public decision, since alternative outcomes and ex ante risk were uncertain.
The post-crisis displayed text of Federal Reserve Act section 13(3) specifies broad-based eligibility and other conditions for emergency programs. The historical AIG facility preceded the current statutory formulation. This distinction prevents using a 2008 legal action as proof of a present general ability to provide bespoke lending to any insurer. Public support is always constrained by the legal mandate in effect at the relevant time.
Market-wide purchases during the pandemic
The Federal Reserve established the Secondary Market Corporate Credit Facility in March 2020 to support credit provision through liquidity in corporate bond markets. It purchased eligible bonds and exchange-traded funds through an institutional vehicle. Insurers holding or trading similar instruments could benefit from improved market functioning without direct access to central bank lending. A market facility can thus transmit a stabilizing effect through asset prices and dealer intermediation across a wider group than the named counterparties.
The New York Fed implementation record describes central bank finance to the special-purpose vehicle and Treasury equity participation, with disposal or maturity of all facility holdings completed by August 31, 2021. The combination identifies the location of the public credit exposure. It also distinguishes closure to new purchases from the subsequent liquidation of outstanding assets. Exit requires the funding and assets of the vehicle to be managed after the emergency flow of operations ends.
A market purchase instrument transfers ownership of eligible assets to the purchasing institution. The seller receives cash without an obligation to repurchase the security. The public balance sheet consequently bears market and credit risk subject to the program terms. By contrast, collateralized finance creates a loan or economically similar repurchase obligation, leaving the borrower responsible for returning cash. The public lender bears counterparty and collateral realization risk. These are different responses to related market disturbances and can have different effects on private balance-sheet leverage.
The 2020 programs also illustrate how the expectation of future purchasing may influence market participants before the first trade. An announcement can alter the anticipated capacity of prospective buyers, affecting the willingness to retain or acquire bonds. A small actual purchase volume can coexist with a large announcement effect, but an observed volume alone cannot identify such an effect. Evaluation requires market data around communications and operations, appropriate control securities, and adjustment for other news and programs.
The British gilt market disturbance
In September 2022, rising yields and deteriorating positions in leveraged liability-driven investment strategies increased the pressure to obtain cash and collateral. Pinter's (2023) An Anatomy of the 2022 Gilt Market Crisis reports transaction-level findings linking greater pre-crisis repo and swap exposure to more gilt selling within the combined LDI-pension-insurance category. The source groups several institutional forms, so the findings do not establish that insurers were the primary cause of the episode.
The Bank of England's October 2022 intervention statement explains the time-limited purpose of its purchases, including the opportunity for LDI funds to strengthen their resilience. The separate expanded collateral repo arrangement operated for a further period through bank counterparties. Restoration of trading capacity and the availability of funding worked through different operational mechanisms. Their beneficiaries overlapped only insofar as institutions were connected to the relevant market and eligible intermediaries.
The Bank's gilt portfolio unwind announcement records GBP 19.3 billion of purchases from September 28 through October 14, 2022, divided into GBP 12.1 billion of long-dated conventional gilts and GBP 7.2 billion of index-linked gilts. The temporary portfolio had been fully sold by January 12, 2023. The completed unwind demonstrates the operational separation of a temporary market-stability transaction from an indefinitely maintained portfolio of securities. It leaves open questions about the counterfactual degree of stress absent the intervention.
An important institutional lesson concerns the capacity of dealers. A large institutional sale requires a buyer who can bear changes in price or intermediate the transaction until another buyer appears. Even when fundamental credit risk in sovereign assets is low, the risk from carrying inventory during volatility may restrict available trading capacity. A backstop buyer can interrupt forced disposals by providing an additional market outlet. An appropriately designed credit facility may instead reduce pressure to sell if the institution can pledge the assets and later repay from credible resources. An intervention directed at trading conditions must be assessed against the actual source of the selling pressure.
Contingent lending in Canada and the United Kingdom
The Bank of Canada's April 2020 CTRF activation notice specifies Canadian-dollar funding for a one-month term against qualifying government-related securities. The operation was activated effective April 6, 2020. The standing Contingent Term Repo Facility description identifies eligible non-bank institutions and primary dealers, discretionary activation, application and master agreement processes. It also distinguishes planned readiness tests from interventions prompted by market conditions. This comparison shows why preparatory access and actual usage should be represented separately.
The Bank of England's January 2025 CNRF market notice opened a distinct facility for applications, including insurance companies subject to eligibility and appropriate financial health. The notice envisages secured short-term finance against gilts when serious dysfunction threatens financial stability, ordinary market mechanisms are insufficient and lending can address the problem. Terms are to be calibrated at activation; the notice itself is not an activation. The July 2026 Financial Stability Report reiterates the role of NBFI vulnerabilities and the opening of this facility for applications in January 2025.
A contingent facility has a preparatory function even without active lending. Counterparties must establish legal agreements, configure settlement accounts, demonstrate the ability to deliver securities and comply with limits. A market-wide shock is an unsuitable moment to begin solving these prerequisites. Readiness processes reduce the time between a central bank's decision to activate an instrument and its capacity to deliver finance. They also help identify which institutions, currencies and securities remain outside the operational perimeter.
An eligible insurer can therefore have access in principle without having drawn any public money. The contingent facility should never be entered as risk-free cash in a baseline funding plan. Its use depends on the authority's activation decision, continued eligibility, acceptable collateral, borrowing limits and functioning operations. The central bank's mandate is directed to financial stability rather than an unconditional guarantee of the insurer's liabilities.
Public Support Terms and Quantitative Evaluation
Eligibility, financial health and loss allocation
Public liquidity support can protect market functioning while transferring risks to public institutions. A core design question is whether assistance finances a temporary timing mismatch, provides leverage for asset disposal, supports a troubled but viable borrower, or contributes capital against an economic loss. Economic and legal solvency must be assessed independently of the immediate shortage. An institution that cannot restore viability with credible changes requires a framework for allocating losses and maintaining protected functions.
The design of a broad-based facility must define its counterparty scope. Eligibility tied to sufficiently material market participation targets the instruments that can address a systemic externality. Solvency or financial health requirements and collateral limits reduce public exposure. At the same time, a narrow collateral or size threshold can exclude firms whose correlated actions contribute to dysfunction. A supervisor may need to assess the aggregate transmission of institutions outside the public lending perimeter and their reliance on commercial banks and other private channels.
The British CNRF notice illustrates the balance through an ordinarily expected gilt-holding threshold of GBP 2 billion and a continuing financial-health condition. Its indicated borrowing limits reflect 50 percent of gilt holdings rounded upward to the nearest GBP 1 billion, subject to discretion. Its expected lending horizon is roughly one to two weeks, with precise terms determined when activated. The design also contemplates haircut and concentration controls. These parameters are neither a universal insurance liquidity requirement nor a commitment to lend a fixed amount in every market environment.
Pricing affects participation and incentives. A lending rate above ordinary private-market terms, but accessible under exceptional stress, can help preserve normal private intermediation. A rate set excessively high relative to the borrower's repayment capacity may intensify solvency concerns. The initial AIG facility's pricing was subsequently changed; the official audit connects the restructuring to worsening conditions and a need to reduce the burden. This illustrates the trade-off between protecting public interests and establishing a feasible repayment path. Different counterparties and lending objectives justify different terms.
Collateral rights reduce expected loss only when the public lender can legally enforce them and realize an asset's value following counterparty failure. Haircuts provide a financial margin against price movement, while concentration limits address exposures to individual securities and markets. The haircut should reflect realistic liquidation conditions and wrong-way risk. A lending program that systematically accepts collateral at inflated valuations can function as an implicit subsidy. Conversely, abruptly raising haircuts at the peak of turmoil can force the sale of additional assets. Stable, transparent collateral practices and stress-calibrated cushions can mitigate this trade-off without eliminating risk.
Cash-flow arithmetic and stress-sensitive collateral
The following accounting illustrations employ invented numbers in millions of a single currency. They are not data on a real insurer, estimates from a fitted model, or regulatory thresholds. For a single legal entity and horizon, define C as available opening cash, I as collectible cash inflows, O as binding cash outflows, and F as executable net new financing. The simplified funding balance is C + I + F − O. An uncovered funding need exists when this quantity is negative.
Suppose an insurer begins a five-day window with 120 of cash. Collectible premiums, investment payments and recoveries yield 45. Claims and expenses require 75, surrender payments 60 and cash variation margin 90. Total required payments are 225, and the funding gap without new finance is 60. The calculation treats both claims and surrender payments as cash commitments and assumes the collateral demand cannot be met directly in securities. A privately arranged loan or equivalent sale producing 60 would balance the cumulative flows only if funds arrive in time.
Suppose the insurer has unencumbered, eligible government bonds worth 100 at current market prices. At a 10 percent haircut, collateralized financing can generate up to 90, assuming all bonds remain eligible and no separate counterparty limit. Financing the 60 gap requires collateral worth 60 divided by 0.90, or 66.67. If the asset value falls to 80 and the haircut rises to 25 percent, proceeds from the whole pool become 60. At that point the gap is covered only exactly; there is no remaining cushion in the pool.
If a further 15 of margin is required after that repricing, total funding need becomes 75 and the maximum feasible loan remains 60. The residual shortage is 15. This simple arithmetic reveals the importance of jointly varying collateral value and cash obligations. Treating the haircut as a fixed percentage independent of the shock would obscure a substantial part of the mechanism.
| Scenario | Cash gap before financing | Collateral market value | Haircut | Maximum proceeds | Residual gap |
|---|---|---|---|---|---|
| Initial stress | 60 | 100 | 10% | 90 | 0 |
| Repricing and larger haircut | 60 | 80 | 25% | 60 | 0 |
| Additional margin call | 75 | 80 | 25% | 60 | 15 |
Settlement timing adds a separate restriction. If 165 in available cash and inflows is reached only on day five, while 170 in payments is due by day two and only 10 of the inflow has arrived by then, the earlier shortfall equals 170 − (120 + 10), or 40. The end-horizon gap of 60 cannot describe both dates. A facility that settles on day three can contribute to the final cash balance and still fail to prevent the earlier missed payment. A cash-flow ladder should measure each interval, record operational settlement time and identify the maximum cumulative shortfall.
A group example further illustrates the perimeter problem. Assume subsidiary A has 80 of freely available cash and 110 in immediately due obligations, while subsidiary B has 100 of cash and 20 in obligations. The consolidated surplus equals 180 − 130 = 50, yet A has an individual gap of 30. If at most 10 can lawfully and operationally transfer from B before settlement, A retains a gap of 20 and B retains surplus cash of 70. The group-level balance remains 50, while the ability to make the individual payments depends on legal and operational transfer capacity. These examples deliberately exclude interest, tax, intra-day movements and valuation effects beyond the stated assumptions.
Reconciliation of documented amounts and counterfactuals
Historical balances must be treated according to their units. AIG's USD 72 billion maximum draw was approximately 84.7 percent of the original USD 85 billion commitment. This calculation uses the original commitment and peak draw from the official table; it does not represent facility utilization after later restructuring. Keeping these quantities separate prevents the common mistake of adding successive facility commitments to infer public losses.
The British GBP 12.1 billion conventional and GBP 7.2 billion index-linked holdings sum to GBP 19.3 billion. Their respective shares of the purchase portfolio were about 62.7 percent and 37.3 percent. The denominator is the reported total purchase holdings, rather than the stock of all government debt or the amount of insurance-sector exposure. The figures identify the composition of the public transaction without estimating the market-price effect.
The EIOPA scenario's constrained balance-sheet cash position of EUR 61.1 billion exceeds the fixed-balance-sheet result of negative EUR 40.9 billion by EUR 102.0 billion. It remains EUR 49.2 billion below the EUR 110.3 billion baseline. These are calculated differences between scenario positions, not actual money lent by authorities. The baseline comparison demonstrates how modeled reactive actions can restore a positive aggregate position while leaving it below the unstressed result.
The three reconciliations use amounts with different accounting meanings. None supplies a common cross-country measure of net economic benefit. A meaningful welfare evaluation would account for executed financing, interest income, financing cost, expected credit loss, guarantees, risk borne during the intervention, distributional effects and changes in payment continuity. Even a positive ex post financial return can coexist with a significant ex ante public risk transfer.
Private resilience, public accountability and exit
The FSB (2024) margin and collateral preparedness recommendations provides a cross-sector approach including stress tests, funding contingency plans, governance and operational collateral practices. Insurer preparedness reduces the need for exceptional public finance by diversifying available liquidity and making it executable before stress. The relevant private cost includes carrying cash, purchasing commitments and maintaining systems. These costs should be weighed against disrupted payment, fire-sale discounts and the social consequences of market impairment.
Public reporting can separate operation size, outstanding balances, collateral type, realized losses, termination dates and conditions attached to access. Aggregate disclosure protects legitimate operational confidentiality while providing information about public exposure. Facility governance also needs a documented decision process for activation and deactivation, procedures for conflicts of interest and an allocation of responsibility among the central bank, supervisor, fiscal authority and resolution authority.
Exit is the transition from temporary public capacity back toward private risk bearing. In a loan, it requires repayment and release of collateral; in asset purchases, it requires cessation of additions and management or disposition of the acquired portfolio. An early exit may revive dysfunction if the private adjustment is incomplete. An indefinite presence may alter pricing, reduce incentives for private market-making and sustain vulnerable business models. A time horizon tied to the cause of market impairment and observable market functioning is consequently important for accountability.
Integrated Scientific Synthesis and Future Research
Comparative findings
The first finding of this investigation is that a shortage at an insurer and illiquidity in the markets supporting it are linked but distinct institutional states. The source of a shortage may be a surge in insurance payments, unwinding of a securities financing arrangement, derivative settlement or a failure to transfer resources from another entity. The resulting sale of assets can disseminate the shock through prices and funding conditions. It follows that market liquidity is part of the practical ability of insurers to pay but is not equivalent to their economic solvency.
The second finding concerns the instruments used to address different problems. A firm-specific facility can directly meet a legal entity's payments, yet it raises particularly demanding questions about viability and the distribution of credit risk. A market-wide asset purchase adds trading capacity while transferring asset exposure. A secured contingent loan may permit a qualified institution to avoid forced sale while remaining responsible for repayment. A fiscal capital injection addresses the burden of economic loss more directly. Resolution can preserve essential functions when the institution's viability cannot be restored. Each response imposes its own legal burdens and distributional effects.
The third finding is that effective contingent finance depends on preparation. A central bank cannot rely on the economic value of a collateral pool without the operational ability to receive it. An insurer cannot assume the availability of a credit commitment without checking when it can be activated and drawn. Authorities must identify the institutions capable of transmitting or mitigating the shock, and the infrastructure that connects them. Preparedness testing, valid settlement instructions and reliable documentation give practical effect to the stated lending architecture.
The fourth finding concerns the interaction of market and prudential policy. Regulatory limits support loss-bearing capacity and confidence, yet synchronized attempts to restore portfolio ratios can create additional selling. A supervisor examining a sector-wide shock should anticipate the likely sequence and price impact of management actions. Maintaining regulatory credibility requires transparent, legally permitted decisions rather than concealing impairment by redefining measurement after a loss. The assessment of capital relief, temporary liquidity measures and orderly resolution should remain institutionally differentiated.
Implications for insurance markets and policyholder protection
The social purpose of liquidity assistance becomes clearest through the obligations whose interruption would otherwise occur. Timely payment enables recovery after insured events, funds retirement income and supports the contractual expectations of policyholders. These functions can be protected through firm continuity, orderly transfer of contracts, market stabilization or other arrangements suited to the circumstances. Assistance to an insurance-affiliated entity should be evaluated in terms of the affected functions and externalities, including who receives the immediate benefit and which parties bear residual risk.
Market stability is also valuable to insurers without shortages. An orderly securities market allows an insurer to maintain its asset allocation, adjust hedges, reinvest premiums and evaluate market prices. Forced market closure can impair even firms with considerable unencumbered resources. Conversely, a sufficiently liquid insurer can stabilize markets by buying assets sold by other institutions. An analytical account of insurers as either sources of risk or stabilizers is incomplete unless it identifies the balance-sheet conditions, mandates and incentives that determine their behavior.
The cross-border implications are substantial. An international insurance group may rely on private funding from banks located in different jurisdictions, hold collateral through foreign custodians and have policies denominated in several currencies. Emergency measures in one domestic market may relieve part of its needs while leaving other entities exposed. The information exchanged among supervisors should help identify material payment obligations, actual sources of collateral, transfer restrictions and potential contagion. It should also preserve local legal rights, especially protection of policyholders' assets.
A comparative framework for evaluating public support consists of factual questions: Which legal authority acts? Which entity can receive cash? Which instruments qualify as collateral or purchases? What loss can be transferred to public balance sheets? Is the support discretionary or committed, and which conditions terminate access? How is viability assessed? Which private remedies are expected first? What event constitutes an exit? Answers are specific to each instrument, jurisdiction and date. The method organizes the comparison without claiming a new scientific theory or universal policy formula.
Future empirical research
The most useful extension would combine entity-specific payment schedules with asset-level financing and transaction information. A structured dataset would include the legal entity identifier, business model, jurisdiction, currency, reporting date, surrender terms, expected and realized claims, reinsurance recovery dates, derivative margin, collateral type, existing liens and eligible private or public facilities. Transaction identifiers should distinguish trade, settlement and contractual maturity. Standardization of these records would allow researchers to investigate not just whether resources exist, but whether those resources can meet the deadlines generated in stress.
Market effects could be studied using bond transaction data, quotes, spreads, turnover, pricing dispersion and dealer balances around the relevant intervention. Identification would require the definition of eligible and comparison securities, controls for preexisting trends, dates of anticipated announcements, and methods to address endogenous selection. Observed post-intervention changes could be compared with securities or markets differently exposed to the same macroeconomic information. Any event study should distinguish announcement effects from executed transactions and consider spillovers into the comparison group.
The private behavior response could also be assessed through changes in the distribution of unencumbered collateral, contingent funding lines, hedge margin arrangements and management actions before and after a policy announcement. Research should test whether more reliable backstops reduce excess defensive sales or encourage greater risk taking and lower private reserves. Both responses are economically possible. The effect depends on the credibility of support, access conditions, supervision and pricing, and may vary across life insurance, non-life underwriting and financial-market activities.
An international sample would need careful adjustments for local rules and reporting measures. Aggregate stress exercises and global liquidity indicators offer starting points, but their observations cannot be combined mechanically. Differences in business composition and denominators can drive variation that might otherwise be attributed to institutional quality. The more useful aim is to understand which specific arrangements make a unit of liquidity available when a given obligation is due and whether that availability reduces costs imposed on other market participants.
The documentary investigation establishes the conceptual and historical conditions for this research. Its own limitations include purposive case selection, lack of confidential legal-entity observations, limited access to selected academic full texts and the absence of a causal welfare estimate. The primary official material supports a comparison of announced terms and recorded operations. Numerical results from stress exercises are explicitly scenario-dependent. The arithmetic examples are accounting demonstrations, and the interpretation makes no claim of an independently discovered empirical law.
Conclusion
Insurance sector liquidity support during systemic shocks is a problem of connecting timely contractual payment capacity with the functioning of financial markets. The investigated cases establish that funding pressure can arise from insurance obligations and from the financial activities used to invest resources and manage risk. Short-term cash demands can coexist with substantial long-term assets. The ability to mobilize those assets in the proper entity, currency and period determines whether an institution can satisfy its immediate obligations.
The AIG experience reveals the interaction of emergency finance, asset transfer and financial restructuring. The pandemic and gilt-market cases illustrate the outward transmission of liquidity shortages through trading and intermediation. Contingent facilities in Canada and the United Kingdom demonstrate an institutional means of preparing access to secured public finance before market dysfunction becomes severe. Supervisory monitoring and stress exercises show why aggregate liquidity measures, individual payment capacity and market-wide behavioral responses must be assessed together.
The comparative evidence supports a defined division of institutional responsibilities. Insurers need to maintain credible liquid resources and private contingencies, supervisors need to examine common vulnerabilities as well as firm-level compliance, and public authorities need instruments that match the market failure while preserving legal accountability. Facility terms and exit conditions must reflect financing needs, collateral risk, economic viability and the allocation of financial losses. Assistance that preserves a viable payment function and reduces disruptive sales serves a different purpose from the rescue of investors against permanent losses.
The central analytical conclusion is that liquidity resilience is a temporal and institutional property of the insurance-financial system. The value of assistance depends on the actual sequence of payments, access to markets and finance, eligibility and the conversion of assets into settlement resources. This interpretation provides a grounded basis for comparative research and the evaluation of public support, with the documented limits of historical evidence preserved.
References
Bank of Canada. (2020, April 3). Bank of Canada Announces Activation of the Contingent Term Repo Facility. Source
Bank of Canada. (n.d.). Contingent Term Repo Facility. Source
Bank of England. (2022, October 14). Bank of England statement on end of gilt market operations. Source
Bank of England. (2023, January 12). Bank of England completes unwind of recent financial stability gilt purchases. Source
Bank of England. (2024, November). The Bank of England's system-wide exploratory scenario exercise final report. Source
Bank of England. (2025, January 28). Contingent NBFI Repo Facility (CNRF)—Market Notice 28 January 2025. Source
Bank of England. (2026, July). Financial Stability Report—July 2026. Source
Board of Governors of the Federal Reserve System. (n.d.). Federal Reserve Act, Section 13—Powers of Federal Reserve Banks. Current displayed statutory text. Source
Board of Governors of the Federal Reserve System. (n.d.). Secondary Market Corporate Credit Facility. Source
Brunnermeier, M. K., & Pedersen, L. H. (2009). Market liquidity and funding liquidity. Review of Financial Studies, 22(6), 2201–2238. DOI: 10.1093/rfs/hhn098. Source
Ellul, A., Jotikasthira, C., & Lundblad, C. T. (2011). Regulatory pressure and fire sales in the corporate bond market. Journal of Financial Economics, 101(3), 596–620. Source
European Central Bank (ECB). (2020, May). Financial Stability Review, section 4.3. Source
European Insurance and Occupational Pensions Authority (EIOPA). (2024). 2024 Insurance Stress Test. EIOPA(2024)0103699. Source
European Insurance and Occupational Pensions Authority (EIOPA). (2024). Insurance stress test 2024: exercise and results. Source
Federal Reserve Bank of New York. (n.d.). Actions Related to AIG. Source
Federal Reserve Bank of New York. (n.d.). Maiden Lane Transactions. Source
Federal Reserve Bank of New York. (n.d.). Secondary Market Corporate Credit Facility. Source
Financial Stability Board (FSB). (2020, November 17). Holistic Review of the March Market Turmoil. Source
Financial Stability Board (FSB). (2024, December 10). Liquidity Preparedness for Margin and Collateral Calls: Final report. Source
International Association of Insurance Supervisors (IAIS). (2020, June 29). Application Paper on Liquidity Risk Management. Source
International Association of Insurance Supervisors (IAIS). (2022, November 23). IAIS finalises liquidity metrics as an ancillary indicator for its Global Monitoring Exercise. Source
International Association of Insurance Supervisors (IAIS). (2025). 2025 Global Insurance Market Report. Source
International Association of Insurance Supervisors (IAIS). (2026, July). IAIS mid-year Global Insurance Market Report 2026 reflects insurance sector stability amid global uncertainty. Source
Pinter, G. (2023, March 31). An anatomy of the 2022 gilt market crisis. Bank of England Staff Working Paper No. 1,019. Source
U.S. Government Accountability Office. (2011, September 30). Financial Crisis: Review of Federal Reserve System Financial Assistance to American International Group, Inc. GAO-11-616. Source