When a Wildfire Fund Absorbs One Utility's Losses, Every Member Utility Pays

A wildfire that ignites in one utility's service territory now shows up as a non-cash charge on a different utility's income statement, in a different part of the state, several quarters later. That is not a metaphor. In the second quarter of 2026, Pacific Gas and Electric Company recorded $78 million of accelerated amortization on its share of a state-level insurance asset, and the disclosed trigger was settlement activity arising from the January 2025 Eaton Fire — a fire in Southern California Edison Company's territory, not its own.

That line item describes a regime change. The framing that dominated analysis from 2017 through 2020 — that a confirmed equipment-caused wildfire creates liability capable of exceeding a utility's equity value and forcing it into bankruptcy — was accurate for its moment. It is no longer the primary mechanism in California. Statute has converted an idiosyncratic, single-name solvency risk into a mutualized, sector-wide exposure with a defined depletion path. The risk was not removed. It was moved, and it now travels differently.

The Current Situation: A $1.6 Billion Loss That Produced a $9 Million Charge

Through June 30, 2026, Southern California Edison had recorded roughly $1.6 billion in losses tied to Eaton Fire settlements with insurance subrogation claimants and participants in its direct compensation program. Against that, it recorded expected recoveries of approximately $917 million from customer-funded self-insurance, $645 million from the state Wildfire Fund's Initial Account, and $70 million through federally regulated transmission rates. The residual charge to earnings, after tax, was approximately $9 million — the arithmetic consequence of a required shareholder contribution of about $12.5 million attached to the self-insurance layer.

Two details matter more than the ratio. The utility stated plainly that it remains unable to reasonably estimate a range of total losses: the $1.6 billion is settlements concluded to date, not an endpoint. And the customer-funded self-insurance layer was exhausted as of February 11, 2026, so everything settled after that date draws on the next layer down.

The fund administrator has indicated that roughly $21 billion of claims-paying capacity is available for the Eaton Fire. That is a large number and a finite one, and it is not reserved for a single event or a single utility. It is the shared balance behind every large investor-owned electric utility in the state.

The Transmission Mechanism: How the Exposure Moves

The architecture was created by California's 2019 wildfire legislation and extended in 2025. Statute read alongside the quarterly disclosures produces a four-layer picture.

Layer one — capitalization of the pool

The statute required initial contributions from large electrical corporations of $7.5 billion multiplied by a per-utility allocation metric, plus annual contributions of $300 million on the same metric. Smaller regional corporations contribute per customer — $625 per account initially, $25 annually. A parallel ratepayer contribution is collected through a non-bypassable charge, subject to the regulator finding it just and reasonable. The result is a pool funded roughly half by shareholders and half by customers, modeled with an estimated 20-year life.

Layer two — the burden-of-proof shift

The less visible half of the statute is procedural. A utility holding a valid annual safety certification is treated as having acted reasonably unless an intervenor can create a serious doubt about its conduct. Without that certification, the utility must affirmatively demonstrate reasonableness by a preponderance of the evidence. The governing standard is whether costs are consistent with actions a reasonable utility would have undertaken in good faith under similar circumstances.

This provision converts a wildfire from a solvency event into a cost-recovery proceeding. It does not eliminate the liability; it changes who carries the evidentiary burden. A lapsed or contested certification would move a utility back toward the pre-2019 posture without any change in the physical fire risk.

Layer three — the reimbursement cap

Draws on the fund are, in principle, reimbursable by the utility that caused the fire. But reimbursement is capped over a trailing three-calendar-year period at 20% of the equity portion of the utility's transmission and distribution rate base, excluding general plant and intangibles, measured in the year of the fire's ignition. For the Eaton Fire, that cap works out to approximately $4.3 billion.

The cap is the load-bearing element, converting an unbounded tort exposure into a bounded, rate-base-indexed obligation. It also creates a feature worth stating: because the cap scales with rate base, a utility that invests more in its network raises its own maximum assessment. Mitigation spending and assessment capacity move together.

Layer four — the shared asset, and why it reaches other balance sheets

Each participating utility carries its contribution to the pool as a long-lived asset and amortizes it over the fund's expected life. That expected life is a function of how fast the pool is being drawn down. When one member's settlements accelerate the draw, every other member's asset is worth less time, and the amortization schedule shortens for all of them.

That is the mechanical explanation for the $78 million of accelerated amortization recorded in the second quarter of 2026 by a utility with no involvement in the fire. Its own fund-related receivables at the same date were roughly $638 million for a 2021 fire and $61 million for a 2022 fire, against about $128 million already received for a 2019 fire. The pool is simultaneously an asset it draws on and one others draw down.

The 2025 amendment adds a further layer: a Continuation Account, effective September 19, 2025, available for fires igniting on or after that date. It is not automatic. It requires the fund administrator to determine that continuation is necessary, the regulator to authorize an extension of the non-bypassable charge, and potentially additional utility contributions. The same legislation carries a rate-base consequence — roughly $2.9 billion of wildfire mitigation capital expenditure approved on or after January 1, 2026 is expected to be excluded from the equity portion of one utility's rate base, which is a direct reduction in the earnings base of the mitigation itself.

The Counterfactual Is Still Running Elsewhere

The best evidence that the fund architecture is doing the work is a jurisdiction that does not have one. A utility serving a Pacific island system, whose equipment has been connected to a catastrophic fire on August 8, 2023, entered the second half of 2026 still describing conditions that could raise substantial doubt about its ability to continue as a going concern, and still describing dependence on raising capital and obtaining lender waivers to avoid debt acceleration. In the first half of 2026 alone it paid out approximately $478.75 million in settlement payments and recorded a further $154 million loss contingency, with its wildfire-related claims liability declining by roughly $136.2 million over the period.

That is the pre-2019 mechanism, observable in the present tense: causation attaches, claims exceed what operations can absorb, the capital structure becomes the shock absorber, and the credit question turns existential rather than actuarial. The physical hazard in both cases is similar. The divergence in credit outcomes tracks almost entirely to whether a statutory pool existed before the fire.

The same reading applies within California's own history. A Chapter 11 filing on January 29, 2019 by the state's largest utility, followed by a court-supervised victim trust, is what the mechanism looked like before the pool existed. Six years later, one major rating agency still assesses that company below investment grade, and it does not receive unsecured credit from energy counterparties — so collateral postings scale with any further downgrade. Statutory protection changed the trajectory of new events. It did not reverse the credit consequences of the old one.

A Historical Parallel: The Nuclear Liability Pool of 1957

The closest structural precedent is not another wildfire. It is the federal nuclear liability framework enacted on September 2, 1957, which faced an identical problem: socially necessary output, a low-probability accident whose damages could exceed any operator's capitalization, and private insurers unwilling to write the tail.

The solution was a layered pool that reads as a direct ancestor of the wildfire fund. Operators buy roughly $500 million of primary private insurance per reactor site. Beyond that, every licensee is subject to a retrospective assessment of up to about $158 million per reactor for a single incident — functionally identical to the shared-asset amortization above, in that a plant with no connection to the accident pays. Across about 95 reactors the secondary tier holds roughly $15 billion, bringing total coverage above $16 billion. The framework runs through December 31, 2065.

The most transferable element is the third tier. Retrospective assessments are capped at roughly $7.9 million per reactor per year. That is not a cap on total liability but on the rate at which it is collected, set so that funding a catastrophe does not bankrupt the parties funding it. California's 20%-of-rate-base, trailing-three-year cap performs the same function through a different variable. In both regimes, pacing the assessment matters as much as sizing the pool.

The parallel carries a warning. Nearly seven decades of nuclear pooling produced a system whose stated coverage has needed periodic legislative reauthorization to stay credible against inflation and changing severity estimates. Pools do not self-index. They require a legislature to revisit the arithmetic — which is what the September 2025 continuation mechanism represents, and why its conditionality matters.

When This Pattern Has Broken Before

Several conditions would break the framing above.

The pool may be adequate, in which case the mutualization channel is noise. A $21 billion stated capacity against $1.6 billion of settlements recorded to date is not obviously strained. If the eventual total lands well inside that capacity, the accelerated amortization at non-involved utilities is a one-time timing adjustment, not the leading edge of a depletion trend. Reading a single quarter's $78 million charge as a structural signal risks over-interpretation.

The endpoint is genuinely unknown, and the igniting utility says so. The disclosure that no range of losses can be reasonably estimated should be read literally, not as boilerplate. Any analysis assigning a confident total asserts something the primary source declines to assert.

Mutualization may not be the dominant channel. The exclusion of roughly $2.9 billion of mitigation capital from the equity rate base is a larger and more durable earnings effect than a quarter's amortization adjustment. If the question is what wildfire risk costs utility shareholders, the answer may lie mostly in how mitigation spending is treated for recovery, not in fund depletion at all.

The pool is jurisdictional, and the mechanism does not travel. None of this applies to a utility outside a state with such a statute. For those, the pre-2019 framing remains the correct one, and the comparison to California is actively misleading.

A safety certification is a condition, not a guarantee. The entire recovery architecture is downstream of the presumption of reasonableness. A contested certification, or a fact pattern that creates the serious doubt the statute contemplates, would return a utility to the older regime while leaving the mutualization effects on its peers unchanged.

What to Watch Next Week

• Fund administrator communications on remaining capacity. The administrator issued a study report on April 7, 2026 setting out policy options. Subsequent capacity statements are the highest-information disclosure in this system, because every member's amortization schedule derives from them.
• Regulatory filings on the non-bypassable charge. The Continuation Account cannot function without an authorized extension of that charge. Any procedural step on that authorization is a leading indicator for whether the pool is being extended or allowed to run to its original life.
• Settlements concluded after February 11, 2026. These draw directly on the shared pool rather than a customer-funded buffer, so the pass-through to other members is more immediate.
• Annual safety certification status for each member utility. Certification renewals are scheduled events with a binary character. A lapse would be more consequential than several quarters of amortization drift.
• Rate case treatment of mitigation capital. Watch specifically for whether newly approved mitigation spending is included in or excluded from the equity portion of rate base, since that determination sets the return on the largest category of wildfire-related expenditure.
• Rating agency commentary distinguishing pooled from unpooled jurisdictions. If spreads for member utilities begin trading closer to unpooled peers, that is the market pricing pool depletion directly.

Concrete Framework — What to Track, in Order

A checklist that follows the mechanism, not the headline.

1. Classify the jurisdiction first. Determine whether a statutory risk pool and a liability cap existed before the relevant ignition date. This separates two entirely different credit models; applying pooled logic to an unpooled utility, or the reverse, produces a wrong answer however carefully the rest is done.
2. Locate the cap and compute it. Where a cap is indexed to rate base — 20% of the equity portion of transmission and distribution rate base, excluding general plant and intangibles, in the ignition year — the number is calculable from public filings. Roughly $4.3 billion in the current case. Treat that figure, not total damages, as the relevant bound on the igniting utility's fund reimbursement.
3. Track pool capacity as a shared variable. Record the administrator's stated claims-paying capacity at each disclosure and cumulative draws against it. Falling capacity is a sector-wide input flowing into every member's amortization assumption.
4. Read the amortization line at utilities that had no fire. Accelerated amortization of a fund-related asset at a non-involved utility is the earliest observable evidence that the shared pool's assumed life has shortened.
5. Verify safety certification status annually. The presumption of reasonableness is the hinge of the recovery chain. Confirm it is current for each utility under review and note any intervenor challenge — the statutory test is whether serious doubt can be created, not whether negligence can be proven.
6. Separate mitigation-capex treatment from liability exposure. These distinct earnings channels are frequently conflated. Capital excluded from the equity portion of rate base — roughly $2.9 billion here — reduces the earnings base permanently, whereas a fund draw is a timing and capacity question.
7. Hold the unpooled case as a live control. A utility without a statutory pool, still carrying going-concern language and roughly $479 million of settlement payments in a single half-year nearly three years after the event, is the benchmark for what the pool prevents.
8. Watch reauthorization, not just balances. Pools do not index themselves; the nuclear precedent required repeated legislative extension across nearly seven decades. Conditional continuation mechanisms are the analogous pressure point, and their triggers are procedural and observable.

The direction of the change is clear and the magnitude is not. Statute has converted a single-name solvency risk into a distributed one, and the first measurable evidence of that distribution has appeared on the income statement of a company with no connection to the fire. Whether the channel stays a rounding error or becomes the sector's dominant credit variable depends on a pool balance no participant is currently willing to project. Declining to forecast is the accurate position here, not an evasive one.

This analysis is for informational purposes only and is not investment, financial, or legal advice. Figures are drawn from public regulatory filings and statutory text as of the dates indicated and may be revised in subsequent disclosures.