California Let a Utility Wildfire Liability Cap Die Without a Vote, and the Cost Did Not Go Anywhere

The California Legislature adjourned its session without advancing Senate Bill 492, Governor Gavin Newsom’s proposal to cap utility wildfire liability. The equity market priced the result within a day. PG&E stock dropped 20 percent. Edison International fell 23 percent, its largest single-day decline in more than 25 years.

A 23 percent single-session repricing of a regulated utility is not a statement about the current quarter’s earnings. It is a statement about who carries a liability that nobody disputes exists.

What the Legislature did and did not do. SB 492 would have capped utility wildfire liability. It did not advance before adjournment, which is a different outcome from a floor defeat: the bill was never voted down, it simply ran out of session. The practical effect is the same for 2026, and any successor measure starts over as new legislation.

The same session was not idle on energy. It sent Newsom SB 1168, which directs the California Public Utilities Commission to assess rate structures ensuring data centers pay a reasonable share of transmission and distribution costs. It also passed SB 913, on aggregated distributed energy resources counting toward resource adequacy; AB 2493, requiring independent audits of interconnection submissions and network upgrades; and AB 1738, on remote inspection, residential only.

Liability that is not capped is liability that is recovered. This is the part of the September repricing that deserves more attention than the share-price headline, and it is an inference about mechanism rather than a filed number. Wildfire exposure that the Legislature declined to socialize does not leave the system. It stays with the companies, and regulated companies recover what regulators permit them to recover: through rate base, through memorandum accounts, through the next general rate case. What the equity market repriced on the day of adjournment was the residual risk sitting above whatever the commission ultimately allows.

Two channels carry that forward into rates. The first is the authorized cost of capital. An equity decline of 20 to 23 percent raises the return both utilities will argue they need in their next general rate cases, and authorized returns apply to the value of approved long-term infrastructure. Wildfire hardening is the least discretionary category in that spend, which means the base to which any authorized return is applied keeps growing.

The second channel is the cost of debt, which reprices faster than any commission proceeding. If lenders now price California wildfire liability as a shareholder risk rather than a socialized one, borrowing costs move before a commissioner votes on anything. No California utility has yet put a post-SB 492 financing-cost adjustment in front of the commission. Whether one arrives is the near-term test of whether the September move was sentiment or repricing.

Where the recovery lands. In California, non-residential distribution revenue is collected substantially through dollars per kilowatt of measured peak demand. San Diego Gas and Electric already reaches $68.80 per kilowatt on its medium commercial summer tariff. Rate trackers put the September 2026 US average commercial retail rate at 14.19 cents per kilowatt-hour against 18.34 cents residential, with demand charges, billed on a customer’s peak 15 or 30 minute interval, accounting for 30 to 50 percent of a typical commercial bill.

That split determines which regulatory events actually reach a building owner. A one-day move in Edison International’s share price is not a billing determinant. The level of the kilowatt charge in a commercial tariff is, and it sits downstream of a revenue requirement that grows with mandated wildfire capital.

SB 1168 reopens the allocation question. A directive to assess rate structures ensuring data centers pay a reasonable share of transmission and distribution costs is, mechanically, an instruction to reopen non-residential cost allocation. A commission cannot decide what share of T&D costs a large load carries without implicitly deciding what share every other non-residential class carries. Demand-charge design is on the table whenever that question is open.

The timing places two pressures on the same set of tariffs. On one side, a revenue requirement growing with uncapped wildfire liability and the capital spending meant to reduce it. On the other, a legislative instruction to re-examine how non-residential customers are charged for the distribution system. Commissions frequently respond to affordability pressure by trimming revenue requirements, but the demand component is the piece least likely to be reduced, because it is the piece utilities are actively trying to make more cost-reflective.

What this means for storage economics. Storage underwritten on demand-charge avoidance earns its return from a metered tariff line rather than from a program budget or an incentive cycle. That is the structural difference between it and projects whose returns depend on which way a legislature votes. California’s September session moved the political line and left the tariff line pointed upward, and only one of the two clears through a customer’s meter every month. Payback models built on a flat demand charge are modeling the less likely case.

Three things are checkable from here. Whether any California utility files a cost-of-capital or financing-cost adjustment citing post-September credit conditions. Whether the CPUC’s response to SB 1168 opens into a general non-residential rate design phase or stays confined to a large-load tariff. And whether a 2027 liability bill arrives with a cap in it, which would reverse the pricing the market applied at adjournment and unwind part of the pressure now pointed at the rate base.

Sources