PG&E Corporation reported third-quarter results on October 22, and the quarter reads, on first pass, as a contradiction: a company shrinking its capital ambition while insisting its earnings trajectory is unchanged. GAAP earnings fell to $0.34 a share from $0.37 a year earlier, non-GAAP core earnings fell to $0.41 from $0.50, and yet management reaffirmed full-year 2026 core guidance of $1.64 to $1.66 a share and 2027 guidance of $1.78 to $1.82. The bridge between those two facts is the story of this quarter — a $2 billion reduction to the 2027 capital plan, a 2–4% non-fuel cost-cutting target the company says it is on track to meet, and a 5,000-mile undergrounding program that is the largest single bet in its history.
The print
Start with the two EPS numbers, because PG&E’s release makes the distinction load-bearing. GAAP earnings were $749 million, or $0.34 per diluted share, against $823 million and $0.37 in the third quarter of 2025. Non-GAAP core earnings were $928 million, or $0.41 a share, against $1,142 million and $0.50 a year ago. The gap between the two — $156 million after tax, or $0.07 a share — is smaller than last year’s $295 million and $0.13, and the reconciliation shows why: amortization of the Wildfire Fund contribution accounted for $102 million pre-tax, strategic repositioning costs for $42 million after tax, and investigation remedies for $8 million.
The core decline of $0.09 a share, roughly 18%, is the number the market noticed. Core EPS of $0.41 missed a consensus clustered around $0.42 by a penny, and revenue of $6.19 billion fell about 1% from $6.25 billion a year earlier, also below consensus. Electric revenues slipped to $4.66 billion from $4.755 billion while natural gas revenues rose to $1.528 billion from $1.495 billion. Zacks attributed the revenue decline in part to roughly $140 million less in interim rate relief under the 2023 wildfire mitigation proceeding and $111 million lower revenues tied to electricity procurement costs.
The company’s own explanation for the earnings decline is more instructive than the miss itself. PG&E said GAAP earnings fell primarily because of a lower California Public Utilities Commission return on equity in effect during 2026 compared with 2025, one-time catch-up impacts from a tax methodology change implemented in 2025, and increased Wildfire Fund expense — $141 million in the quarter against $86 million a year earlier — partly offset by earnings on higher customer capital investment and net operations-and-maintenance savings. Read that list in order and the quarter stops looking like an operational stumble. Two of the three headwinds are regulatory and accounting mechanics, not service failures. The third, Wildfire Fund expense, is the recurring cost of the liability regime California built after the fire years, and it is rising.
The nine-month picture sharpens the point. Through September, GAAP earnings were $1.06 a share against $0.89 a year earlier, and core earnings were $1.24 against $1.14. The year is up; the quarter is down. A quarter in which the year-to-date trend and the quarterly trend point in opposite directions is a quarter in which the drivers matter more than the direction, and here the drivers are a lower authorized return and a bigger Wildfire Fund bill — both set by institutions other than PG&E’s operating managers.
The guidance math
Guidance is the promise that matters most, and PG&E did not move it. The 2026 range of $1.64 to $1.66 implies, by simple arithmetic against the $1.24 already banked through nine months, a fourth quarter of roughly $0.40 to $0.42 a share — essentially a repeat of the third quarter, not a heroic finish. The company said on the call that the 2026 range represents a 10% increase from 2025 at the midpoint, and that the 2027 range of $1.78 to $1.82 represents a 9% increase at its midpoint. On the guidance-credibility checklist, reaffirming after a down quarter rather than trimming is a statement of confidence; it is also a statement that the fourth quarter has very little room for a surprise in either direction.
That is worth contrasting with the pattern in the October 23 NextEra piece, where management reaffirmed guidance and pointed at the high end after a beat. PG&E is reaffirming after a miss. The two postures are not equivalent. NextEra’s reaffirmation was momentum management; PG&E’s is a defense of the range, and the market’s job now is to decide whether the defense holds through a fourth quarter that must deliver almost exactly what the third quarter just did.
The $2 billion cut
The most consequential line in the release is also the shortest: the company is “implementing previously announced $2 billion reduction to 2027 capital plan.” On the earnings call, management put the 2027 plan at $11.4 billion after the reduction, which was announced on September 2 to address high financing costs, and said rate base guidance for 2027 was unchanged despite the deferrals, because many of the affected projects have construction timelines extending into 2028 or later.
This is where the earnings-report checklist’s capex section does its work. Capex today becomes rate base tomorrow, which becomes earnings the day after — so a utility cutting its capital plan is, in the ordinary telling, cutting its future earnings base. PG&E’s answer is that the cut is a timing shift, not a scope reduction: projects deferred, not canceled, with rate base guidance intact because the spending lands later. That answer deserves the skepticism the checklist prescribes for any management claim about deferred spending. A deferral that preserves rate base guidance is only costless if the deferred projects return on schedule and at the budgeted cost, and if the commission accepts the revised timeline without reopening the underlying need. Neither is guaranteed.
The reason for the cut matters as much as its size. PG&E framed it as a response to financing costs, and the release pairs it with a strategic review “advancing with the goal of becoming a financially strong, investment-grade company.” Read together, those two lines describe a company choosing balance-sheet repair over growth speed. That is the opposite of the posture in the NextEra quarter, where regulated and transmission capital is up roughly 25% since December 2025 and the company is advancing more than 20 gigawatts of large-load projects. One company is accelerating into the load-growth era; the other is slowing its spending to protect its credit standing. Both can be rational. Only one of them can claim to be doing both.
The cost-discipline line is the third leg of the same stool. PG&E says it is on track for a 2–4% reduction in non-fuel O&M, and management said on the call that those savings contributed about $0.04 a share through the first nine months. Under the regulatory-lag mechanism the checklist describes — run the business cheaper than the rate case forecast, keep the savings until rates reset — O&M discipline is the most direct lever a utility’s managers control. PG&E is pulling it hard, and the nine-month numbers suggest it is working. The question is whether O&M savings can carry a guidance range while the authorized return is lower and the Wildfire Fund bill is higher. So far, the arithmetic says yes, barely.
The undergrounding bet
Against that backdrop of restraint, the quarter’s biggest number points in the opposite direction. PG&E filed a 5,000-mile undergrounding plan with California regulators, proposing $16.6 billion of spending between 2028 and 2037. The company claims the program would reduce wildfire risk by 98% and outages by 90% on the lines moved underground, and deliver an estimated $117 billion in long-term benefits. Treat every one of those figures as a company claim, not a fact: the 98% and 90% are engineering estimates for the specific lines selected, and the $117 billion is a benefits model whose assumptions — avoided fire costs, avoided outage costs, discount rates — the filing, not the press release, will have to defend.
The tension is the point. A company cutting $2 billion from next year’s capital plan is simultaneously proposing a $16.6 billion decade-long program. The two are not contradictory if the cut is about financing costs today and the program is about risk reduction tomorrow — but they are only compatible if the commission approves the program on terms that let PG&E finance it without returning to the capital markets at the costs that forced the cut in the first place. Undergrounding is the most capital-intensive form of hardening, and its economics depend entirely on the recovery mechanism: how much enters rate base, how fast, and at what return. The filing is the opening of that negotiation, not its conclusion.
Chief executive Patti Poppe’s prepared remark points at the other half of the financing question. “While we still believe California needs a durable wildfire liability framework that delivers for customers and investors, safety remains our foundation,” she said. That sentence is doing two jobs: it endorses the undergrounding program as safety work, and it reminds Sacramento that the liability regime — the rules that decide who pays when a fire starts — is still, in the company’s view, unfinished. A $16.6 billion hardening program does not solve a liability problem; it reduces the probability of the event the liability regime prices. Investors should not confuse the two.
The operational texture of the quarter supports the safety narrative without overstating it. Management said on the call that the weather-normalized ignition rate is down 74% since 2017, marking a fourth consecutive year without a major equipment-related wildfire. The release also notes more than 4,240 electric customers and 3,000 electric-vehicle charging ports connected in the quarter, a virtual power plant demonstration with Google and Rewiring America targeting more than 20,000 customer devices and 12 megawatts of distributed resources by 2028, and nearly $15 million in bill assistance to more than 46,000 low- and moderate-income customers. None of these moves the EPS line. All of them are the kind of record-building a utility does when its next big ask is a commission decision.
The load-growth contrast
The earnings-preview piece from October 21 framed September’s record load as the receipt for the sector’s load-growth thesis: the demand showed up, and the question was which utilities could serve it. PG&E’s answer is modest by that standard. The company has energized 174 megawatts of new data center projects since 2024, with a multiyear pipeline targeting 1.8 gigawatts online by 2030, according to the call. The release adds the company’s claim that, with appropriate pricing, every gigawatt of new data center load could help customers save 1% or more on their monthly electric bill. That “with appropriate pricing” qualifier is doing real work — it is the difference between load growth that subsidizes existing customers and load growth that burdens them, and it is exactly the cost-allocation fight now running through Virginia’s merger docket.
Set against the October 23 NextEra piece, the contrast is the column’s cleanest illustration of two different utility strategies in the same week. NextEra beat, pointed at the high end, and grew its backlog to roughly 36.7 gigawatts. PG&E missed by a penny, held its range, and cut its capital plan. NextEra’s problem is political — whether Virginia lets it buy Dominion, and on what terms. PG&E’s problem is financial — whether it can fund the safety program California demands at a cost its balance sheet can carry. The sector’s load-growth story is common to both; the constraints are not.
One more call detail belongs in the financing file. Management proposed $456 million in interim rate recovery for collection starting in January 2027, aimed at stabilizing customer bills and aligning cash flow with costs. Interim recovery is a cash-flow tool, not an earnings event, but for a company managing financing costs it matters: cash collected sooner is debt not issued.
What to watch
Four markers, in order of consequence. First, the undergrounding filing’s reception at the CPUC — whether intervenors accept the company’s benefit claims or force a smaller, slower program, and what return the commission allows on the spending it does approve. Second, the strategic review’s outcome: the release says it is advancing, not what it will conclude, and any structural change — asset sales, a reorganization, a financing plan — would reframe the entire investment case. Third, the January 2027 interim recovery proposal, which tests whether the commission will help PG&E’s cash flow while the capital plan shrinks. Fourth, the fourth-quarter print itself, which must land in a narrow $0.40-to-$0.42 window to validate a guidance range the company chose to defend rather than move.
PG&E’s quarter is best read as a financing statement disguised as an earnings report. Earnings fell for reasons largely outside management’s control; guidance held because of levers inside it — cost discipline, deferred spending, and a bet that a decade of undergrounding, approved on acceptable terms, will do what the liability regime has not: make California’s largest utility financeable on the strength of its own safety record. Whether the commission prices that bet the way the company models it is a question no quarterly release can answer.
None of this is investment advice, and nothing in this analysis recommends buying or selling any security. It is a reading of one quarter’s evidence — a smaller ambition, a held promise, and the largest safety bet this utility has ever proposed.