PG&E Corporation PCG

Key data

The business

PG&E is described in company materials as the largest investor-owned utility in the United States by customers, delivering electricity and natural gas to about 16 million people across northern and central California. It is a legal monopoly: no one else owns the wires and pipes in its territory, and in exchange the state regulator, the CPUC, sets the rates. The economics are simple. PG&E invests in poles, wires, substations and pipelines, that pile of capital is the rate base, and the regulator lets it earn a set return on the base. Spend more, earn more. So the company is spending furiously, with capital deployment described in recent commentary as targeting roughly $12 billion for 2026, much of it burying thousands of miles of power line so a falling wire cannot spark the next fire. Second-quarter revenue was flat at $5.9 billion, but the first half rose 8% to $12.8 billion as electric rates stepped up. The thing customers actually feel is the bill, and it keeps rising.

The numbers

Two things matter in the sequence below: earnings that grind higher every year, and a cash flow statement that cannot fund the growth by itself.

¹ Derived as FY2025 annual less filed Q1-Q3 quarters

Revenue is seasonal and looks flat, second-quarter 2026 landed exactly where it was a year earlier, because summer is the low season and much of the top line is pass-through fuel cost that never touches profit. Watch operating income and EPS instead: both step up through the year, and the first quarter is now the big one. On the core number the company steers to, PG&E has beaten expectations every quarter for the past year, most recently $0.40 against $0.36.

² H1 2026: $0.39 (Q1) + $0.33 (Q2)

Zoom out and the story is a steady climb off the 2019 bankruptcy floor. Diluted EPS went from a penny of loss in 2021 to $1.18 in 2025, and core earnings have compounded about 12% a year since 2022. Management has communicated guidance for earnings growth of at least 10% annually through 2028, and for a regulated monopoly that guidance is unusually credible, because the growth is mechanical: rate base times allowed return. The catch is the next table.

PG&E spent $11.8 billion on its system last year and generated $8.72 billion in operating cash. The $3 billion gap, every year, is filled with borrowing and new stock. Long-term debt has climbed to $57.4 billion, net debt is 6.05 times EBITDA, and interest eats most of operating profit, coverage is only 1.67 times. That is the whole argument in one line: the market pays 10 times earnings, half the peer multiple, not because it doubts the earnings growth but because it is unsure the balance sheet and the fire seasons let PG&E collect it. Priced correctly is a real possibility. My read is that the discount runs a touch too wide, and the print that settles it is a peak fire season, July to October, that passes without a PG&E-caused catastrophe.

Management

The record since bankruptcy is better than the share price suggests. Company leadership has hit or raised core EPS guidance every year since 2021 and delivered the safety milestones tied to executive compensation. Insiders, though, are net sellers: twelve sales worth $8.26 million against a single $119,700 buy over the past year, led by utility CEO Marlene Santos unloading $2.85 million the day before the July earnings release, plan status not disclosed. The common dividend, wiped out in the 2019 bankruptcy, has only been restarted at a token rate, so shareholders are paid almost nothing to wait. And PG&E is issuing stock, not retiring it, so every share of your ownership is slowly diluted to help fund the poles and wires.

How it fails or surprises you

A fire it caused. CAL FIRE naming PG&E equipment as the cause of a catastrophic fire in the July-to-October dry season. The state's wildfire fund (dimensioned in company disclosures at roughly $21 billion) and the safety certification cap some liability, but a Camp Fire-scale event could exceed coverage and reopen the solvency question that ended in Chapter 11 last time. The print: a CAL FIRE cause determination.

Paying for growth with paper. Operating cash flow was $8.72 billion in 2025 against $11.8 billion of capex, a $3 billion hole filled with debt and new shares. Net debt is already 6.05 times EBITDA and interest covers only 1.67 times. Each equity raise near 1.1 times book dilutes the per-share growth the whole thesis rests on. Print: the size of the next financing update.

The discount closes (right tail). At 10 times earnings and 1.1 times book against peers near 2.2 times, a re-rate needs only ordinary competence: clean fire seasons plus Silicon Valley data center load pushing rate base above the current plan. Two or three quiet summers and the multiple drifts toward peers, worth 50% or more with no change in earnings. Print: an upward rate base revision.

Closing thoughts

Nothing in the next four quarters settles this. What matters is that PG&E gets through the peak fire season, keeps its credit rating above junk, and funds the building plan without a punishing equity raise. The left tail is genuinely fatal and genuinely possible: one catastrophic fire it caused. The right tail is a slow re-rate that pays 50% or more if the summers stay quiet. The downside is a permanent loss, the upside is patient, and that asymmetry only works if the balance sheet survives the wait, which is why the credit rating and the net-debt line matter more than any quarter's EPS.

The bet is still that PG&E keeps burying power lines and growing what it earns a return on, that California keeps paying for it, and that it gets through the dry summers without starting a fire. What breaks it is a fire it caused. The two numbers that tell you first are net debt to EBITDA, now 6.05 times, and the utility's credit rating: a downgrade below investment grade means the market has stopped believing the fund and the undergrounding will hold.