Expand Energy Corporation EXE

Key data

The business

Expand Energy is the largest natural gas producer in the United States, formed in October 2024 when Chesapeake bought Southwestern. It pumps about 7.5 billion cubic feet of gas equivalent a day from two places: the Haynesville shale in Louisiana, next door to the Gulf Coast LNG export docks, and the Marcellus and Utica in Appalachia. It sells almost entirely gas, with a sliver of oil and natural gas liquids. The economics are simple and brutal: revenue is production times a price the company does not set, so the only levers management holds are how cheaply it drills and how little debt it carries. Its edge is scale and low-cost acreage, and sitting beside the LNG terminals matters because that is where the new demand is landing. The product is a molecule, identical to a rival's, which is why cost per unit is the entire game.

The numbers

Read the quarters as a sequence, not a level, because a gas producer's headline earnings swing on the weather and on paper derivative marks. Q2 2025's $4.02 of GAAP EPS was mostly an $825M unrealized derivative gain, not cash.

The winter quarter, Q1 2026, was the peak: $4.40B of revenue as cold weather lifted prices. Then spring came, realized gas fell to $2.42 per Mcf against a $2.90 benchmark, and the just-reported Q2 2026 landed at $1.33 of adjusted EPS, ahead of the $1.13 the Street looked for and the fourth straight beat, on revenue of approximately $1.8B. Production held at about 7.5 Bcfe a day. The point is that the same wells earn triple in January and a third of that in April, so cash across a full year, not any single quarter, is what to judge.

FY2024 is the trough: a gas price crash on top of a merger that closed in October, so it holds barely a quarter of the combined company. FY2025 is the first clean year of the merged entity, and it turned $12.1B of revenue into $1.8B of profit and $1.84B of free cash. What matters more than profit is that the business self-funds with room to spare.

The discipline has held: H1 2026 reinvestment ran a lighter 42%, free cash rose to $2.0B, while shares moved from about $99 to $98. The balance sheet is the whole thesis's backstop, net debt at 0.45x EBITDA, interest covered 16 times, $4.2B of liquidity. The variant view is that structural LNG and data-center power demand raise the floor under gas prices through 2028, and that at 3.9x EBITDA the market is still pricing gas as permanently cheap; the two prints that settle it are realized price and free cash flow over the next year.

Management

Insiders bought and never sold: seven open-market purchases worth $0.9M over the year, director Marcel Teunissen adding about $379K across May and June, Michael Wichterich $108K in March. The board doubled the buyback authorization to $2.0B in July and pays a $0.575 base-plus-variable quarterly dividend, and since FY2025 repurchases were only $100M, that firepower is new. The record shows four consecutive adjusted-EPS beats. One flag: cash fell from $2.2B in March to $0.7B in June, and a late-July 8-K discloses a pending acquisition, so capital is being committed just as the buyback expands. Plan status is not disclosed on the purchases, but director buys with zero sales is the cleaner read.

How it fails or surprises you

The price floor. Q2 2026's $2.42 realized gas is the fact this read explains least well: cash still came at that price, but a $2.00 strip through a warm winter would erase the cushion. Watch realized price over the next two quarters against the $2.90 benchmark; sustained below $2.50 and the buyback, not the drilling, gets cut first.

LNG and power demand (right tail). US LNG export capacity is set to roughly double into 2028 as Gulf Coast trains start up, with data-center load a second new draw. If gas settles structurally above $3.50, Expand's 7.5 Bcfe a day turns a $1.8B free-cash year into something far larger and the 3.9x multiple re-rates. The tell is realized price and 2027 guidance.

The acquisition. The July 8-K commits cash toward a purchase, draining balance to $0.7B. Bought well, low-cost acreage near the LNG docks deepens the moat; bought at a cyclical high on borrowed money, it echoes the leverage that bankrupted the old Chesapeake in 2020. Watch the price paid and the debt taken on.

Closing thoughts

This is a commodity exposure with a fortress bolted to it. You cannot know next year's gas price and neither can the market, so the edge is not a forecast, it is that Expand sits at the low end of the cost curve carrying almost no debt, which lets it outlast the cheap-gas years that sink weaker producers and compound through the rich ones. The left tail, gas stuck near $2.40 through a mild winter, is real but survivable at 0.45x leverage, so what is at risk there is the pace of buybacks, not the company. The right tail, the LNG and power demand pull, looks the fatter one over three years and is barely paid for at under 4x EBITDA.

The bet is still that America keeps burning more natural gas, for power at home and to ship overseas as LNG, and that Expand sells enough of it to make money. It breaks if gas sits near $2.40 through a warm winter while the new acquisition piles on debt, and the pair that tells you first is realized gas price against the $2.90 benchmark and net debt climbing above 1x EBITDA from today's 0.45x.