Case studies › VST
Vistra's call put cash flow and demand risk in the same paragraph
Vistra used a third-quarter call to raise the profit outlook, guide 2027, and put a per-share cash flow target next to a list of things that could stop it. Both halves are on the record and both belong in the same note.
Third-quarter earnings were US$3.23 per share against a US$2.94 consensus. Revenue was US$6.35 billion against US$7.11 billion expected. Earnings beat by 10% while revenue missed by 11%.
The guidance, as given
| Metric | 2026 | 2027 |
|---|---|---|
| Adjusted EBITDA | $7.1–7.5bn | $7.7–8.5bn |
| Adjusted FCF before growth | $4.3–4.7bn | $4.3–5.1bn |
| FCF before growth per share | — | ~$14 |
| Share repurchases, modelled | — | ~$3bn/yr |
Record quarterly adjusted EBITDA of US$2.378 billion was attributed to fleet performance and commercial execution in unusually hot weather. The 2026 EBITDA range was narrowed to a higher midpoint, and the 2027 range was introduced including Cogentrix and the Meta agreement at Perry.
Per-share free cash flow of about US$14 in 2027 against just under US$10.50 in 2025, and roughly 70% further growth to 2030. The mechanism is disclosed: about US$3 billion a year of repurchases, of which roughly US$2 billion is additional available cash allocated for modelling purposes.
Per-share cash flow is a share count story
US$14 per share is not a forecast of US$14 of cash per share of the business. It is that cash divided by a count that is being shrunk by US$3 billion a year. The distinction matters because the numerator is guided in a range of US$4.3 to US$5.1 billion, which is a 19% width, while the denominator is a policy.
Divide the low end of the 2027 range by the low end of the implied count and divide the high end by the high end, and most of the spread disappears into the repurchase line. Anything that raises the share price mechanically raises the cash needed for the same count reduction, so the buyback is most effective when the stock is cheapest, which is the opposite of how it behaves in a re-rating.
Cogentrix, and what it cost
The October close added roughly 5,500 megawatts of gas-fired capacity and took the portfolio to about 50,000 megawatts. An 11% capacity increase, with the earnings forecast improving after announcement.
Gas capacity is the part of the fleet that is priced off the spark spread, and the 2027 guide takes the fuel curve as it stands rather than a recovery in Texas power prices. That is a conservative assumption on price and it is the right one. It also means the EBITDA range is levered to a curve the company does not control.
Demand risk, in the company's own words
Slower data-centre connections in Texas. Regulatory uncertainty. Both were addressed on the call and neither is a small matter: data centres are the incremental load the 2030 growth path depends on, and permitting is the gate in front of it.
A merchant generator with a 70% cash flow growth target to 2030 needs somewhere for the electrons to go. If the interconnection queue slows, capacity additions still happen while the demand they were bought for arrives later. On the spread between cost of capital and realised price, that is the whole risk.
Where the case breaks
The revenue miss of 11% alongside an earnings beat is the first crack and it is the one to track. Revenue at US$6.35 billion against US$7.11 billion expected means less energy sold or less sold at the expected price; the beat came from cost and weather, not from volume.
If Texas connections keep slipping, the 2030 path compresses toward the buyback alone, and a buyback-only growth story is a low multiple. The regulatory question sits on top of that.
What the arithmetic says
US$14 per share of 2027 cash flow, built from US$4.3 to US$5.1 billion of free cash flow, a US$3 billion repurchase, a 5,500 megawatt acquisition, a flat Texas price assumption, and two named risks in the demand line. Earnings beat, revenue missed.
A 19% wide cash flow range converted into a single per-share number, held together by US$3 billion of annual repurchase. The cash flow is guided. The denominator is policy.
Sources
Earnings, revenue, guidance ranges, capacity, acquisition size, repurchase assumptions and stated assumptions are the company's own third-quarter 2026 results and call disclosures. Consensus figures are third-party. Per-share derivation and the width comparison are arithmetic on the stated ranges and can be re-run.
Published for information only. Nothing here is investment advice or a recommendation. This note is not prepared in accordance with the legal requirements designed to promote the independence of investment research. No position is held in the security covered.