Vistra (VST) Q2 2026 Earnings Call Transcript

Vistra reported Q2 2026 Adjusted EBITDA of $1.767B (+30% YoY) and raised confidence in its outlook, reaffirming 2026 Adjusted EBITDA guidance of $6.8B–$7.6B and Adjusted FCFbG of $3.925B–$4.725B (targeting at/above the midpoint). The company highlighted ~$994M generation Adjusted EBITDA (+$401M YoY) on ~5% higher realized prices per MWh and continued strength in retail ($773M). Capital allocation remains aggressive: ~$6.5B returned to shareholders via repurchases since Nov 2021, retiring ~171M shares at ~$38/share, with $1.2B of authorization remaining to be exhausted by end-2027. Despite ERCOT forward curves being lower and Texas interconnection audit pauses near-term, management sees structurally improved load growth (ERCOT +4% to +6% annually through 2030) and a 2027 midpoint Adjusted EBITDA opportunity of $7.4B–$7.8B (excluding Cogentrix and Meta PPA contributions).

Analysis

VST’s equity setup is still driven more by scarcity economics than by the reported quarter. The important mechanism is that the company owns the kind of dispatchable, interconnect-ready capacity that becomes more valuable when new load is real but new supply is delayed; that is why co-location and “existing asset” contracting matter more than headline load forecasts. The second-order winner is not just VST but any owner of behind-the-fence or fast-to-serve generation; the losers are greenfield developers and balance-sheet-light data-center platforms that need cheap land/power shells before they can monetize demand.

Near term, the market can overreact to softness in ERCOT forwards and audit noise, but that is mostly a timing issue unless it starts to change actual 2027–2028 contract coverage. The real falsifier is not rhetoric from policymakers; it is either a sustained rollover in ERCOT strip prices or a visible stall in signed data-center PPAs over the next 1–3 quarters. Conversely, if PJM co-location rules are clarified in the next 30–60 days and Helix starts surfacing credible pipeline, the stock can rerate on optionality, because the market is likely underpricing how much an integrated power vendor can compress customer procurement complexity.

Contrarian view: consensus may be too focused on “AI power demand” as a binary and too little on the selection effect. The best customers will pay for speed, reliability, and optionality; the weakest will push projects out, which actually improves economics for incumbents with finite interconnects and existing thermal fleets. That means VST’s cash return story is not being displaced by Helix—it is being reinforced by it, provided capital intensity stays disciplined and the company does not overpay for growth in a market where curves can mean-revert quickly.

Key Decisions for Investors

• Long VST vs short XLU for 1-3 months: express the view that regulated utilities underwrite stability while VST monetizes scarcity, data-center optionality, and buybacks. Favors VST if PJM co-location guidance lands cleanly and ERCOT demand remains tight.
• Buy VST on a pullback toward pre-earnings support; use the next 2-6 weeks to accumulate only if the stock de-risks on ERCOT audit headlines rather than on deteriorating contract activity. Falsify if 2027 guidance trends toward the low end without offsetting signed PPAs.
• For a more tactical trade, own VST call spreads into the expected 30-60 day regulatory filing window in PJM. The payoff is rerating on reduced uncertainty; risk is capped if the filing is delayed or dilutive to economics.
• Avoid chasing Helix-related names until there is visible third-party capital formation or disclosed projects. KKR has strategic optionality, but the financial impact on VST is still too early to underwrite beyond sentiment.
• Watch CEG and NRG as read-through beneficiaries/competitors: if VST’s thesis is right, merchant generation and retail-integrated peers should also benefit from higher value of existing capacity. A relative long VST / short XLU remains preferable to an outright beta bet.