Vistra: Data-center load growth rebuilds EBITDA off a trough 2025, but the stock already prices the recovery

Stevie AI on Vistra Corp. (VST-USA | vistracorpvs)

9/20/2026

Summary

Vistra is an integrated competitive power generator and retailer with ~40 GW of nuclear, gas, coal, solar, and battery capacity split between ERCOT and PJM, plus a retail arm that naturally hedges wholesale volatility. The structural insight is straightforward: hyperscale data center demand is driving 4-6% load growth in ERCOT and 2-3% in PJM, PJM capacity prices have re-rated sharply higher, and the pending Cogentrix acquisition adds roughly $700M of incremental EBITDA at the 2027 midpoint. Management's $6.8-7.6B 2026 and $7.4-7.8B 2027 Adjusted EBITDA guidance embeds a genuine earnings rebuild after a depressed FY2025 in which hedge roll-off and mark-to-market drag crushed reported results. This is a real, identifiable recovery story, not merely a narrative. The historical numbers show just how violent the swing has been: FY2024 revenue of $19.4B and net income of $2.5B (EPS $7.00) collapsed to FY2025 revenue of $17.0B and net income of just $0.7B (EPS $2.21), a roughly 68% earnings decline driven by hedge timing and softer realized pricing rather than a structural impairment of the asset base. The forecast path assumes a steady climb back — EPS of $2.96 in FY2026, $4.64 in FY2027 (aided by Cogentrix), $5.62 in FY2028, and $6.55 in FY2029 — alongside free cash flow scaling from $1.3B to $2.5B and net debt drifting up modestly from $21.1B to $23.8B as growth capex of $3.8-4.5B/year funds Permian Peakers, the Meta PPA, Oak Hill 2 solar, and Helix commitments. Applying a 16x forward P/E — a premium to the traditional IPP/utility average given above-sector load growth and data-center optionality, but discounted from hypergrowth multiples given commodity-price sensitivity and leverage — to the EPS path yields price targets of $47 (2026), $74 (2027), $90 (2028), and $105 (2029) on a pure forward-EPS basis. However, the current price of $140.67 already reflects a market view that looks through the FY2026-2027 trough years to the FY2028-2029 EBITDA and per-share cash flow normalization, effectively pricing in a multiple closer to 21-25x FY2027 EPS or capitalizing normalized 2028-2029 earnings today. At current levels, the stock is pricing full credit for the data-center-driven demand thesis and Cogentrix accretion with limited room for ERCOT forward-curve disappointment, warranting a HOLD rather than a fresh BUY.

Thesis

1. **Structural Demand Tailwind Is Real But Already Reflected in Price** The data center-driven load growth story in ERCOT (4-6% annually) and PJM (2-3% annually) is not speculative — July 2026 saw record all-time peak demand in both markets (PJM 168+ GW, ERCOT 91+ GW), confirming the thesis is playing out in real time rather than remaining a forward promise. This is a genuine structural shift in power demand that supports higher capacity prices and utilization across Vistra's fleet. The issue is not whether the thesis is correct — it likely is — but whether the current $140.67 share price has already discounted several years of this growth. With FY2026 EPS of only $2.96 and FY2027 EPS of $4.64, the stock trades at roughly 48x and 30x those near-term years respectively, multiples that only make sense if investors are underwriting the FY2028-2029 normalized earnings power today. 2. **Integrated Retail-Generation Model Provides Genuine Earnings Stability** Vistra's vertically integrated structure — combining ~40 GW of generation with a retail electricity business — provides a natural hedge that reduces cash flow volatility relative to pure-play generators. This hybrid model helped cushion (though did not eliminate) the FY2025 earnings collapse and should support more resilient margins as the portfolio re-hedges into stronger PJM capacity pricing and Cogentrix's contracted cash flows layer in. The nuclear fleet (Comanche Peak and co-owned units, plus Martin Lake coal/gas) provides baseload stability and downside protection via the nuclear production tax credit, a feature management explicitly cites as an offset to ERCOT forward curve softness. 3. **Cogentrix and Growth Capex Program Add Real, Quantified Upside** The pending Cogentrix acquisition is expected to add ~$700M to 2027 EBITDA at the midpoint, a substantial and specific contribution that is embedded in guidance. Combined with the Permian Peakers, Meta PPA, Oak Hill 2 solar, and Helix commitments, Vistra's ~$3.8-4.5B annual growth capex program is directly targeting data-center and industrial demand rather than speculative growth. The capital allocation framework — roughly $3B to equityholders, $4.5-5B to growth, $2-2.5B of flexibility over 2026-2027 — is disciplined and specific, giving investors visibility into how >$10B of expected cash generation will be deployed. 4. **Margin Normalization Off a Depressed FY2025 Base Drives Most of the Earnings Recovery** The jump from FY2025 net income of $0.7B to FY2026's $1.0B, FY2027's $1.5B, and FY2028's $1.7B is substantially a mean-reversion story: FY2025 was depressed by hedge roll-off and mark-to-market drag, not a permanent impairment of earnings power. Reported FY2024 net income of $2.5B (EPS $7.00) is a more representative base case for normalized earnings power once hedges roll forward and PJM capacity pricing strength flows through, which the forecast trajectory (EPS $6.55 by FY2029) roughly re-approaches by the end of the forecast window. 5. **Leverage Discipline Keeps Credit Risk Contained While Funding Growth** Net debt is forecast to rise only modestly from $21.1B (FY2026) to $23.8B (FY2029) even as growth capex remains elevated, implying management is holding leverage near its stated 3.0-3.5x net debt/EBITDA target. The stated goal of mid-investment-grade ratings across all three agencies (currently achieved at two of three) suggests further deleveraging discipline that should support multiple stability, though it also constrains the pace of buybacks relative to prior years. 6. **Catalyst Path Is Clear But Binary on ERCOT Pricing** The Q3 2026 earnings call (October 2026) should provide the next major data point — updated 2026/2027 guidance and Cogentrix closing status. If ERCOT forward curves stabilize or PJM capacity strength more than offsets softness, the stock has room to re-rate toward the 2027-2028 EPS-based targets. If ERCOT weakness persists, 2027 EBITDA could land toward the $7.4B guidance floor, and the premium multiple currently embedded in the share price would compress.

Risks

1. **ERCOT Forward Curve Weakness Persisting Into 2027-2028** Current ERCOT forward prices sit materially below the October 31, 2025 baseline used to set 2027 guidance. Management's offsets — PJM strength, nuclear PTC protection, and hedging — are real but not unlimited; if ERCOT softness persists or worsens, 2027 EBITDA could land at or below the $7.4B guidance floor, and uncontracted 2028+ volumes face a genuinely uncertain pricing environment that the current premium valuation does not appear to discount. 2. **Texas Data Center Load Growth Assumptions Prove Overstated** The entire bull case rests on 4-6% ERCOT load growth materializing as forecast. Any slowdown in data center buildout — due to financing constraints, grid interconnection queue delays, chip supply issues, or a broader AI capex pause — would directly undermine the demand assumptions embedded in both management guidance and the market's implied premium multiple on VST shares. 3. **Regulatory and Queue Policy Risk in Texas** Ongoing scrutiny of the Texas data center interconnection queue and potential policy changes around large-load tariff structures or curtailment obligations could delay or reduce the pace of new demand connecting to ERCOT, directly impacting the load growth trajectory management has guided to through 2030. 4. **Execution and Integration Risk on Cogentrix and Growth Capex** The ~$700M EBITDA contribution from Cogentrix is contingent on deal closing and integration proceeding without disruption; delays or cost overruns on the $3.8-4.5B/year growth capex program (Permian Peakers, Meta PPA, Oak Hill 2 solar, Helix) could push out the earnings ramp embedded in FY2027-2029 forecasts. 5. **Elevated Leverage and Rising Net Debt Amid Capital-Intensive Growth** Net debt is forecast to rise from $21.1B to $23.8B over the forecast period even as management targets 3.0-3.5x leverage; any EBITDA shortfall relative to guidance would push leverage above target, potentially constraining the ~$3B equity return program and pressuring the credit rating upgrade path (currently investment-grade at only two of three agencies). 6. **Valuation Already Reflects a Favorable Outcome** At $140.67, the stock trades at a substantial multiple of near-term EPS ($2.96 FY2026, $4.64 FY2027), implying the market has already priced several years of successful execution on load growth, Cogentrix accretion, and PJM capacity strength. Any disappointment on these fronts — even if earnings still grow — could trigger material multiple compression given how much good news appears embedded in the current share price.

📈 Price Targets