Valuation -- Concerns/Risks 7/10
| Company | Fwd EV/EBITDA 2026E | Fwd EV/EBITDA 2027E | Fwd P/E 2026E | FCF Yield 2026E | Net Debt / EBITDA | Investment Grade |
|---|---|---|---|---|---|---|
| Vistra (VST) | ~9-10x | ~8-9x | ~17-18x | ~7-8% | 2.6x / 2.3x YE27 | YES (x2) |
| Constellation (CEG) | ~13x | ~12x | ~25-26x | ~3.8% | ~1.5x | YES |
| Talen (TLN, PF) | ~10-11x | ~9-10x | ~25-28x | ~6-9% | <3.5x | NO (BB) |
| NRG Energy (NRG) | ~10x | ~9x | ~17x | ~6-7% | ~3.5x | NO (BB+) |
| AES Corp (AES) | ~7-8x | ~7x | ~9x | ~10%+ | ~5.5x | BBB- |
| Key Takeaway | VST trades at a ~3-4 turn discount to CEG on EV/EBITDA and ~8 turns on P/E despite signing the two largest hyperscaler nuclear PPAs in US history, achieving investment-grade ratings at two agencies, and running a more diversified merchant + retail (TXU) book. FCF yield ~7-8% is competitive across the cohort. The discount looks unjustified on commercial execution; it is justified only if you assign meaningful weight to coal/gas mix overhang and AI-capex narrative risk. | |||||
| Metric | 2024A | 2025A | 2026E (Guide) | Drivers |
|---|---|---|---|---|
| Adjusted EBITDA | ~$5.66B | $5.84B (core) / $7.20B (inc. all items) | $6.80B - $7.60B | ERCOT scarcity + PJM capacity prints + Energy Harbor full-year + Meta PPA initial |
| Adj. FCF before Growth | ~$2.65B | $3.50B | ~$3.5-4.0B | ~7-8% FCF yield on $50B mkt cap; supports $2.5B buyback + IG metrics |
| Cogentrix Contribution | -- | -- | +~$500M run-rate | 5.5 GW gas, $4B EV / 7.25x 2027E; Q3 2026 close, partial-year 2026 |
| 2027 Midpoint Opportunity | -- | -- | $7.4 - $7.8B EBITDA | Cogentrix full year + Meta ramp + Amazon Comanche Peak commencement |
EV/EBITDA and FCF yield are the right primary metrics for an IPP like VST. P/E is distorted by mark-to-market hedge accounting, large depreciation on the fossil base, deferred tax timing, and Moss Landing one-time write-offs -- which is why headline 2026E P/E of ~17-18x is less useful than EV/EBITDA for cross-IPP comparison.
- EV/EBITDA captures the underlying generation economics across nuclear (PTC floor at $43.75/MWh), gas (capacity + energy), coal (depreciated stranded base), and retail (TXU). At ~9-10x 2026E and ~8-9x 2027E, VST sits ~3-4 turns below CEG and at parity with TLN pro-forma -- despite a stronger commercial slate and IG balance sheet.
- FCF yield (~7-8% 2026E) is the cleanest cash-on-cash measure. Adj. FCF before growth of $3.5B in 2025 funds a $2.5B buyback authorization (~5% of market cap) and maintains the 2.3x YE27 leverage target.
- Net Debt / EBITDA validates the IG ratings: 2.6x current trending to 2.3x YE27 even after Cogentrix close, well below TLN (<3.5x), NRG (~3.5x), and AES (~5.5x).
The stock has round-tripped from a ~$219 peak to ~$147, compressing forward EV/EBITDA from the low-teens to high-single-digits on 2027 midpoint. The sell-side PT distribution has not followed the multiple down -- mean PT ~$233 implies ~59% upside, the widest PT-to-spot gap across the IPP cohort. This is a setup where either the price is wrong or the published targets are wrong; the asymmetry of fundamentals (signed PPAs, IG ratings, Cogentrix accretion, $2.5B buyback) suggests the spot is closer to the dislocation.
| # | Catalyst | Impact | Timeframe |
|---|---|---|---|
| 1 | Next hyperscaler PPA -- ~3.2 GW remaining uncontracted nuclear (Beaver Valley + Comanche Peak) plus uprate optionality | Each 1 GW PPA = mid-single-digit FCF accretion; re-rates fleet to contracted-asset multiple | 2H 2026 -- 2027 |
| 2 | Cogentrix close + integration -- 5.5 GW gas, $4B/$4.7B EV at 7.25x 2027E | ~$500M run-rate EBITDA; immediately accretive; broadens gas exposure ahead of PJM tightness | Q3 2026 expected close |
| 3 | ERCOT PCM shelved -- PUCT voted Dec 2024 to drop Performance Credit Mechanism | Removes regulatory overhang; market design now focused on real-time co-optimization + DRRS, friendlier to existing dispatchable fleet | Already happened |
| 4 | PJM capacity auction prints -- last two BRAs cleared at cap ($329.17 and $333.44/MW-day) | Cap removal would be a clear positive given expanded PJM footprint post Energy Harbor + Cogentrix | Mid-2026 next auction |
| 5 | Brownfield gas / Permian expansion -- 860 MW Permian gas plant + 4,500 MW pipeline | Supports ERCOT load growth; gas brownfield economics support attractive contracted returns | 2027 -- 2029 |
| 6 | Buyback execution -- $525M repurchased in the first four months of 2026; $2.5B authorized through YE27 | ~5% of market cap remaining; accretive at depressed prices | Ongoing |
| 7 | Meta PPA ramp + Comanche Peak SLR/uprate -- 2,176 MW operating + 433 MW uprate; SLR + ~200 MW uprate to backfill AWS PPA | 8-10% incremental FCF from operating ramp; 5-7% from uprate at full ramp by 2034 | Late 2026 onward; 2027-2030 |
| Risk | Severity | Probability | Detail |
|---|---|---|---|
| Moss Landing residual liabilities | MEDIUM | MEDIUM | ~$400M write-off taken Q1 2025; multiple lawsuits ongoing; remediation costs not fully bounded; reputational risk on battery pipeline. Manageable, not existential. |
| Coal retirement timing + stranded costs | MEDIUM | MED-HIGH | Lignite/coal in Texas + Midwest on retirement glide path. Martin Lake, Coleto Creek have life-extension optionality but face EPA pressure. Stranded-cost risk modest given depreciated book. |
| FERC PJM behind-the-meter overhang | MED-HIGH | MEDIUM | Meta-Vistra structured front-of-meter with grid contributions (license renewals already secured) -- less acute than TLN-AWS BTM exposure. Sector-wide FERC rule-making remains an overhang. |
| Power price / gas normalization | HIGH | MEDIUM | Merchant fossil fleet hedged 90%+ 2026, ~70% 2027; outer years exposed. Sustained low gas + mild weather + DC demand under-delivery compresses sparks. |
| Multiple compression (AI capex narrative) | MEDIUM | MEDIUM | Already 33% off highs -- much AI froth removed. ~9-10x still above pre-2024 IPP norms (6-8x) but well below CEG. Another 2 turns compression = 20-25% downside. |
| ERCOT regulatory (post-PCM) | LOW-MED | LOW | PCM shelved is a tailwind. New designs (real-time co-optimization, DRRS) net positive-to-neutral for existing dispatchable fleet. Co-located load rules at ERCOT/PUCT are still evolving. |
| Scenario | Probability | 2027E EBITDA | EV/EBITDA | Fair Value Range | Drivers |
|---|---|---|---|---|---|
| Bull | 30% | $8.5-9.0B | 10-11x | $200-$240 | Third hyperscaler PPA + Cogentrix accretes on schedule + PJM cap removal |
| Base | 50% | $7.6-8.0B | 9-10x | $155-$185 | Cogentrix closes on time; Meta/Amazon ramp on schedule; no incremental PPA |
| Bear | 20% | $7.0-7.4B | 7-8x | $105-$135 | FERC BTM tightens; AI capex pauses; gas softens; Moss Landing legal escalates |
| Probability-Weighted Fair Value | ~$160-$180 | Spot $147 sits below base case; bull upside $50-90 vs bear downside $15-40 -- asymmetric upside | |||
VST is the only IPP that combines (1) signed hyperscaler PPAs across both ERCOT and PJM, (2) a diversified merchant fleet that captures both ERCOT load growth and PJM capacity tightening, (3) an investment-grade balance sheet at 2.3x target leverage, (4) a competitive retail business (TXU) that monetizes its own generation, and (5) a depressed entry multiple after a 33% pullback.
The Cogentrix deal at 7.25x 2027 EBITDA was disciplined, not desperate. Two more nuclear PPAs likely sign within 18 months. Buyback flywheel + EBITDA growth = ~12-15% annual FCF/share compounding through 2028.
You are buying a merchant generator with a 30%+ coal-and-gas fossil revenue mix at ~9-10x EBITDA -- still 50% above pre-2023 IPP norms. The two hyperscaler PPAs you are paying for do not ramp meaningfully until 2027-2032; in the meantime cash flows depend on ERCOT/PJM power prices and capacity markets that are politically vulnerable in election cycles.
Moss Landing reminded everyone that battery + nuclear assets carry tail-risk exposure that does not show up in EBITDA models. If AI capex pauses for two quarters, the multiple compresses to 7-8x and the stock goes to $110-$120.
Score of 7/10 on the Concerns/Risks dimension reflects a favorable risk-reward tilt: meaningful asymmetric upside on probability-weighted fair value, IG balance sheet, and a regulatory backdrop (ERCOT PCM dead) that is a net tailwind -- offset by genuine fossil-mix and FERC-sector tail risks that cap the score below 8+.
Why not higher (8-9): Moss Landing residual liabilities are not fully bounded (remediation + ongoing litigation). Coal retirement timing carries political-cycle risk. The PJM behind-the-meter FERC rule-making remains a sector-wide overhang even though the Meta deal is structured front-of-meter. ~30% fossil revenue mix caps multiple expansion below CEG until the fleet transitions further toward contracted carbon-free generation. The AI demand thesis is still the bull case; if AI capex narrative cracks, the multiple compresses 2 more turns.
Why not lower (5-6): Stock is already 33% off the high -- much of the AI froth has been removed. The two hyperscaler PPAs and Cogentrix transaction are signed and in the price -- execution risk only from here. IG ratings at two agencies remove balance-sheet tail. $525M of buyback in the first four months of 2026 evidences management conviction at current prices. Mean sell-side PT $233 is +59% above spot, the widest gap in the IPP cohort. ERCOT PCM shelved is a real regulatory tailwind. VST is more diversified than TLN (no single asset >15% of EBITDA) so binary outage risk is materially lower.
Net assessment: VST scores higher than TLN's 6/10 on Concerns/Risks because (1) the multiple has already compressed by 3-4 turns vs TLN's still-stretched ~13x ex-Cornerstone, (2) diversification reduces binary single-asset risk, (3) IG balance sheet provides defensive optionality, (4) ERCOT regulatory backdrop is a tailwind where TLN's PJM BTM is a headwind, and (5) the catalyst slate is just as loaded but with less in the price. Probability-weighted fair value $160-180 vs spot $147 = asymmetric upside, with bull upside $50-90 against bear downside $15-40.