Thematic Exposure -- 9/10

Vistra is the most diversified and arguably strongest local-monopoly franchise in US merchant power. Four distinct local oligopolies stacked on top of each other: (1) ERCOT merchant nuclear -- Comanche Peak is the ONLY contractable nuclear plant in Texas (a single-asset monopoly); (2) ERCOT thermal generation -- top-4 oligopoly with VST #1; (3) ERCOT competitive retail -- TXU Energy at ~25-30% share in an effective duopoly with NRG; (4) US merchant nuclear -- #2 nationally at ~6.4 GW inside a 6-player national oligopoly. The franchise is already monetized via the two largest corporate nuclear PPAs in US history: Amazon-Comanche Peak (1,200 MW, 20-yr) and Meta-VST PJM (2,609 MW across Perry / Davis-Besse / Beaver Valley). Held one notch below 10 because multi-region gas, coal, retail, and renewables dilute the pure-monopoly framing (vs TLN's single-asset focus); ~8 GW of coal still on the books; and thematic lift is partly priced after the stock round-tripped from $219 to ~$147. Weight: 25%
Oligopoly Hard Gate: PASS on FOUR Separate Dimensions
ERCOT Nuclear Monopoly -- ERCOT Thermal Top-4 -- ERCOT Retail Duopoly -- US Merchant Nuclear Top-2
VST is the only ERCOT player that simultaneously dominates all three layers of the Texas competitive electricity stack -- generation, nuclear, and retail -- and layers a #2 national merchant nuclear position on top. This is a local franchise with no full US peer.
The Four Local Oligopolies
Market VST Position Share / Rank Structural Status
1. ERCOT merchant nuclear Comanche Peak Units 1 & 2 (~2,400 MW). ERCOT has exactly two nuclear plants: Comanche Peak (VST, merchant) and South Texas Project (rate-based / IOU-owned, NOT contractable to hyperscalers). 100% (sole merchant) SINGLE-ASSET MONOPOLY -- for a Texas hyperscaler PPA on existing nuclear, Comanche Peak is the only game in town. Monetized via Amazon 20-yr 1,200 MW co-located PPA.
2. ERCOT thermal generation Vistra ~16-17 GW gas + coal. Other top-4: Calpine (~10 GW), NRG (~9 GW), LCRA / utility-equivalents. #1 thermal Top-4 OLIGOPOLY, VST is the largest dispatchable thermal generator in ERCOT.
3. ERCOT competitive retail TXU Energy (plus Ambit and other VST-owned brands) serves over 5 million residential customer accounts. Next-largest competitor: NRG / Reliant + Direct Energy at ~20-25%. ~25-30% (#1) EFFECTIVE DUOPOLY at the top of competitive Texas retail (TXU + NRG/Reliant). TXU brand dominates net-promoter and customer-count rankings.
4. US merchant nuclear VST ~6.4 GW (Comanche Peak, Perry, Davis-Besse, Beaver Valley). #2 behind Constellation (~22 GW). #2 nationally 6-player OLIGOPOLY (CEG, VST, PSEG, NEE, TLN, Dominion). Only merchant nuclear is contractable to hyperscalers without state PUC rate-base interference.
The vertical integration -- owning generation, retail, and the only merchant-contractable Texas nuclear plant, in a state forecasting +54 GW of peak load by 2030 -- is a local franchise that has no full US peer.
ERCOT Nuclear Share
100%
Sole merchant-contractable plant (Comanche Peak)
Total Merchant Nuclear
~6.4 GW
#2 in US 6-player oligopoly (CEG #1)
Texas Retail Customers
~5M
TXU + Ambit; ~25-30% share, #1 in ERCOT
Contracted Nuclear PPAs
~3.8 GW
Amazon 1.2 GW + Meta 2.6 GW (20-yr)
Already Monetized -- The Two Mega-PPAs
Amazon-Comanche Peak (Sept 2025) + Meta-VST PJM (Jan 2026) = ~3.8 GW Locked at 20-yr Tenor
1. Amazon - Comanche Peak (September 2025). 1,200 MW co-located, 20-year PPA. Initial energization Q4 2027, full ramp 2032. This is the visible monetization of the ERCOT single-asset nuclear monopoly. Amazon also took an Oak Hill solar PPA -- existing relationship.

2. Meta - VST PJM (January 2026). 2,609 MW (2,176 MW operating + 433 MW uprates) across Perry, Davis-Besse, and Beaver Valley. 20-year tenor. Initial deliveries December 2026; full 2,609 MW online 2034. Per the VST Q4 2025 call, this is the LARGEST nuclear-uprate-supported corporate PPA in US history.

Beaver Valley (~1.8 GW) remains uncontracted and is the next monetization target ("we continue to see a very high level of interest in Beaver Valley" -- Stacey Dore, Q4 2025 call). The 2027 midpoint opportunity ($7.4-7.8B EBITDA) excludes Cogentrix and Meta, so contracted earnings will step up materially from here.
Generation Portfolio by RTO -- Where the Monopolies Live
Asset / Sub-fleet Fuel RTO MW Local Monopoly / Oligopoly Status
Comanche Peak (Units 1 & 2) Nuclear ERCOT (North) ~2,400 SOLE merchant nuclear in ERCOT -- single-asset monopoly. Amazon 1,200 MW PPA.
Perry Nuclear PJM (OH) ~1,268 One of <10 merchant nuclear plants in PJM. Meta 20-yr PPA delivery Dec 2026.
Davis-Besse Nuclear PJM (OH) ~908 PJM merchant nuclear oligopoly. Meta 20-yr PPA delivery Dec 2027.
Beaver Valley (Units 1 & 2) Nuclear PJM (PA) ~1,800 Adjacent to MAAC/DOM data-center demand sink. Uncontracted -- next monetization target.
ERCOT thermal (Permian, Midlothian, Lamar, Forney) Gas CCGT / peakers ERCOT ~13,000+ Top-4 oligopoly, VST #1. ~60% utilization -- pure operating leverage on existing assets.
Permian Power 1 & 2 (under construction) Gas peakers ERCOT (West TX) 860 New build for West TX oil & gas + data-center load; online 2028.
PJM gas fleet (EH + Lotus + legacy) Gas CCGT PJM ~7,000+ Capacity-revenue heavy; PJM 2026/27 BRA cleared ~$329/MW-day (near cap).
Cogentrix (pending H2 2026 close) Gas CCGT (10 plants) PJM + others ~5,500 Competitive; ~$730/kW net of tax benefits.
Lotus (closed Oct 2025) Gas CCGT / CT (7 plants) PJM / ISO-NE / NYISO / CAISO ~2,600 Geographic diversification; competitive markets.
MISO / ISO-NE / NYISO / CAISO gas + coal Gas / Coal Multi-region ~5,000+ Capacity + energy; coal on glide-path retirement.
Texas coal (Martin Lake, Oak Grove) Coal ERCOT ~3,500 Capacity; retirement pushed out by MATS extension. Local but declining.
Midwest coal (Miami Fort, Baldwin, Newton, Kincaid) Coal PJM / MISO ~4,500 Retirement glide-path; Miami Fort coal-to-gas conversion optionality.
Vistra Zero (solar / BESS) Solar / BESS ERCOT / PJM / MISO ~1,200 Contracted PPAs (Amazon Oak Hill, Microsoft Pulaski). Competitive market.
Total -- -- ~41,000 MW pre-Cogentrix ~46,000 MW post-Cogentrix close. Nuclear ~6.4 GW = #2 US merchant nuclear operator behind CEG (~22 GW).
2025 adjusted EBITDA $5.91B (Generation $4.29B + Retail $1.62B record). 2026 guide $6.8-7.6B. 2027 midpoint opportunity $7.4-7.8B EXCLUDES Cogentrix and Meta contributions.
US Merchant Nuclear Oligopoly -- VST is #2
Operator Merchant Nuclear (GW) PJM-Deliverable (GW) Notes
Constellation (CEG) ~22 ~14 #1 nationally; largest PJM-deliverable nuclear footprint.
Vistra (VST) ~6.4 ~4.0 #2 nationally. Perry + Davis-Besse + Beaver Valley PJM-deliverable.
PSEG ~3.8 ~3.8 Salem, Hope Creek -- all PJM-deliverable.
NextEra (NEE) ~2.3 minimal Primarily outside PJM.
Talen (TLN) ~2.2 2.2 Susquehanna -- single-asset focus (vs VST's multi-asset diversification).
Dominion ~2.0 minimal Primarily outside PJM.
Top 6 combined ~38.7 ~24 6-player oligopoly; only contractable nuclear capacity for hyperscalers without state PUC rate-base interference.
Total US merchant nuclear contractable to hyperscalers: ~30 GW. Hyperscaler PPA demand to date (Amazon + Meta + Microsoft + Google + Oracle/OpenAI Stargate) already exceeds 50 GW of interest -- demand materially exceeds available merchant nuclear supply.
Theme Growth -- AI Power Demand (Bull-Case Trajectory)
Datapoint Value Source / Note
ERCOT 2025 LTLF3 peak load forecast (2030) 139 GW (vs ~85 GW current) +54 GW; ~85% from data centers / crypto / electrification. ERCOT 2025 LTLF.
ERCOT data-center share of growth potential 77.9 GW by 2030 (vs 29.6 GW prior year) ERCOT 2025 RFI -- step-change in interconnect demand.
ERCOT large-load interconnect queue 136 GW Energization dates 2025-2030. ERCOT, April 2025.
VST view of realistic ERCOT incremental load by 2030 30-40 GW (10-15 GW large DC) VST Q1 2026 call (Jim Burke) -- management haircut to LTLF.
ERCOT annual peak-load growth (VST view) 5-6% through 2030 VST Q1 2026 call.
PJM 2025 LTLF peak load growth by 2030 +32 GW (94% data centers) PJM 2025 LTLF -- structural step-change driven almost entirely by AI/DC load.
PJM annual peak-load growth (VST view) 2-3% through 2030 VST Q1 2026 call.
PJM 2026/27 BRA clearing price ~$329/MW-day (near cap) PJM, July 2025 -- structural capacity tightness.
Hyperscaler 2026 CapEx ~$700B (+50% YoY) VST Q4 2025 call commentary.
US electricity consumption 2025 ~4,200 TWh (+2.5% YoY) Returning to growth after 15 years of flat demand.
VST CCGT fleet currently runs at ~60% utilization. Jim Burke (Q1 2026 call): this can rise to mid-80s as load grows -- pure operating leverage on existing assets, no incremental capex required.
Texas Retail Moat -- TXU Energy
~5M Customer Accounts -- #1 Brand -- $1.62B Record Retail EBITDA in 2025
TXU Energy is the largest competitive residential retailer in Texas. Including Ambit and other Vistra-owned brands, VST serves over 5 million residential customer accounts (Daloopa-confirmed). Of an addressable competitive ERCOT residential market of ~8-9M premises, VST's share is roughly 25-30%, with the next-largest competitor (NRG / Reliant + Direct Energy) at ~20-25% -- a two-player oligopoly at the top of competitive Texas retail.

Switching costs and brand value: TXU dominates net-promoter and customer-count rankings; brand recall is high; switching friction (contract terms, deposits, billing migration) is material at the individual-household level. Retail delivered record $1.62B EBITDA in 2025, well above the "medium-term" management target of ~$1.4B -- the moat is widening, not eroding.

Counter-cyclical ballast: the retail book naturally hedges generation -- when wholesale prices spike, retail margins compress but generation profits expand, and vice-versa. This is the structural advantage VST has over pure-play merchant generators like TLN.
Replacement Supply Economics -- Substitutes Are Throttled
Substitute Status Time to Market
New AP1000 nuclear builds $25-35B per unit; first-of-a-kind risk; permitting + construction. 10+ years
Small modular reactors (SMRs) Pre-commercial; NRC licensing in progress; supply chain immature. Post-2030
New gas turbines (GE Vernova, Siemens, MHI) OEM backlog sold out into 2030. Cost of new-build gas "more than doubled in the last 5 years" per Jim Burke Q1 2026 call. 3-4 years (sold out)
Renewables (solar + wind) Cannot meet 24/7 hyperscaler reliability requirement at GW scale. Intermittency is structural. Cannot deliver 24/7
Battery storage (BESS) Useful for grid balancing, not bulk supply. VST mgmt: ERCOT battery ROI has been "virtually nothing." Insufficient duration
VST advantaged new-build position Pre-ordered turbines; 70 existing sites with permits / interconnects; lower marginal cost than greenfield competitors. Structural edge
Substitutes are throttled across every dimension. Existing merchant nuclear + dispatchable gas is the only supply that can serve hyperscaler 24/7 load at scale before 2030 -- VST sits inside both oligopolies.
Buyer Concentration -- Demand Materially Exceeds Supply
Hyperscalers Need >50 GW -- Merchant Nuclear Contractable ~30 GW -- Demand >> Supply
Buyers (hyperscaler-dominant):
  • Amazon (AWS): Comanche Peak 1,200 MW 20-yr co-located PPA + Oak Hill solar PPA -- existing relationship.
  • Meta: 2,609 MW 20-yr PPA across Perry / Davis-Besse / Beaver Valley uprates -- largest in US history.
  • Microsoft: Pulaski solar PPA + ongoing nuclear discussions.
  • Google, Oracle / OpenAI Stargate: actively shopping; VST has multiple "advanced" gas + nuclear discussions per Q1 2026 call.
  • Retail base: ~5M residential + millions of C&I customers via TXU / Ambit -- diversifies away from pure hyperscaler concentration.
Demand vs supply gap: hyperscaler-stated 2026-2030 power demand exceeds 50 GW; total US merchant nuclear contractable supply is ~30 GW. The structural shortage means PPA pricing power sits with sellers (VST, CEG, TLN, PSEG) -- not with the four hyperscaler buyers, despite their concentration. Mitigated further by the 5M-customer retail base which provides a base-load earnings stream uncorrelated to hyperscaler PPA negotiations.
Why Held at 9 -- Not 10
Factor Why It Matters Drag
Four monopolies dilute the pure-monopoly framing TLN is a single-asset pure-play (Susquehanna). VST is materially more diversified -- multi-region gas, coal, retail, renewables -- which adds operating complexity and dilutes the "single-monopoly" narrative even though it strengthens the franchise overall. Mild -- vs TLN narrative
Residual coal exposure ~8 GW of coal still on the books (Martin Lake, Oak Grove, Miami Fort, Baldwin, Newton, Kincaid). Retirement glide-path with conversion optionality (Miami Fort coal-to-gas), but ESG overhang and retirement uncertainty remain. Modest
Thematic lift partly priced Stock round-tripped from $219 peak (Sep 2025) to ~$147 (May 2026). Better entry than at peak, but still embeds meaningful AI optionality. Not a clean contrarian setup. Modest
Regulatory clarity still pending PJM co-location rules and ERCOT SB6 batching still being finalized. FERC December 2025 co-location order is favorable directionally, but rule clarification is ongoing. Modest
Buyer concentration in PPAs 4 hyperscalers dominate large PPA decisions. Partially offset by 5M-customer retail base, but the marginal PPA negotiation is concentrated. Mild
Rubric for 10: ">50% local segment share + theme growing >10% + oligopoly (≤3 players)." VST hits MONOPOLY on ERCOT merchant nuclear and OLIGOPOLY on three other dimensions, with theme growth well above 10% on the ERCOT peak forecast -- only the diversification, residual coal, and partly-priced valuation hold it one notch below 10.

Score Rationale
Factor Assessment Impact
Local oligopoly -- ERCOT merchant nuclear 1-player monopoly (Comanche Peak) for hyperscaler PPAs Strong +
Local oligopoly -- ERCOT thermal generation Top-4 oligopoly, VST #1 Strong +
Local oligopoly -- ERCOT retail #1 share (~25-30%); 2-player effective duopoly with NRG Strong +
National merchant nuclear position #2 (~6.4 GW) of 6-player US oligopoly Strong +
Theme growth -- ERCOT load Peak forecast +54 GW by 2030 (LTLF3); ~85% data centers Strong +
Theme growth -- PJM capacity prices 2026/27 BRA cleared ~$329/MW-day (near cap) Strong +
Contracted revenue locked ~3.8 GW nuclear PPAs + Oak Hill / Pulaski; ~50% EBITDA shifting to stable Strong +
Diversification (gen + retail, multi-region) Smooths volatility but dilutes pure-monopoly framing of TLN Mild -- (vs risk +)
Buyer concentration 4 hyperscalers dominate large PPAs; mitigated by 5M retail customers Mild --
Substitutes / new-build cost inflation New gas/nuclear sold-out and expensive -- favors VST existing fleet Strong +
Regulatory (ERCOT SB6, FERC co-location) SB6 passed; FERC Dec 2025 co-location order favorable; ongoing rule clarification Moderate +
Coal exposure ~8 GW coal on books; retirement glide-path; conversion optionality Neutral
Valuation reflects theme Stock round-tripped from $219 peak to ~$147 May 2026 -- better entry than at peak Mild +
9/10 — VST is the most diversified and arguably the strongest local-monopoly franchise in US merchant power. In ERCOT -- the highest-growth power market in the country (peak 85 to 139 GW by 2030) -- VST owns the only merchant-contractable nuclear plant (Comanche Peak), is the #1 thermal generator, and runs the #1 retail brand (TXU). Layered on top: #2 US merchant nuclear position with ~4 GW PJM-deliverable nuclear (Perry, Davis-Besse, Beaver Valley) sitting inside a 6-player national oligopoly, already monetized via the 20-year, 2,609-MW Meta PPA (largest US corporate nuclear PPA ever).

Oligopoly hard-gate: PASS on at least 4 separate dimensions -- ERCOT merchant nuclear monopoly, ERCOT thermal oligopoly, ERCOT retail duopoly, US merchant nuclear oligopoly.

The 2027 midpoint opportunity ($7.4-7.8B EBITDA) excludes Cogentrix and Meta contributions, so contracted earnings will step up materially. Per CFO Moldovan (Q4 2025 call), full execution of nuclear PPAs plus Cogentrix plus share-repurchase deployment of all unallocated cash through 2030 implies adj. FCF/share could reach $22-25 vs. 2026 starting point of $12.50. AI-driven demand growth (bull-case ERCOT LTLF3 trajectory) provides multi-year tailwind structurally undersupplied by the new-build pipeline.

Held one notch below 10 because: (a) the company is materially more diversified than a pure-play nuclear name like TLN -- multi-region gas, coal, retail, and renewables dilute the "single-monopoly" framing and add operating complexity; (b) ~8 GW of coal still on the books with retirement / conversion uncertainty; (c) valuation, while well off the $219 peak, still embeds meaningful AI optionality at ~$147; and (d) regulatory clarity on PJM co-location and ERCOT SB6 batching is still being finalized.
Sources: ERCOT 2025 Long-Term Load Forecast Update; PJM 2025 LTLF; Grid Strategies 2025 National Load Growth Report; EIA -- rapid electricity demand growth in Texas and mid-Atlantic; Vistra & Meta Announce 20-Year Nuclear PPAs (Vistra IR, Jan 2026); ANS -- The Meta-Vistra deal: A closer look; Power Magazine -- Meta locks in 6.6 GW nuclear power; TXU Energy public market-share rankings; VST Q1 2026, Q4 2025, Q3 2025, Q2 2025, Q1 2025, Q4 2024 earnings call transcripts; Daloopa fundamentals as of May 2026.