NRG Energy

▲ Bull Case

• ◆LSP synergy + new-build flywheel. $200M+ in cost-side LSP synergies (shared O&M, hedging consolidation, debt refi) layered on top of acquired EBITDA, plus all 5.4 GW new-build commercial by 2030 at $80–90/MWh net spark spread, drives 10% EBITDA CAGR to ~$8.5B by 2030. Combined with multiple re-rate to 10.5x (convergence with VST + premium for new-build visibility) → ~$220/share, +63%.
• ◆Data center capture exceeds plan. NRG hits 1+ GW contracted in 2026 and scales toward the 6.5 GW pipeline, adding $300–400M of incremental EBITDA by 2030. Bull-case 2030 EPS of $22–24 vs. management's stated $14+ target.
• ◆Buyback compounds the per-share story. $1B+/yr buyback for five years takes share count from 215M to ~180M, with no growth-capex starvation because FCFbG reaches $4.5B by 2030. FCF/share runs to ~$21–22 from $13 today — a 65% per-share compounding even at flat multiple.

▼ Bear Case

• ◆LSP integration drags and multiple compresses. Cost-side LSP integration is messier than expected; one moderate ERCOT weather year hits 2027 or 2028; data-center MOUs convert at lower-than-hoped margin because hyperscalers route to direct-PPA with renewable developers. EBITDA CAGR collapses to 2.2%; terminal multiple compresses to 7.5x as market re-rates NRG as a higher-debt, lower-quality merchant generator → ~$95/share, −30%.
• ◆ERCOT extreme weather tail still binds. Winter Storm Uri cost NRG ~$800M pre-tax in a single week in Feb 2021. Post-Uri winterization + market reforms reduce probability and severity but cannot eliminate the tail. Climate science suggests more frequent extremes; one Uri-class event in the next five years would hit $700M–$1B in a single quarter and likely suspend the buyback.
• ◆Leverage is now the real constraint. Net debt post-LSP sits at ~$23B against ~$5.6B EBITDA (~4.1x trailing, ~3.5–3.7x on forward 2026E); rating agency tolerance is the binding constraint on buybacks. Any combination of weather loss + LSP miss + commodity stress could push leverage above 4.0x, trigger a downgrade to sub-investment grade, suspend the buyback, and remove the single-largest per-share return mechanism for multiple years.

“Sell-side consensus runs constructive (~60% Buy, ~40% Hold). The active debates are: (1) LSP integration execution — Is $200M+ in synergies achievable, and will the 5.4 GW new-build pipeline commercial on schedule (first 1.2 GW in 2029)? (2) Data-center capture vs. bypass — Does NRG-the-retailer or NRG-the-generator capture ERCOT data-center load? Bull view: NRG's Reliant C&I sales force + new-build gas fleet is uniquely positioned. Bear view: hyperscalers prefer direct PPAs with renewable developers. (3) Multiple convergence vs. permanent discount — Should NRG re-rate to VST's 10x or beyond, or does the post-LSP higher-leverage profile justify a permanent discount? What consensus is not debating but probably should: the persistence of the leverage drag if the 2026 weather year is mild (lower retail load + lower spot premium → leverage normalization slips → buyback halved → multiple stalls).”

• ◆Q2 + Q3 2026 earnings — first full quarters with LSP. Synergy progress disclosure; data-center contract progress against 1 GW 2026 target; capex pacing on new-build gas; refreshed 2027 guidance shape. Late July / early November 2026.
• ◆Data-center contract announcements. Conversion of the 6.5 GW LOI pipeline into binding multi-year supply agreements with named hyperscalers. Each large deal is a multi-hundred-million dollar EBITDA datapoint.
• ◆Buyback pace and authorization refresh. Continued ~$200M+/quarter execution and a 2027 authorization at or above $1B would lock in the per-share compounding story.
• ◆2026 ERCOT summer outcome. Normal-to-hot summer = retail load + premium spot pricing tailwind ($150–200M EBITDA); mild summer = ~$100–150M headwind.
• ◆Leverage milestone. First quarterly disclosure showing Net Debt / EBITDA at or below 3.0x triggers ratings stability and buyback acceleration discussions.
• ◆2027 investor day. Expected refresh of the "$14+ EPS by 2030" target with new-build schedule, synergy update, and data-center pipeline disclosure.
• ◆GE Vernova partnership milestones. Confirmation that first 1.2 GW of new-build is on schedule for 2029 commercial ops + permitting progress on remaining 4.2 GW.

• ◆ERCOT extreme-weather event (Uri-class). Single event could cost $700M–$1B and suspend buyback. Probability ~5%/yr; cannot be fully hedged in merchant retail model.
• ◆LSP integration failure / impairment. $200M+ synergy miss, integration drag into 2028, possible $1–2B goodwill impairment if 2030 EBITDA targets miss materially. Probability ~20%.
• ◆Leverage downgrade to sub-IG (BB+). Net debt $23B / EBITDA $5.6B = 4.1x trailing; any miss pushes to 4.5x+ territory and sub-IG downgrade, suspending buyback and removing primary per-share return driver.
• ◆Data-center hyperscaler bypass. If data centers route to direct-PPA with renewable developers, NRG retail loses the C&I capture story and misses $300–400M incremental EBITDA by 2030.
• ◆Vivint subscriber erosion. Vivint churn at 12–14% is load-bearing for $1.15B segment EBITDA. Amazon Ring + DIY competition is structural pressure.