Valuation -- 6/10
| Metric | TLN (PF) | TLN ex-CS | CEG | VST | NRG | AES |
|---|---|---|---|---|---|---|
| Fwd EV/EBITDA (2026E) | ~10-11x | ~13x | ~13x | ~10x | ~10x | ~7-8x |
| Fwd EV/EBITDA (2027E) | ~9-10x | ~11-12x | ~12x | ~9x | ~9x | ~7x |
| Fwd P/E (2026E) | ~25-28x | -- | ~26x | ~18x | ~17x | ~9x |
| FCF Yield (2026E) | ~6-7% | -- | ~3.8% | ~7.2% | ~6-7% | ~10%+ |
| Net Debt/EBITDA (YE26E) | <3.5x | ~2.5x | ~1.5x | ~3.0x | ~3.5x | ~5.5x |
| Key Takeaway | TLN trades roughly in line with VST/NRG on PF EV/EBITDA when giving credit for Cornerstone and the AWS ramp, at a modest discount to CEG (IRA-PTC certainty, IG balance sheet), and at a premium to AES. Ex-Cornerstone standalone TLN is ~13x -- 2x historical IPP norms of 6-8x. FCF yield is the cleaner lens: ~6-7% in 2026 stepping to ~9% PF 2027 is competitive with VST/NRG and superior to CEG. | |||||
| Metric | 2024A | 2025A | 2026E Guide |
|---|---|---|---|
| Adjusted EBITDA | $770M | $1,035M | $1,750M - $2,050M |
| Adjusted Free Cash Flow | $283M | $524M | $980M - $1,180M |
| YoY EBITDA Growth | -- | +34% | ~+83% (mid) |
Per the IPP/merchant power rubric, EV/EBITDA and FCF yield are the correct primary metrics -- trailing/forward P/E is distorted by:
- Heavy D&A on long-lived nuclear and CCGT assets (Susquehanna alone is ~2.5 GW dual-unit, 42 years old, with a license through 2042/2044 and an SLR application pending to extend 60->80 years).
- Nuclear PTC mechanics -- Susquehanna did NOT receive the PTC in 2025 because realized prices stayed above the $43.75 strike. PTC is a price-based floor, not a steady credit, so EPS swings on policy and forward curves.
- Cornerstone purchase accounting -- $3.45B price (~$2.55B cash + 2.4M shares + $2.6B new debt) temporarily inflates leverage and amortization above steady-state.
EV/EBITDA captures asset-level cash earning power across the fleet on an apples-to-apples basis with CEG/VST/NRG. FCF yield captures the actual cash available for buybacks ($2B authorized through 2028) and incremental M&A -- the real currency of IPP capital allocation. On these two metrics TLN looks expensive vs. history (13x vs. 6-8x norm) but reasonable vs. AI-power peer set, with FCF yield stepping from ~6-7% in 2026 to ~9% PF 2027.
| Scenario | Probability | 2027E PF EBITDA | EV/EBITDA | Fair Value Range | Drivers |
|---|---|---|---|---|---|
| Bull | 30% | $2.8-3.0B | 11-12x | $450-$525 | Second 1 GW hyperscaler PPA signed by YE26, RBP rules favorable, BRA cap removed |
| Base | 50% | $2.5-2.7B | 9-10x | $340-$410 | Cornerstone closes on time, AWS ramps on schedule, no incremental PPA in 2026 |
| Bear | 20% | $2.2-2.4B | 7-8x | $200-$275 | FERC tightens BTM further, no second PPA, gas prices soften, cap stays in place |
| Probability-Weighted Fair Value | ~$355-$400 | Spot $374.61 sits inside base case; bull upside ~$100-$150, bear downside ~$100-$175. Asymmetric tilt depends entirely on hyperscaler PPA cadence in 2H 2026. | |||
| # | Catalyst | Impact | Timeframe |
|---|---|---|---|
| 1 | Second hyperscaler PPA (post-Montour pivot) | Mgmt pipeline active across multiple sites; each 1 GW = ~$1-1.5B incremental NPV. Delivery post-2028 regardless of signing date. | 2H 2026 - 1H 2027 |
| 2 | PJM RBP / Reliability Backstop Procurement | A pay-as-bid 15-yr procurement would unlock financeable new-build and add upside to existing-asset capacity revenue. Mgmt in active coalition. | 2H 2026 ruling |
| 3 | Cornerstone close (summer 2026) | Adds ~$500M EBITDA run-rate, >$4/share incremental FCF. Already announced, regulatory close pending. | Q3 2026 |
| 4 | PJM Capacity Auction prints | 2026/27 BRA cleared at $329.17/MW-day cap ($16.1B total); 2027/28 cleared at $333.44/MW-day cap ($16.4B total) -- both at price collars. Removal/raising the cap is clear optionality. | Mid-2026 next auction |
| 5 | AWS PPA ramp acceleration | Existing contract allows acceleration up to 480 MW earlier than ramp; each 480 MW tranche ~$2/share FCF impact per mgmt slide. | Ongoing 2026-2028 |
| 6 | Susquehanna re-licensing (60 -> 80 yrs) | Existing license runs to 2042/2044. SLR application planned; extension underpins long-dated contract value. | 2027-2030 process |
| 7 | Hedge book monetization | TLN intentionally under-hedged in '27/'28 outer years; rising forward sparks (+15% Q4 2025) provide opportunistic hedge layering. | Ongoing |
| 8 | Buybacks ($2B authorized through 2028) | At ~3.0x leverage today, room to lean in on dips. Average price paid ~$149 to date. | Ongoing |
| # | Risk | Severity | Probability | Detail |
|---|---|---|---|---|
| 1 | FERC behind-the-meter overhang | HIGH | HIGH | FERC rejected the amended Susquehanna ISA in Nov 2024 (300 -> 480 MW BTM); in March 2026 FERC denied rehearing. AEP/Exelon claim $140M/yr cost-shift to ratepayers. Amazon 2.0 PPA pivot to front-of-the-meter mitigates but at lower spread economics. Single biggest structural risk to the IPP-hyperscaler thesis. |
| 2 | Montour / second PPA slippage | HIGH | MEDIUM | Montour County Commission vote went against TLN; mgmt explicitly walked away from "discussing development in public forum." Any 6-12 month delay in a second hyperscaler PPA could compress the multiple given how much is priced in. |
| 3 | Single-asset concentration at Susquehanna | HIGH | LOW-MED | Susquehanna (~2.5 GW nuclear) is ~50% of equity value and anchor of the AWS PPA. A multi-month forced outage (e.g. Q2 2025 Unit 2) materially impairs cash flow. A nuclear safety event anywhere in the US fleet would crater the multiple. |
| 4 | Valuation reset / multiple compression | HIGH | MEDIUM | At ~13x 2026E EBITDA ex-Cornerstone, well above historical IPP norms (6-8x) and at parity with VST despite less geographic diversity. If AI capex narrative cracks, 3-4 turns of multiple = 30-40% downside. Consensus PT ~$455 is only ~21% above spot. |
| 5 | PJM capacity cap / political backlash | MEDIUM | MED-HIGH | Last two BRAs cleared at the cap. Governor Shapiro and PA/Ohio politicians focused on affordability. RBP rules could include cost-allocation provisions or carve-outs that limit upside. The +71% PPL-zone load forecast invites political intervention on ratepayer protection. |
| 6 | Power price / gas price normalization | HIGH | MEDIUM | Meaningful share of EBITDA is merchant on the fossil fleet (Freedom, Guernsey, Cornerstone CCGTs). Sustained low gas + mild weather + DC demand under-delivery would compress sparks materially -- 2026/27 sparks already moved +15% but can reverse. |
| 7 | Cornerstone integration / leverage | MEDIUM | LOW-MED | $3.45B price (~$2.55B cash + 2.4M shares), $2.6B new debt. Pro-forma leverage temporarily above 3.5x target. If standalone EBITDA disappoints, balance sheet less flexible. |
| 8 | Nuclear PTC mechanics | MEDIUM | LOW | Susquehanna did not receive the PTC in 2025 (price-based mechanism). If realized prices stay above the $43.75 strike, no PTC -- bullish on energy revenue but removes a floor. Repeal/modification of the IRA nuclear PTC is a tail risk under the current administration. |
The "powering AI" thesis is real and TLN owns the single most valuable nuclear asset positioned for it -- Susquehanna, 2.5 GW dual-unit, dedicated transmission, AWS already contracted at scale. Management is best-in-class commercial operators with a proven flywheel: contract -> acquire -> hedge -> repurchase. FCF per share doubles 2025 -> 2026 and grows again in 2027; Cornerstone is highly accretive (~$4+/share); a second hyperscaler PPA is a matter of when, not if; and PJM capacity prices clearing at cap signal a structurally tight market for 5+ years.
You are paying ~13x 2026 standalone EBITDA -- 2x historical IPP multiples -- for a single-asset-concentrated merchant generator whose flagship deal structure was rejected twice by FERC and whose next big PPA just got blocked by a county commission. The hyperscaler bid is exquisitely sensitive to capex cycles; AWS, MSFT, GOOG could rationalize spend in any 6-month window. Susquehanna's spring 2025 outage cost ~$100M+ of EBITDA and the asset is 42 years old. Buybacks have already happened at the cheap prices (~$149 avg); from here management is buying at a premium.
Score of 6/10 reflects balanced-to-favorable risk/reward with thin margin of safety.
Why not higher (7-8): The FERC BTM rejection is now adverse and final on rehearing (March 2026), removing a structural option that was central to the original Amazon 1.0 deal economics. Single-asset concentration at Susquehanna (~50% of equity value) is real and the asset is 42 years old. The stock has already moved +55% YoY -- much of the bull case is priced. Consensus PT of ~$455 is only ~21% above spot, signaling sell-side is also out of room. At ~13x standalone 2026 EBITDA, you are paying 2x historical IPP multiples with limited cushion if any one of the 2027/2028 hyperscaler PPAs slips or RBP rules disappoint.
Why not lower (4-5): The catalyst slate is genuinely loaded -- Cornerstone close in Q3 2026, second PPA pipeline active across multiple sites, RBP ruling expected 2H 2026, capacity auction cap removal optionality, AWS ramp acceleration. Cornerstone-adjusted valuation (~10-11x 2026E) is in line with VST/NRG and below CEG. FCF yield of ~6-7% in 2026 stepping to ~9% PF 2027 is competitive. The AI demand thesis remains intact per management ("data centers are coming and they're not slowing down") and PJM capacity clearing at cap two auctions running confirms structural tightness for 5+ years.
Net assessment: Probability-weighted fair value ~$355-$400 brackets spot of $374.61. Bull upside of ~$100-$150 vs. bear downside of ~$100-$175 is symmetric, with asymmetric tilt depending entirely on hyperscaler PPA cadence in 2H 2026. Suitable as a core AI-power position sized for the PPA-cadence binary, with willingness to add on pullbacks tied to Montour/FERC headlines rather than thesis breaks.