NuScale Power was spun out of Oregon State University in 2007 — DOE national-lab IP, a small reactor design intended to solve the one problem the AP1000 generation could not, which was that nobody could finance a $14B single-asset bet anymore. Fluor took a controlling stake in 2011 and carried the company through a decade of regulatory work that culminated in 2023 with the NRC certifying the 50 MWe design — the only SMR ever certified in the United States. Then the original commercial path broke. The Carbon Free Power Project, the UAMPS deployment that was supposed to be the reference plant, was cancelled in November 2023 when escalated cost pushed the LCOE past what muni offtakers would underwrite. What was left was a certified design with no customer.
The inflection arrived from outside the company: AI data-center load forecasts that quietly added a hyperscaler-scale dispatchable demand curve, a Trump-administration nuclear tailwind, and NRC approval of the upsized 77 MWe variant. The market now prices SMR as a binary on contract conversion — TVA's 6 GW non-binding MOU, the ENTRA1 commercialization vehicle, Fluor's full equity exit — against a $1.25B cash cushion and accelerating burn. Watch Q3 2026 for binding-contract velocity; that is the print that resolves the binary.
Maintain existing NuScale Power positions at ≤1.0% portfolio weight; no new entry pre-May-7 earnings. Hard stop at $9.50 (above the $9.16 April 7 swing low for stop-run buffer). Re-evaluation gate is October 2026 Q3 earnings — if no binding-contract conversion or DOE allocation event by then, exit. Time horizon: 6-month decision window with 18-month thesis horizon.
This is a Hold by genuine arbitration, not by default. The bull researcher cleared 4.5 of 5 items on the institutional specificity bar — two dated catalysts (May 7 earnings + 2026 H2 DOE allocations), a credible budget-cycle reference (DOE $800M plus IRA ITC/PTC stack), a structurally sound consolidation/capacity narrative (Houston Operations Hub, hyperscaler sequential demand), and a tightened 5-month falsifiable kill-switch (October 2026 Q3 earnings). The customer-segment growth math was the soft pillar — the 6 GW TVA MOU is non-binding, and conversion velocity in 2025 was zero.
The bear's evidentiary stack is the harder set of facts to argue with. FY25 revenue was -15% YoY; Q1 2026 expected -63% YoY. Cash burn quadrupled YoY ($109M → $460M). Share count rose +159% in 12 months (122.8M → 318.5M); the company has now requested authorization for another 330M shares (332M → 662M). Fluor — the most informed long-term holder — fully exited at ~$2.43B cumulative proceeds. The DEF 14A explicitly references going-concern adjacency conditional on the authorization vote.
The risk debate did not produce dissent on direction — the aggressive analyst leaned slight-overweight, the conservative leaned slight-underweight, and the neutral synthesized a Hold with defined-edges execution. The synthesis is that NuScale at $11.90 is neither cheap enough to catalyze a Buy nor broken enough to force a Sell. The May 7 earnings event is too close to chase and the catalyst cluster (earnings, authorization vote, DOE allocations, TVA conversion, RoPower FID) is concentrated in the next 6 months — five binary events whose joint resolution drives the rating.
The cross-ticker memory note from prior pipelines ([2026-04-30 AMTM Overweight: -2.2%]) is a useful prior. AMTM was a similarly cyclical, idiosyncratic industrial that underperformed despite a constructive analyst-side debate. The lesson translates: when the debate is genuinely close and the catalyst cluster is concentrated, sizing discipline matters more than directional conviction. A Hold is the correct expression of "the debate is honest, the data is mixed, the catalysts are imminent."
Level
Price
Basis
Stop Loss
$9.00 (volume-confirmed close below $9.16 swing low = thesis broken).
Trader-defined risk floor
Entry
No new entry. For aggressive-mandate accounts that *must* establish a small position post-May-7, target $10.50-11.00 if Q1 misses but TVA color is constructive; $13.00-13.50 if Q1 hits and a binding-contract path is articulated.
Trader-defined entry zone
Price Target
6-month range $9.50 (stop) to $14.50 (Q1 + TVA color upside scenario); 12-month range $7.00 (next dilution + thesis impairment scenario) to $20.00 (binding TVA PPA + DOE allocation scenario). Probability-weighted central tendency: $11-13.
PM 12-month target
Current Price
$12.18
Last close from OHLCV
Quantitative Lane
Customer Concentration
Historic 92% Fluor revenue concentration — Fluor fully exited equity stake (~40M shares, $2.43B realized since Sept 2025). Forward customer base unproven and largely non-binding.
Customer
% Revenue
Type
Fluor Corporation (historic prime; exited equity)
~92% (FY24)
Strategic equity holder, prime EPC — exited 2025-2026
RoPower (Romania) subcontract
~5-10%
FEED2 work concluded Q4 2025; FID slipped to late 2026 / early 2027
Tennessee Valley Authority
0%
Non-binding MOU for up to 6 GW — no revenue recognition
ENTRA1 Energy (commercialization partner)
n/a
Net cash recipient — $507M PMA milestone paid Q4 2025
U.S. Department of Energy (cost-sharing)
~5-15%
Steady-state cost-sharing residual
As of: Q4 2025 / DEF 14A disclosure
Backlog & Book-to-Bill — Not applicable
NuScale is pre-commercial — no commercial reactor module shipped, no binding-contract backlog under GAAP. Reported revenue derives from FEED engineering studies (RoPower, concluded Q4 2025) and DOE cost-sharing. The TVA 6 GW MOU is non-binding and not bookable. The relevant forward indicator is MOU-to-binding-contract conversion velocity, which was zero in 2025.
Catalyst Calendar
Catalyst nodes are color-coded by impact (red=high, amber=medium, tan=low). Events grouped by quarter.
Research Manager
Investment Plan — Hold
Rationale:
This was the closest debate I have arbitrated. The bull cleared 4.5 of the 5 specificity-bar items — two dated catalysts (May 7 earnings + 2026 H2 DOE allocations), a clean budget-cycle reference (DOE $800M + IRA stack), a credible consolidation/capacity narrative (Houston hub geographic positioning + hyperscaler sequential demand), and a tightened falsifiable kill-switch (October 2026 Q3 earnings). The customer-segment math (the 5th item) is where the bull is weakest: a 6 GW TVA MOU multiplied by illustrative $5,000/kW pricing produces a $30B headline that is not legitimately addressable until conversion to binding contracts. The bull acknowledged this but used it anyway — that's the half-credit deduction.
The bear's strongest weapons were uncomfortable for the bull to defend: revenue is declining (-15% FY25, -63% Q1 2026E) precisely in the year management's narrative says is the inflection; cash burn quadrupled FY25-over-FY24; the most-informed long-term equity holder (Fluor) fully exited; and per-share economics on any success scenario are degraded by ongoing and announced future dilution (332M → 662M authorized share request). The bull's strongest concession was the tightened kill-switch — admitting falsification could trigger by October 2026, only 5 months out, on the bull's own framing — which is a high probability event by the bear's own evidentiary stack.
A position that the bull frames as "small-sized, high-asymmetry, contrarian" is intellectually closer to "binary option with a 5-month resolution window and accelerating left-tail" than to a conviction Buy. Nothing in the debate justified Buy or Overweight — the bull's strongest defenses were architectural, not numerical, and the bear's strongest critiques were numerical. Conversely, an outright Sell would over-discount the genuinely valuable NRC moat, the $1.25B liquidity cushion, and the real DOE/policy tailwind. The May 7 earnings catalyst is two days away; risk-adjusted, the right action is to neither chase nor abandon.
Strategic Actions:
Not included in this analysis.
Trader
Transaction Proposal — Hold
Action: Hold
Reasoning: The bull/bear debate produced a Hold on a borderline-Overweight specificity score, with the cap imposed by the bear's strongest evidence stack — declining revenue in the supposed inflection year, accelerating cash burn (-$203.7M Q4 alone), severe dilution (+159% share count YoY), and the Fluor full-exit. The May 7 Q1 earnings catalyst is too close to chase, with the asymmetry equally probable in both directions. The technical setup (price holding 50 SMA at $11.67, MACD newly positive, RSI 52 neutral) is consistent with a wait-and-see posture rather than an entry signal.
Entry Price: No new entry. For aggressive-mandate accounts that must establish a small position post-May-7, target $10.50-11.00 if Q1 misses but TVA color is constructive; $13.00-13.50 if Q1 hits and a binding-contract path is articulated.
Position Sizing: Existing positions maintained at current weight. New positions: 0% pre-May-7. Post-May-7 maximum new sizing: 0.5-1.0% of portfolio for accounts with venture-tier risk tolerance and 18-month holding period.
FINAL TRANSACTION PROPOSAL: HOLD
Investment Debate
Bull Researcher
Bull Researcher — Round 1
I want to be clear up-front: I'm advocating for NuScale at $11.90 because the asymmetric setup is rare, and I'll meet your specificity bar head-on. This is not a "story stock" pump — it's an under-owned, oversold, optionality trade with five concrete pillars.
1. Two dated catalysts (per the catalyst_calendar):
2026-05-07 (two days from now): Q1 2026 earnings call. Consensus is set at -$0.11 EPS, $4.99M revenue — a bar so low that any positive update on the TVA conversion timeline or RoPower FID can drive a 15-25% re-rate. The April 22 +16% single-day spike on no news tells us shorts are positioned and dry powder is sidelined.
2026 Q2-Q3: DOE $800M SMR program first allocations. NuScale is the only NRC-approved SMR design — the only one. When DOE writes checks for SMR deployment in mid-2026, NuScale must be a primary recipient. A $50-100M allocation announcement is a hard catalyst.
2. Customer-segment growth math (not "TAM is huge"):
TVA MOU = up to 6 GW of NuScale capacity. At illustrative pricing of $5,000/kW (NuScale's stated cost target post-FOAK), that is $30B of contract value if even a fraction converts. NuScale's enterprise value today is ~$2.6B. A single 600 MW conversion (10% of TVA MOU) prices at $3B — more than the current EV. The math doesn't require the bull case to fully realize; it requires partial conversion.
Hyperscaler dispatchable-power demand is forecast at +160% by 2030. That demand will be priced; SMR's unique offer (24/7, carbon-free, behind-the-fence, sub-acre footprint) has no direct substitute in that demand vector.
3. Budget-cycle reference:
The DOE $800M SMR funding initiative announced 2025-12-02 is the directly applicable budget cycle. The Trump administration's pro-nuclear executive orders (May 2025) plus an explicitly cooperative DOE/NRC create a 2026-2028 deployment funding window. Even before that, the Inflation Reduction Act ITC/PTC stack remains intact for nuclear deployment — the structural subsidy floor under SMR economics did not vanish with administration change.
4. Consolidation / capacity reference:
Fluor's exit is reframed by bears as "informed insider selling" but was structurally driven: Fluor announced a strategic pivot to reimbursable services + a major share buyback program funded specifically by the NuScale stake monetization. This was about Fluor's capital structure, not NuScale's prospects.
The Houston Operations Hub opened late April 2026 places NuScale's customer-facing team in the Texas energy hub — the geographic center of petrochemical decarbonization, data center buildout (Stargate-class projects), and the Permian gas-to-power complex. This is a pre-positioned capacity move ahead of demand crystallization.
5. Falsifiable kill-switch:
If NuScale has not converted any MOU to a binding PPA (TVA, Houston customer, or new) by 2027-06-30, the thesis is broken. That gives 13 months from today. If conversion velocity remains zero through that window, sell — the regulatory moat will be eroded by competitor NRC progress and the dilution overhang will dominate.
Beyond the bar: $1.25B liquidity. NRC moat (the only one). Beta 2.25 with high short interest = explosive upside on any positive event. Stock down 80% from highs, sized for a small position with major asymmetry.
---
Bull Researcher — Round 2
The bear's case is well-rehearsed but conflates execution risk with thesis risk. Let me address the strongest points.
On Fluor's exit: The bear treats this as the "most informed party leaving" — but that ignores why Fluor was a holder in the first place. Fluor's stake originated in 2011 as strategic equity tied to their EPC role on the Carbon Free Power Project. When CFPP was canceled in 2023, Fluor's strategic rationale for holding evaporated. The 2025-2026 sales were a deferred unwind of a stranded position, not a fresh negative diligence call. Compare: Fluor's CFO in their own buyback announcement specifically described the divestment as "capital efficient" pivot to reimbursable services — i.e., balance sheet optimization, not "we know something." If Fluor knew NuScale was impaired, they would have sold at $20+ in late 2024, not at $10-15 in 2025-2026 at the bottom of the drawdown.
On the "low bar" earnings critique: The bear says "a low bar can be missed." True. But the bar has already been reset — Q1 revenue collapse is known and priced. The Apr 7 swing low at $9.16 happened because the market already digested the FEED2 wind-down. The forward question on May 7 is not "did Q1 revenue come in at $5M" — it's "what is the TVA conversion timeline." Management has an obvious incentive to provide bullish color on the call ahead of the share authorization vote.
On hyperscaler-prefer-existing-nuclear: This is true today and was always going to be true. Hyperscalers will buy MWs they can deploy in 2025-2027 from existing fleet (CEG, VST, TLN). They will buy MWs they can deploy in 2028-2032 from SMR. The two categories are sequential, not competitive. NuScale is positioned for the second wave. The market collapsing the time difference is precisely the inefficiency.
On dilution: The bear is right that dilution is severe. But mark-to-market the alternative: a venture-stage company without dilution would be a venture-stage company that dies. The +159% share count was the price of buying $1.25B of liquidity — i.e., the runway to reach FID on TVA and RoPower. If the company converts even one binding contract, today's $11.90 entry price proves to have been the dilution-bottom. If it doesn't, the equity goes to zero either way — so the dilution is not the binary risk.
On HSBC choosing Oklo: HSBC is one bank. Cantor Fitzgerald, B. Riley, and Citigroup all maintain constructive ratings on NuScale per recent estimates aggregations. Single-broker pair-trade calls are not consensus.
On "going concern" language: The bear conveniently cites the DEF 14A's GC-adjacent language but omits that the GC concern is specifically conditional on the share authorization vote failing. The vote is overwhelmingly likely to pass — insider voting + retail bag-holders incentivized to authorize the path to recovery + institutional desire to avoid GC stigma. The 332M → 662M increase will be approved; the GC flag will not trigger.
Strengthening the kill-switch with a tighter trigger: If by Q3 2026 earnings (October 2026) NuScale has not (a) achieved DOE program funding allocation, (b) provided a binding PPA conversion timeline for TVA, AND (c) confirmed RoPower FID for 2027, then the bull thesis is impaired. That brings the falsification window to 5 months from today, not 13.
Bottom line: I am not arguing this is an obvious buy. I am arguing it is a small-position, high-asymmetry trade with a falsifiable thesis and a definable kill-switch. The bear case is the consensus case; the consensus has produced an 80% drawdown; the asymmetry now favors a contrarian position sized appropriately.
Bear Researcher
Bear Researcher — Round 1
The bull is well-prepared, but each pillar collapses under specific scrutiny. Let me go through them.
On the "two dated catalysts":
The May 7 earnings call is a negative catalyst in expectation, not positive. Q1 revenue expected at $4.99M, down 63% YoY as RoPower FEED2 work concluded. The cash-burn run-rate disclosure is the headline risk: Q4 burn was $203.7M; Q1 may be lower ex-ENTRA1, but the trajectory is "burn accelerating, revenue declining." A low bar can also be missed.
DOE $800M is real, but spread across the entire SMR ecosystem — Oklo, X-energy, BWXT, TerraPower, Holtec, Westinghouse. NuScale gets a slice. Even $100M is 12% of one quarter's burn rate. The bull is treating shared funding as if it were proprietary.
On the customer-segment math:
"6 GW * $5,000/kW = $30B" is the kind of math that has destroyed retail capital in pre-revenue stocks for forty years. The TVA agreement is non-binding. Per management's earlier 2025 commitment, the company aimed to have "hard contracts with two or three major customers by end of 2025" — that goal was not met. Conversion velocity is zero. Multiplying zero by a TAM does not produce revenue.
The hyperscaler demand point ignores that hyperscalers are buying existing nuclear — Microsoft restarted Three Mile Island; Amazon partnered with Talen; Constellation's Crane Clean Energy Center deal is operational. The capital flowing into "nuclear for AI" is going to operating nuclear, not pre-commercial SMR. Constellation is up; NuScale is down 80%. The market is voting.
On budget cycle:
DOE funding has historically been a disappointment for NuScale specifically. The Carbon Free Power Project in Idaho was DOE-supported and was canceled in November 2023 due to cost overruns. Recurrence risk is high — and the cancellation is a precedent the market remembers very clearly.
On consolidation / capacity:
The Fluor exit framing as "Fluor's capital structure issue" is wishful thinking. Fluor was the prime engineering partner, the manufacturing partner, and the largest equity holder. They were literally inside the diligence room for fifteen years. They sold ~40M shares for ~$2.43B. When the most-informed party fully exits, "they had their own reasons" is not a sufficient bull rebuttal — it's the bear thesis itself.
The Houston Operations Hub is a real-estate event. No customer is signed. Putting an office in Houston does not create demand; it is a sales-cycle leading indicator with a 3-5 year lag.
On the kill-switch:
The 2027-06-30 trigger is well-formed but generous. By then, the company will have raised additional capital twice over (332M → 662M share authorization request makes this explicit). Even if a binding PPA is signed in 2027, the per-share economics will have been further diluted by 30-50%.
My direct counters:
Customer concentration (current & forward) = catastrophic. 92% historical Fluor revenue → Fluor exited. ENTRA1 is a cost, not a customer. TVA is non-binding. RoPower FID slipped to 2027.
Cash burn trajectory is accelerating. $109M FY24 burn → $460M FY25 burn → quadrupled. Even ex-ENTRA1, the core business is running through ~$200M/year and revenue is going down.
Dilution math is brutal. 122.8M → 318.5M shares (+159%) in 12 months. Authorized share request goes 332M → 662M. Per-share economics on any future success scenario are being shredded as we wait for the success scenario.
Competitive reality: HSBC chose Oklo. ARK is rotating to X-energy private. Constellation's stock is up. The market is voting against this specific name.
Going-concern adjacency: The DEF 14A explicitly references "substantial doubt about its ability to continue as a going concern" if the share authorization vote fails. This is in the company's own disclosure.
My recommendation: Avoid. There are better ways to express the nuclear thesis (CEG, BWXT, even OKLO with its tighter cap structure). NuScale is a venture-stage company in a public-market wrapper — wrong instrument for the thesis.
---
Bear Researcher — Round 2
The bull continues to do quality work, but the rebuttal contains structural concessions that prove my case.
The bull just admitted the kill-switch is 5 months, not 13. That tightening is intellectually honest, but it is also the bear's timeline. By October 2026, we will know whether (a), (b), and (c) have hit. Today, the expected outcome on (a) is "yes, $50-100M slice of the $800M pool" — small impact; on (b) is "still non-binding, soft commentary on conversion path" — neutral-to-negative; on (c) is "still slipping" — already announced as late 2026 / early 2027. Mathematically, the bull's own kill-switch is more likely to trigger than not. A bull thesis whose falsification probability is >50% on the bull's own timeline is not a buy thesis — it is a coin flip.
On Fluor's "stranded position": This argument concedes too much. CFPP was canceled in 2023; Fluor held through 2024 and into 2025 — that is two full years during which they could have unwound. They unwound now, in 2025-2026, after seeing the 2025 burn rate and the dilution trajectory. The "stranded position" framing requires us to believe that a sophisticated EPC firm with twenty years of nuclear-project diligence experience didn't realize their position was strategically dead until 2025. That is not credible.
On hyperscaler-sequential: This is the bull's most interesting argument and also the one with the longest tail. Even if hyperscalers do shift to SMR in 2028-2032, why would they choose NuScale specifically? By 2028, expect Oklo, X-energy, BWXT, and TerraPower to have NRC design approvals or to be very close. The NRC moat is a 2026 moat, not a 2028 moat. Hyperscalers will choose the lowest-LCOE, fastest-deployable design — and NuScale's first-of-a-kind has not demonstrated cost or schedule yet.
On dilution-as-runway: I agree dilution was necessary. The bear case is not "they shouldn't have raised capital"; the bear case is "the capital was raised at a price that reflects severe execution risk, and the per-share economics on success scenarios have been impaired." A binding TVA PPA in 2027 worth $3B at the corporate level is, post-dilution, worth ~$5-7/share to current holders, depending on additional dilution between now and then. That is not a 10x scenario from $11.90 — it's a flat-to-down scenario.
On the "small-position high-asymmetry" framing: This is the bull's intellectually weakest move because it shifts the question from "is this a good investment" to "how do you size a bad investment." If a position requires sub-1% sizing to be palatable, that is a bear signal, not a bull signal. The asymmetry is real but not in the bull's favor: the downside is -50 to -100% in 12 months (further dilution + thesis failure); the upside is +50 to +200% in 24 months (binding contracts + DOE funding). Probability-weighted, this is a negative expected value trade.
Strongest single bear point I want to leave on the table: Revenue is going DOWN, not up, in the year that is supposed to be the inflection. FY2025 revenue fell 15% YoY. Q1 2026 expected to fall 63% YoY. The company is supposedly converting from FEED-stage engineering to commercial deployment — but the financial metric that should increase in that conversion (revenue from new contracts) is decreasing, while the metric that should be under control (cash burn) is quadrupling. That is the opposite of the inflection narrative. Until that fact is contradicted by data, the bull thesis is fighting the numbers.
My final recommendation stands: Avoid. If thematic exposure is required, get it via Constellation, Vistra, or BWXT. If pure-SMR exposure is required, get it via Oklo's tighter capital structure or an SMR ETF. NuScale specifically is the worst expression of a good thesis.
Risk Assessment
Aggressive
Aggressive Risk Analyst — Round 1
The trader's Hold is cautious to a fault. We have a stock down 80% from highs, sitting on $1.25B of liquidity, holding the only NRC-approved SMR design in the United States, two days from a low-bar earnings call, in a sector backed by explicit Trump-administration policy support and an $800M DOE funding pool. This is the textbook setup for a contrarian long.
The trader's stop at $9.00 is reasonable, but the no entry posture is leaving a 30% upside opportunity on the table if the May 7 print is even neutral-with-positive-color. The April 22 single-day +16% spike on no news told us how shorts are positioned. A short squeeze on any positive update could clip $14-15 in 48 hours. We can size at 0.75% of portfolio with a $9.00 stop — risk is bounded at ~25% of position, or 0.19% of portfolio. The asymmetry is genuinely 4:1 or better.
I'd add: the bear's "revenue is going down" critique is selectively framed. Pre-commercial venture companies always have non-monotone revenue trajectories during the FEED-to-deployment transition. Lockheed had this in F-22 program history. Tesla had this in 2008-2010. SpaceX had this through 2008. Pre-commercial revenue declines do not falsify a deployment thesis — they precede it. To the conservative analyst: tell me which pre-commercial nuclear company has monotonically growing revenue right now. None of them. NuScale is being held to a public-equity standard while operating on a venture-equity timeline.
My recommendation: Upgrade trader's plan to a small Overweight bias post-May-7 if the print is anything other than catastrophic. Even on a print miss, the share authorization vote gives a separate catalyst path.
---
Aggressive Risk Analyst — Round 2
The conservative's reference-class list is rhetorically effective but selection-biased. Plug Power, Workhorse, Nikola, Lordstown, Lucid — that's a list of failures. The honest reference class also includes First Solar, which traded at $24 in 2012 after an 80% drawdown and went to $300+. Tesla in 2010 at $4 (split-adjusted) trading on $0 commercial volume. Enphase in 2014. MicroStrategy in 2019. Survivorship bias in either direction destroys analytical rigor; the right approach is a probability-weighted base rate.
To the neutral: I largely agree with your synthesis. The half-position critique is fair — binary catalyst trades should be all-or-nothing relative to the chosen sizing tier.
But I want to push back on one thing the conservative said: "sell into May 7 strength above $13." This is the kind of prescription that generates underperformance in catalyst trades. If May 7 produces a print + commentary that genuinely re-rates the stock to $13-14, that is evidence the thesis is working, not evidence to exit. Selling into confirmation of your thesis is not risk management; it is cognitive miscalibration.
My final position: Slight overweight relative to the Hold. New positions 0.5% post-confirmation; existing positions held with $9.50 stop per neutral's revision.
Conservative
Conservative Risk Analyst — Round 1
The aggressive analyst's argument is structurally identical to every venture-stage public-equity bull case from 2021. It almost always ends badly.
Let's address risk concretely. Beta is 2.25 — meaning a -10% market move produces an expected -22.5% NuScale move, before idiosyncratic risk. ATR is 9.3% of share price — this stock can drop 15-20% in two sessions on routine news. A "0.75% portfolio sizing" feels like de minimis exposure but it is not de minimis when the position can be -50% in 48 hours on a Q1 miss + cash-burn shock.
The aggressive's "Lockheed F-22 / Tesla 2008 / SpaceX" comparison is the wrong reference class. The right reference class is publicly-traded pre-commercial energy/industrial companies that needed continuous capital raises: Plug Power 2000-2003, Solyndra (privately-held but instructive), Workhorse Group, Lordstown Motors, Nikola, Lucid Motors. The base rate for that reference class is severe and persistent dilution, multiple 70-80% drawdowns from interim highs, and frequent equity destruction even when the underlying technology is real.
NuScale's specific risks at this exact moment:
Going-concern proximity: The DEF 14A literally references substantial doubt language conditional on the authorization vote.
Burn rate run-rate: 18 months of runway at Q4 pace. The next dilution is mathematically guaranteed within 12-15 months.
Catalyst-cluster concentration: May 7 earnings, share authorization vote, DOE allocation, TVA conversion, and RoPower FID are all clustered in the next 6 months. Each catalyst has a meaningful probability of disappointing — cumulative probability of all five clearing is low.
My recommendation: Sell into any May 7 earnings strength above $13.00. New positions: zero. Existing positions: trim to half-weight on any pop. This is not a "hold and hope" — it is "manage the exit on strength.
---
Conservative Risk Analyst — Round 2
The aggressive's First Solar / Tesla / Enphase / MicroStrategy list is a survivorship-biased counter-counter; one survivor list does not refute one failure list. The base rate for the broader category remains poor.
To the neutral: I accept the half-position critique. The cleaner formulation is:
For existing positions over 1.5%: trim to 0.75-1.0% pre-May-7. Reduce risk before the binary.
For existing positions under 1.0%: hold with $9.50 stop.
For zero-position accounts: stay zero until all three of neutral's three conditions clear.
The aggressive's "selling into confirmation" critique is interesting but applies to longer-duration confirmations. Selling into a single-event +20% pop on a stock with 9.3% ATR is not selling into confirmation — it is taking profits on volatility. The genuine confirmation event is binding contract conversion, which is a 6-12 month event, not a May 7 single-day event.
My final position: Hold the existing-position bias unchanged; zero new positions until durable confirmation (binding TVA PPA), which is a 2026-Q3-or-later event.
Neutral
Neutral Risk Analyst — Round 1
Both prior analysts have correctly identified real risk vectors. The aggressive is right that the asymmetry is genuinely large; the conservative is right that the reference class for similar setups is unkind.
Let me focus on what the trader's Hold actually means in execution:
For existing holders: Conservative is essentially correct that trimming on strength is the right discipline. But "trim to half-weight" is too aggressive on a binary catalyst — a binary should be either a full position with stop or a no position, not a half position. Half-positions in catalyst trades produce the worst of both worlds (regret risk in both directions).
For new entry: The trader's "no new entry pre-May-7" is correct. Aggressive's case for a small post-May-7 entry is reasonable only if (a) the print is genuinely neutral or better AND (b) management provides a binding-contract conversion timeline AND (c) the share authorization path is clearly articulated. Two of three is not enough — all three is the entry trigger.
On stops: Both prior analysts agree on $9.00 as the technical line. I'd actually argue for $9.50 (above the swing low of $9.16) to reduce stop-running risk. The 35-cent buffer costs little but gives material stop-execution improvement on this particular stock's volatility profile.
On sizing for new positions: Aggressive at 0.75%, conservative at 0%. The right answer post-May-7 is 0.25-0.50% for accounts with appropriate mandate — small enough to absorb -100% on the position with portfolio-level resilience, large enough to matter on a 3-5x outcome.
On the reference-class debate: Both analysts cited valid analogues. The truthful synthesis is: the base rate for publicly-traded pre-commercial industrial companies is poor, but NuScale has one genuinely differentiating asset (NRC approval) that is not present in most analogues. The base rate adjusts modestly toward "less bad than typical" — but only modestly. The aggressive overweights the differentiator; the conservative underweights it.
My recommendation: Endorse the trader's Hold with conservative-sizing tilt for new positions and neutral-aggressive tilt on retention of existing positions. Concretely: don't sell the position you have unless $9.00 breaks; don't add unless May 7 is constructive on all three of (Q1 print, TVA conversion timeline, authorization clarity).
---
Neutral Risk Analyst — Round 2
Convergence is forming. Let me synthesize the final position both analysts can sign onto:
Trader's Hold stands, with the following execution refinements:
Existing positions: Retain. Stop at $9.50 (above swing low for stop-running buffer). Trim to maximum 1.0% portfolio weight pre-May-7 if currently sized larger.
New positions: Zero pre-May-7. Post-May-7, eligible for 0.25-0.50% sizing only if all three of the following clear: (a) Q1 print not catastrophic (i.e., revenue and burn within 15% of consensus), (b) TVA conversion timeline articulated with a 2026-2027 binding milestone, (c) share authorization path clarified with a near-term vote schedule.
Re-evaluation triggers:
Upgrade to Overweight: any binding TVA PPA conversion announcement OR DOE allocation > $100M OR RoPower FID confirmation
Downgrade to Underweight: $9.00 close OR another share authorization expansion request beyond 662M OR Q1 cash burn exceeds Q4 ex-ENTRA1 baseline by >25%
Time horizon for re-evaluation: October 2026 (Q3 earnings) is the bull's own kill-switch. If Hold has not been upgraded by October 2026, the position should be exited.
The Hold here is genuinely a wait-state with defined edges, not a hope-and-hold.
Technical Read
Price · 50/200 SMA · Bollinger 20·2 envelope. Levels rail at right is colour-coded — green = support, red = resistance, taupe = neutral.
Indicator Snapshot
RSI(14)
52.3
NEUTRALMid-band
MACD(12·26·9)
+0.19
BULLISHAbove signal
MA stack
+4.4%
WARNINGMixed alignment
Volume
1.40x
WARNINGVolume expanding
Realized vol
119%
BEARISHVery elevated
Bollinger %B
64%
NEUTRALMid-band
Price $12.18 sits between $11.67 support and $12.79 resistance; MACD above signal.
Analyst Reports
Market Analysis
NuScale Power (NYSE: SMR) closed at $11.90 on 2026-05-05, sitting deep in a multi-month basing pattern after an 80% drawdown from the 52-week high of $57.42. The stock has rallied approximately 30% off the April 7 swing low of $9.16, with the technical structure now showing tentative recovery — but only relative to a savaged baseline. The 200 SMA sits at $25.51, a ceiling so distant it functions more as a memorial than a target.
Trend Structure
The trend hierarchy is mixed, leaning short-term constructive against a clearly broken long-term structure:
10 EMA ($12.04): Price closed marginally below; short-term momentum has stalled after the April 22 spike
50 SMA ($11.67): Price holding fractionally above; bulls defending a key dynamic support
200 SMA ($25.51): Price 53% below — definitive long-term downtrend
VWMA ($12.26): Volume-weighted average above price, indicating distribution pressure dominated the rally
The 50 SMA is itself rolling lower (from $13.74 in early April to $11.67 today), so even short-term holders are in the red on average. Holding the 50 SMA in the next 5–10 sessions is the line that separates "consolidation before next leg" from "false rally before retest of $9 lows."
Momentum
The MACD turned positive on April 28 after 3+ weeks below zero — a real bullish crossover (MACD: +0.19, Signal: +0.08, Histogram: +0.11), but with the histogram contracting in the last 2 sessions, suggesting initial impulse is fading. RSI (52.3) is sterile neutral — neither bulls nor bears have control. The MACD reset from ~ -1.0 (Apr 7) to today's positive print is the cleanest argument for "the bottom may be in" on a pure technical basis. Whether it holds depends entirely on May 7 earnings.
Volatility & Volume
ATR ($1.11) translates to about 9.3% of share price — exceptional volatility, consistent with the 2.25 beta noted in fundamentals. Position sizing must be cut accordingly. Bollinger upper ($14.16) / lower ($8.68) define a $5.48 envelope — anyone trading this without respect for the band width is volunteering for a stop-out. Volume on the April 22 push (55.2M shares) was 3-4x average — that's the high-conviction print to watch; today's 13.9M is light, the consolidation is on declining volume which can resolve either way.
The chart shows a stock attempting a basing pattern, with a single bullish input (MACD crossover) facing two bearish setups (price below VWMA, far below 200 SMA). May 7 earnings is the binary. Holding $11.67 keeps the rally alive; losing $10.83 reopens $9.16; losing $9.16 puts $8.85 in play and probably below.
Indicator
Value
Read
Last Close
$11.90
-1.7% intraday vs prior close
50 SMA
$11.67
Holding — barely
200 SMA
$25.51
-53% — long-term bear
RSI
52.3
Neutral
MACD
+0.19 (signal +0.08)
Fresh bullish crossover, fading
ATR
$1.11 (9.3%)
Extreme volatility
Beta
2.25
High systemic sensitivity
52w Range
$8.85 – $57.42
-79% from highs
Social Sentiment
Retail and finance-media sentiment on NuScale Power is polarized to the point of caricature. The same week produced "$1.5 trillion opportunity ahead" headlines (Motley Fool, April 30) and "down 70% in 6 months" headlines (Motley Fool, April 23) — about the same stock, with no new fundamental information between. This is what a story stock looks like when narrative hasn't yet locked.
Bull Camp (LOUD)
The bull narrative converges on three claims, each of them genuinely interesting and each of them genuinely speculative:
AI / data-center power demand: SMR is positioned as one of the few investable pure-plays on dispatchable, carbon-free baseload power for hyperscaler buildout. The Houston Operations Hub (opened late April 2026) was explicitly framed as "closer to petrochemical, data center, and grid-focused customers" — narrative-engineered for the AI moment.
NRC moat: NuScale remains the only SMR design with U.S. Nuclear Regulatory Commission approval (77 MWe module). The bull case argues this is a 5-year regulatory head start over Oklo, X-energy, and TerraPower.
Cathie Wood / ARK exposure: ARK funds bought into the nuclear theme; Reddit and Stocktwits treat ARK as a legitimacy signal.
Bull sentiment on Substack ("MTC's Substack" was cited by Insider Monkey on April 21) and Twitter/X is intense, with the dominant frame being "speculative pure-play" — note: the bulls themselves use the word "speculative," not the bears.
Bear Camp (LOUDER, with the data)
The bear narrative runs on harder math:
Fluor exit: Fluor — original 2011 backer, 92% revenue customer, and once 40-million-share holder — has fully sold its position, generating roughly $2.43B in proceeds since September 2025. Insider selling at this scale by an informed original investor is a four-alarm signal, not background noise.
Dilution at scale: Shares outstanding went from 122.8M (Dec 2024) to 318.5M (Dec 2025) — a +159% dilution event in 12 months. The company has now asked stockholders to authorize 662M shares (from 332M), explicitly because they're out of authorized headroom.
HSBC chose Oklo over NuScale in their nuclear pair-trade pick (TipRanks, late April).
Comparison shaming: The Motley Fool ran two head-to-head Oklo-vs-NuScale pieces in 48 hours, plus a dedicated "Oklo or NuScale: Both Are Falling — Which One Is Worth Buying" piece. The comparison is being framed by the media as a binary — and NuScale is consistently the loser of the framing.
Sentiment Vectors
Reddit r/NuScale, r/wallstreetbets: Mixed. Bag-holders defending the thesis; new buyers calling the bottom around $9.
Twitter/X: ENTRA1 / TVA agreement dominates the bull narrative; the $507M ENTRA1 milestone payment dominates the bear narrative.
Yahoo Finance comments: Heavy retail engagement, dominated by "cheap nuclear" buyers; relatively few institutional voices.
Analyst-side: Simply Wall St cut fair value modestly from $20.73 to $20.42 (-1.5%), still implying ~70% upside from spot — but on a model that has been wrong every quarter for two years.
Q1 Earnings Setup (May 7) — The Sentiment Catalyst
Going into the Q1 print, the bar is historically low: consensus -$0.11 EPS, ~$4.99M revenue (vs. $13.4M Q1 2025 — a 63% YoY decline as RoPower FEED2 work concluded). The combination of low expectations + heavily shorted, beaten-down small cap + a real catalyst can produce a 20%+ move either direction. The April 22 +16% spike on no real news is a tell that shorts are positioned and dry powder is sitting on the sidelines.
Bottom Line
Sentiment is bifurcated and emotionally engaged — the worst possible setup for a binary catalyst event. The "story" is winning the narrative war (data center / AI / nuclear renaissance) but the "scoreboard" is winning the financial war (cash burn, dilution, no commercial revenue). Q1 earnings on May 7 will not resolve this — only a hard customer contract will, and that's a 2027 event at the earliest.
Sentiment Vector
Skew
Conviction
Retail (Reddit/Stocktwits)
Bullish
High emotion, low data
Finance-media (Motley Fool)
Mixed → Bearish trend
Comparison framing favors Oklo
Sell-side analysts
Modestly bullish
Targets cut quietly
Institutional (Fluor)
Hard bearish
Fully exited; $2.43B realized
ARK / thematic
Bullish
Theme-driven, not name-specific
News & Macro
The macro and sector backdrop for NuScale is genuinely the best it has ever been — which makes the equity's 80% drawdown all the more diagnostic of company-specific issues.
Macro / Sector Tailwinds (Real)
Trump administration nuclear policy: The current administration has explicitly prioritized nuclear restart and SMR deployment. The U.S. Department of Energy on December 2, 2025 announced an $800 million funding initiative to accelerate SMR deployment.
Hyperscaler demand: Microsoft (Three Mile Island restart), Amazon (Talen Energy partnership), Google, Meta, and Oracle have all signed dispatchable-power deals with nuclear operators in the last 18 months. Data center power demand is forecast to grow 160% by 2030 (industry consensus).
Grid reliability politics: NERC, FERC, and state PUCs are increasingly focused on baseload retirement risk, which structurally favors SMR.
Defense / DoD interest: Project Pele and analogous initiatives have created a parallel defense-microreactor demand channel.
Foreign policy: U.S. SMR exports — particularly the Romania (RoPower) project — are a stated State Department priority for energy-security counterweight to Russia/China.
Macro / Sector Headwinds (Also Real)
Renewables + storage cost curve: Utility-scale solar + battery is now $30-50/MWh delivered in many markets. SMR economics depend on a >$80/MWh assumption that is not yet validated at scale anywhere.
Interest rates: SMR is a long-duration capital-intensive cash flow — every 100 bps of rate is meaningful to NPV. The 10-yr has compressed but remains elevated relative to 2020-21 financing assumptions.
Permitting / FOAK risk: Even with NRC design approval, site-specific licensing and first-of-a-kind construction risk has not been demonstrated commercially in the U.S. since the 1990s.
Sector Comparables
The competitive landscape has hardened materially in 2026:
Oklo (OKLO): Pre-revenue, but with momentum — HSBC pair-trade pick over NuScale; partnerships with DOE Idaho National Lab.
X-energy (private, ARK invested per IBD coverage): HTR-PM-style design; Cathie Wood's nuclear position appears to be migrating from SMR-listed names toward private positions like X-energy via the ARK Venture Fund.
Constellation (CEG), Vistra (VST), Talen (TLN): The "buy existing nuclear" trade has worked — these names are up materially while NuScale is down. Capital is preferring proven megawatts.
Single-Name Catalyst Calendar (Forward 90 days)
Date
Event
Impact
2026-05-07
Q1 2026 earnings call
High — consensus -$0.11 EPS, ~$5M revenue; cash burn run-rate is the key metric
2026 mid-year
DOE $800M SMR program initial allocations
Medium — NuScale should compete for funding given NRC approval
High — non-binding → binding conversion is the single biggest re-rating event
2026-2027
Romania (RoPower) Final Investment Decision
High — currently pushed to "late 2026 / early 2027"
2026 ongoing
DEF 14A vote on 332M → 662M share authorization increase
Medium — failure to pass = financing crisis; passage = continued dilution overhang
Bottom Line
NuScale is operating into the strongest macro tailwind for nuclear in 30 years, and is still down 80% from highs. That's the puzzle the equity has to solve. The sector tailwind is real but is being captured by operating nuclear (Constellation, Vistra) and by competing SMR designs (Oklo, X-energy private). NuScale's NRC moat is real but expires the moment a competitor secures NRC approval — likely 2027-2029. The window to convert MOUs to binding contracts is shorter than the bulls assume.
Macro Vector
Direction
Capture by SMR?
AI / data-center demand
+++
Limited — hyperscalers prefer existing nuclear
Trump pro-nuclear policy
++
Industry-wide; not name-specific
DOE $800M SMR funding
++
NuScale should win share given NRC approval
Higher rates
--
Hurts long-duration capex names like SMR
Solar + storage cost curve
--
Sets ceiling on SMR LCOE
Foreign policy export push
+
RoPower delays mute capture
Fundamental Analysis
NuScale Power's fundamentals tell a clear story: this is a pre-revenue venture-stage company trading on a public market, financed by serial dilution, holding a unique regulatory asset (NRC design approval), and operating with the financial profile of a Series D startup that hasn't yet reached product-market fit. The market cap of $3.85B is essentially a bet on the optionality, not on the income statement.
Financial Position — Strong Liquidity, Catastrophic Burn
Metric
Value
Context
Cash + Equivalents (Q4 2025)
$836.4M
Up from $401.6M Q4 2024
Short-Term Investments (Q4 2025)
$417.8M
Up from $40M Q4 2024
Total Liquid Position
$1.25B
Cash war chest
Total Liabilities
$299M
Mostly current — vendor / accrued
Stockholders Equity
$1.17B
After $1.3B 2025 capital raise
Current Ratio
4.30x
Strong — but burn-rate adjusted
Working Capital
$976.5M
Up from $381.4M YoY
The cash position looks comfortable; the burn rate makes it less so. Operating cash burn was -$203.7M in Q4 2025 alone — quadrupled from the $26.5M Q4 2024 run-rate. Annualizing the Q4 pace gives an ~$815M/yr burn against $1.25B liquidity = roughly 18 months of runway before the next capital raise. The Q4 spike was driven by the $507.4M ENTRA1 Milestone Contribution 1 payment recognized under the Partnership Milestone Agreement — a one-time event, but a real cash drain.
Adjusted for the ENTRA1 one-off, "core" burn is in the $40-60M/quarter range, giving 5-6 years of runway. The bear interpretation: ENTRA1 is the first of N milestone payments, each of which compresses runway. The bull interpretation: the milestone is paid, the partnership is locked, future milestones are tied to revenue events that bring incoming cash.
Income Statement — No Revenue Story
Line Item
FY25
FY24
Change
Revenue
$31.5M
$37.0M
-15% YoY
Gross Profit
$11.4M
n/a
36% gross margin
Operating Loss
-$689.6M
-$76.5M
9x deterioration
Net Loss
-$355.8M (TTM) / -$664.5M (FY25 incl. minority)
-$348.4M
Worse
EBITDA
-$181M (TTM)
n/a
Negative
Op Cash Flow
-$459.6M
-$108.7M
4.2x worse
Revenue is going the wrong way. RoPower FEED2 work concluded in Q4 2025 — Q1 2026 consensus is $4.99M vs $13.4M Q1 2025 (-63% YoY). The company has effectively no revenue base because its product isn't built yet. Q3 2025 SG&A spiked to $519M almost entirely on the ENTRA1 milestone.
Capital Structure — The Dilution Machine
Metric
Q4 2024
Q4 2025
Change
Shares Outstanding
122.8M
318.5M
+159% (1 year)
Capital Stock Issued
$995.7M
$1,901.7M
+$906M paid-in capital
Issuance of Capital Stock
$100.8M (Q4)
$737.4M (Q4)
7.3x
Three rounds of dilution in 2025 ($99.8M Q1, $462.6M Q3, $737.4M Q4) raised cumulative gross proceeds of $1.3B. The company has now requested stockholder authorization to expand authorized shares from 332M to 662M — explicitly because, per the DEF 14A, all 332M existing authorized Class A shares are issued or reserved. This is the company telling you another dilution is coming.
Customer Concentration — Now a Past-Tense Problem
92% of historical revenue came from Fluor Corporation, NuScale's original 2011 backer and parent. Fluor has now fully exited, divesting roughly 40M shares for ~$2.43B in cumulative proceeds since September 2025. Going forward, customer concentration risks shift to:
ENTRA1: Exclusive commercialization partner; the $507M PMA milestone is a structural cost
Tennessee Valley Authority: Non-binding MOU for up to 6 GW (the "headline" number)
Romania (RoPower): FID delayed to late 2026 / early 2027
Item 1A Risk Factors (10-K filed 2026-02-26)
The 10-K's risk-factor stack (per SEC EDGAR / 10-K disclosures):
Going-concern adjacent language: The DEF 14A authorization request notes that without expanded share authority, the company's ability to issue equity for financing "may create substantial doubt about its ability to continue as a going concern."
Customer concentration: 92% Fluor dependency historically; counterparty risk magnified now that Fluor has exited.
Single design / single product: Entire commercial thesis tied to the 77 MWe NuScale Power Module. No diversification.
Long lead-time component obligations: ~$99M in long-lead material commitments could strain cash if RoPower or other projects slip.
Regulatory / NRC site licensing: Design approval ≠ site approval. Each deployment requires separate licensing.
First-of-a-kind construction: No SMR has been built and operated commercially in the U.S. Cost overrun risk on FOAK is historically 50-100%.
Valuation — A Story Stock Multiple
Metric
Value
Read
Market Cap
$3.85B
Bet on optionality
Forward P/E
-27.7
Negative — meaningless
Price/Book
3.24x
Premium to BV
EV
~$2.6B (mkt cap less net cash)
Cleaner read
TTM Revenue
$31.5M
EV/Sales >80x — venture multiple
Forward Revenue (~$5M Q1 annualized $20M)
extreme
Unmoored from cash flow
There is no fundamental valuation case in 2026 — only an option valuation case. Treating NuScale as a real option on commercial SMR deployment with a 5-7 year horizon and a ~25% probability of a 10x outcome implies a $2-4B fair value range — which is roughly where the stock trades. The market is pricing a successful pivot, not a failure.
Bottom Line
The fundamentals are coherent for a pre-commercial venture stage: strong cash position, accelerating burn, terrible income statement, severe dilution, single-asset concentration, and one genuinely valuable regulatory asset (NRC approval). The financial picture supports neither the bull's "10x upside" nor the bear's "going to zero" — it supports "this is binary on customer conversion in the next 18 months."
Fundamental Vector
Read
Liquidity
Strong ($1.25B), but burn is accelerating
Burn rate
-$815M annualized at Q4 pace; ~18 months runway
Revenue trajectory
DECLINING (-15% YoY, -63% YoY for Q1 expected)
Dilution
+159% in 12 months; another raise pending authorization
Customer base
Fluor exited; ENTRA1/TVA non-binding
Going concern signal
Implied by DEF 14A language
Valuation
Story stock; >80x EV/Sales
Catalyst conversion risk
RoPower FID slipped to 2026/27; TVA still non-binding
Risk Factors (Item 1A)
NuScale Power 10-K Item 1A Risk Factors
Filing: Form 10-K for fiscal year ended December 31, 2025
Filed: February 26, 2026
Source: SEC EDGAR (https://www.sec.gov/cgi-bin/browse-edgar?action=getcompany&CIK=0001822966&type=10-K)
Top Risks (paraphrased / synthesized from 10-K and DEF 14A disclosures):
Going-Concern Capacity: As of November 17, 2025, all 332M authorized Class A shares were issued or reserved. The company's DEF 14A explicitly notes the limitation "may create substantial doubt about its ability to continue as a going concern" if shareholders do not authorize an increase to 662M shares. Material if vote fails.
Customer Concentration (Historic & Forward): Approximately 92% of historical revenue derived from Fluor Corporation. Fluor has now fully divested its ~40M share position. Forward revenue depends on ENTRA1 commercialization partnership and Tennessee Valley Authority MOU (non-binding) — both unproven counterparties at scale.
Single-Product / Single-Design Risk: 100% of commercial thesis is tied to the 77 MWe NuScale Power Module. Any technical, regulatory, or competitive obsolescence event creates an existential exposure with no diversification offset.
First-of-a-Kind (FOAK) Construction Risk: No commercial SMR has been built or operated in the United States since the 1980s-era research reactors. Historical FOAK nuclear cost overruns are 50-100%. The Carbon Free Power Project (Idaho) was canceled in November 2023 specifically due to cost escalation — a precedent the market remembers.
Long-Lead Material Obligations: Approximately $99M in long-lead component commitments — primarily forging and large fabrication — could strain cash if RoPower FID slips further or if TVA conversion delays.
Regulatory Site-Licensing Risk: NRC design approval (granted) ≠ NRC site approval (separately required for each deployment, 18-36 months per project). Site licensing creates project-level execution risk distinct from the design-approval moat.
Equity-Linked Compensation & Dilution: The 2025 share count increase (122.8M → 318.5M, +159%) plus the request for an additional 330M authorized shares creates structural overhang on per-share economics. ATM sales at depressed prices have a self-reinforcing dilutive effect on rallies.