Environment & Energy
Should nuclear power be expanded to fight climate change?
Clean baseload or waste legacy?
ANALYSISApril 13, 2026
More than 30 countries endorsed a goal at COP28 to triple global nuclear capacity by 2050, and the World Nuclear Outlook Report 2025 projects that 1,428 GWe of global nuclear capacity is achievable by mid-century. Both the Biden and Trump administrations have independently set targets to expand U.S. nuclear capacity, with goals ranging from 200 GW to 400 GW by 2050. The debate over whether nuclear power should be a central tool in fighting climate change is now a top-tier policy question at international climate forums including COP30.
If nuclear is the only carbon-free source that can reliably replace fossil fuels at scale, is opposing it on safety grounds an environmental position — or an environmental liability? And if we build it, who bears the risk when something goes wrong?
Not familiar with this story? Get context →
Germany's nuclear exit as cautionary tale
Conservative
Germany's Energiewende is the clearest natural experiment we have: after spending over $500 billion transitioning away from nuclear, Germany now emits roughly twice the carbon per kilowatt-hour as France, which kept its reactors running. That is not a modeling projection — it is the observed outcome. When the policy produces more coal combustion in the name of clean energy, the policy has failed on its own terms.
Liberal
The France-Germany comparison is real, but it proves less than you're claiming. France built its nuclear fleet under a centralized state monopoly over decades with no meaningful cost competition — it is not a replicable template for 2025 procurement. Germany's error was closing reactors before replacement capacity was ready, not closing them in principle.
Conservative
If the lesson is 'don't close reactors before replacements are ready,' that is an argument for slower nuclear phase-outs, not faster ones — and it directly undermines the urgency argument for shifting capital away from nuclear now.
Liberal
Agreed: don't phase out without replacements. That's the liberal position too — keep existing reactors running, build new ones, and deploy renewables in parallel. The disagreement is about sequencing and capital allocation, not whether nuclear belongs in the mix.
Nuclear's baseload role versus renewable flexibility
Conservative
Dispatchable baseload power running at 90%-plus capacity factors regardless of weather conditions is not a preference — it is a structural requirement for industrial civilization. The liberal counterargument that sufficient storage could theoretically replace this function is correct on the word 'theoretically.' We do not have grid-scale storage deployed anywhere near the level that argument requires.
Liberal
The word 'theoretically' is doing a lot of work there. Battery storage costs have fallen roughly 90% over the last decade and continue dropping — the same cost trajectory that made solar and wind competitive. Dismissing storage as theoretical in 2025 applies the same logic that would have dismissed offshore wind as theoretical in 2005.
Conservative
Cost trajectories are not deployment realities. The U.S. currently has roughly 26 GW of battery storage against over 1,100 GW of total generating capacity — we are orders of magnitude away from storage covering the baseload gap, and that buildout has its own land, materials, and timeline constraints.
Liberal
Which is exactly why no serious liberal analyst is proposing storage-only grids by 2035 — the argument is for parallel deployment, not replacement sequencing. Nuclear buys time while storage scales; the two aren't competitors.
Construction timelines and the climate window
Conservative
The 14.5-year average build time is a real problem — not a talking point. But the honest diagnosis is that this timeline is inflated by regulatory process, not physics. South Korea completes proven reactor designs in roughly seven years. If the NRC procedural scaffolding is what doubles the timeline, then reforming that process is the climate intervention, not abandoning the technology.
Liberal
South Korea's seven-year timeline is built on continuous serial construction using standardized designs, a national utility with institutional memory, and a regulatory culture that evolved alongside that program. The U.S. hasn't completed a reactor in decades — we would be rebuilding that institutional capacity from scratch, which adds time regardless of what the NRC rulebook says.
Conservative
That is a genuine constraint, but it is an argument for starting now and accepting the learning curve, not for waiting until renewables scale up. Every year we delay the institutional rebuild is a year the timeline stays long.
Liberal
Agreed on starting now — but 'starting now' means also maximizing deployment of technologies that can come online in two to four years while the nuclear rebuild takes shape. The climate math requires both, not a sequential bet on institutional recovery.
True cost comparison: nuclear versus renewables
Conservative
Levelized cost figures showing renewables cheaper than nuclear are systematically incomplete because they exclude the grid-balancing costs that intermittent generation imposes on everything else. The Rockefeller Foundation study found that pairing nuclear with renewables could lower total system costs by up to 31%. The honest question is what a reliable, low-carbon grid costs end-to-end — and when you ask that question, nuclear's economics look considerably better.
Liberal
The grid-balancing cost argument is legitimate and underappreciated. But it cuts both ways: if the system cost of intermittency is the real metric, then nuclear's 8.1 cents per kWh should also include the cost of backup capacity for the periods reactors are offline for refueling or unplanned outages — which existing LCOE figures equally ignore.
Conservative
Nuclear's capacity factor of 90%-plus means it is offline roughly 10% of the time versus solar's average 25% capacity factor in good regions — the backup cost asymmetry is real and substantial, not a wash.
Liberal
Capacity factor isn't the only variable — solar outages are predictable and distributed, nuclear shutdowns can be abrupt and large-scale, as France's 2022 fleet availability crisis demonstrated when half its reactors went offline simultaneously. Honest system-cost accounting needs to model tail risks, not just averages.
SMR technology: promise versus demonstrated reality
Conservative
Small modular reactors are theoretically faster and cheaper, but no commercial SMR has been demonstrated at scale. Acknowledging this is not a reason to abandon nuclear — it is a reason to distinguish between the SMR bet and the proven technology already operating. Extending existing reactors and beginning conventional builds now does not require SMRs to succeed on schedule.
Liberal
That distinction is correct and important, but the political and financial case for nuclear expansion has leaned heavily on SMR promises to justify the investment thesis. NuScale's cancellation of its U.S. project after cost estimates nearly tripled should register as a warning that the 'proven technology plus SMR pipeline' framing obscures how much of the economic case rests on the unproven half.
Conservative
NuScale's cancellation was a first-generation commercial project hitting first-of-kind costs — the same dynamic that made early offshore wind projects expensive before standardization drove costs down. The lesson is to keep developing the technology, not to treat one cancellation as a verdict.
Liberal
The offshore wind analogy cuts against you: early offshore wind was expensive and got cheaper because governments sustained investment through the cost curve. If that's the model for SMRs, the honest ask is for explicit public subsidy and patience — not a claim that the economics already pencil out.
Nuclear waste: solvable problem or permanent liability
Conservative
The 97,000 metric tons of stored spent fuel is a real problem that deserves a real answer, not permanent avoidance. Finland is building a permanent repository right now — demonstrating that this is a solvable engineering and political problem. The waste objection cannot be the permanent veto on a technology that is already doing more for decarbonization than any other single source.
Liberal
Finland's success is precisely the point — it required decades of community engagement, Indigenous consent processes, and sustained political will that the United States has conspicuously failed to replicate. Yucca Mountain was abandoned not because the physics failed but because the politics did. Calling it 'solvable' without accounting for why the U.S. specifically keeps failing to solve it understates the actual obstacle.
Conservative
Yucca Mountain failed because it was sited by political imposition over Nevada's objections — Finland's model succeeded because it used genuine community consent processes. The U.S. failure is a process failure, not evidence that the solution is unavailable.
Liberal
Then fund the consent-based siting process, pass the legislation to replace the 1987 Nuclear Waste Policy Act amendments, and commit to the timeline — not as a condition of nuclear expansion, but as the simultaneous obligation that expansion creates. The waste argument isn't a veto; it's an invoice.
Conservative's hardest question
The 14.5-year average construction timeline combined with the unproven commercial status of SMRs means the nuclear expansion argument depends substantially on regulatory reform and technological demonstration that have not yet happened — if SMRs fail to deliver on cost and speed, the case for nuclear as a near-term climate solution weakens significantly, and capital invested in slow conventional builds could crowd out faster-deploying alternatives during the most critical emissions reduction window.
Liberal's hardest question
The 14.5-year average construction timeline is the argument's most serious vulnerability: even if everything goes right institutionally and financially, reactors ordered today arrive near the end of the critical 2030–2040 emissions reduction window, raising a legitimate question about whether capital deployed in nuclear could deliver faster climate returns if redirected to wind, solar, and storage at current cost trajectories. This is not easily dismissed, and an honest expansion case must be paired with aggressive parallel renewable deployment rather than presented as a sequential alternative.
The Verdict
Both sides agree
Both sides accept that Germany's post-Fukushima nuclear phase-out increased carbon emissions in the near term, treating it as a cautionary example rather than a model to emulate.
The real conflict
They disagree on a factual-methodological question: whether levelized cost estimates for renewables adequately capture grid-balancing, storage, and system-integration costs, which determines whether nuclear is economically competitive or merely ideologically preferred.
What nobody has answered
If both administrations — one aggressive on climate, one hostile to it — independently converged on nuclear expansion targets, does that reveal an engineering consensus, or does it reveal that nuclear is politically useful to constituencies that have nothing to do with decarbonization, and how would we tell the difference?
Sources
- World Nuclear Outlook Report 2025 — World Nuclear Association
- COP28 Declaration to Triple Nuclear Energy (2023)
- COP30 World Nuclear Association coalition announcement
- IEA nuclear emissions reduction estimates
- U.S. EIA 2025 levelized cost of energy estimates
- Biden Administration nuclear capacity announcement, November 2024
- Trump Administration Executive Order on Nuclear Regulatory Commission reform, May 2025
- Rockefeller Foundation-commissioned study on nuclear and renewables complementarity
- U.S. commercial spent nuclear fuel storage data, 2025
- Public opinion survey data on nuclear energy perception
In the news
May 9, 2026
Week in Politics: Redistricting fight in Tennessee and Virginia; latest poll on Trump
May 2, 2026
Event With Links to Oil Industry Teaches Judges “Healthy Skepticism” of Climate Science
Apr 25, 2026
Starmer plans new powers to ban state-backed terror groups
Apr 25, 2026
After 2 failed votes, Mike Johnson unveils new plan to extend key U.S. spy powers
Apr 24, 2026
Unfounded Health Concerns Are Powering a Solar Backlash