I am going to give you numbers first and argument second, because this debate too often runs on feelings and the numbers are decisive.
Capacity factor: the percentage of maximum potential output a power plant actually generates. Onshore wind in the US: approximately 35%. Utility-scale solar: approximately 25%. Nuclear: approximately 93%. That gap is not incremental. It is structural. Wind does not blow all the time. The sun does not shine at night. Nuclear runs continuously.
LCOE — levelized cost of electricity — has become the standard comparison metric and it favors renewables on paper. But LCOE does not capture system costs: the backup generation, storage, transmission buildout, and grid balancing required when intermittent sources drop out. When you add those costs, the economics of nuclear improve substantially relative to wind and solar. A 2021 analysis from MIT's Energy Initiative found that zero-carbon grids with nuclear power required significantly less total system investment than grids relying on renewables alone.
Vogtle Unit 3 came online in 2023, Unit 4 in 2024. Yes, they were over budget and delayed. I will discuss that in exchange. But both units are now generating approximately 1,100 megawatts each of carbon-free power that will run for 60-80 years without fuel price risk. You cannot say that about a solar farm.
The climate case for nuclear is simple: we need to decarbonize the grid in 25 years. Nuclear is the only carbon-free source that provides firm, dispatchable, high-density power at scale. We need it.
I want to start by saying something that will probably surprise you: I am not anti-nuclear. I am anti-building-nuclear-right-now-instead-of-renewables-that-can-actually-be-deployed-in-time.
The timeline is the whole argument. We have approximately 8-10 years to make substantial emissions reductions if we want to stay below 1.5°C. Maybe 15 years if we accept 2°C as the target. In that window, the question is not what is the best long-run energy source. The question is what can we build fast enough to matter.
Nuclear plants in the United States take 10-19 years from permit to operation based on recent experience. Vogtle was first proposed in 2006 and came online in 2023-2024 — 17-18 years. And it cost $35 billion for 2,200 megawatts. Solar and wind projects take 2-5 years to permit and build. The cost of utility-scale solar has dropped 90% since 2010. Offshore wind is expensive but deployable within the climate window.
I accept your capacity factor numbers. I accept the intermittency challenge. The answer to intermittency in the relevant timeframe is storage plus overbuilding plus demand response plus grid interconnection — not nuclear, which cannot be built in time to matter for the critical window.
The best nuclear case I can make for myself: build it for the post-2040 grid. Finance it now, permit it now, design it now. But do not tell me it is the centerpiece of a climate strategy that needs to produce results by 2035.
The timeline argument is the one I take most seriously and I want to engage with it directly rather than pivoting.
You said Vogtle took 17-18 years. Correct. But Vogtle was the first new nuclear construction in the US in 30 years, using a partially new design, built by a workforce that had lost continuity across a generation, in a regulatory environment with almost no institutional memory of licensing new construction. South Korea builds a comparable reactor in 5-7 years. France at its peak in the 1970s and 1980s was commissioning 3-5 reactors per year. The US problem is not nuclear physics, it is regulatory sclerosis and construction ecosystem collapse.
The ADVANCE Act passed in 2024 with bipartisan support. It streamlines NRC licensing. It funds workforce development. If we treat nuclear the way we treated wartime manufacturing — with urgency, standardized designs, supply chain investment — the timeline changes. Not to 2 years. But to 7-8 years for a standardized Gen III+ reactor, which means meaningful capacity by 2033-2034.
And I want to push back on the storage answer. Battery storage at grid scale currently costs approximately $300-400 per kilowatt-hour of capacity. To back up a coal plant that runs at 85% capacity factor, you need multiple days of storage, not hours. The math does not currently work for long-duration storage as a primary backup to wind and solar at continental scale.
We are now arguing about Vogtle specifically and I want to be honest about what it represents.
Vogtle Unit 3 and 4 are a cautionary tale about executing nuclear construction under US conditions as they existed before the ADVANCE Act. They are not a cautionary tale about nuclear physics, or about whether nuclear can be built on a relevant timeline under better conditions. That distinction matters.
France's nuclear experience — 58 reactors, peak construction rate of 5-6 per year in the early 1980s — is the template. Not Vogtle. France gets approximately 70% of its electricity from nuclear. Its grid carbon intensity is among the lowest in Europe, including lower than Germany which shut down its nuclear plants and replaced them partly with gas. That is not a hypothetical. That is what happened.
The iron-air battery point is interesting but commercial deployment at scale for a technology entering the market now will not solve the backup problem by 2030. Iron-air is promising. It is not a 2028 solution at continental scale.
I think we are converging on a position where we agree nuclear should be part of the post-2035 grid and disagree about whether it should be the centerpiece now. That is an honest disagreement about risk and timeline and I am not sure more rounds will resolve it. I am open to closing if you are.
The France comparison is the strongest thing in your case and it is genuinely strong. 70% nuclear, low grid carbon intensity, and they managed it at scale. The Germany comparison — shutting nuclear and partially replacing with gas — is also fair and something my side of this debate needs to reckon with more honestly.
Here is where I land: I think the honest version of my argument is not "nuclear is bad" but "nuclear cannot be the centerpiece of a strategy that needs to produce results in a 10-year window." France built its fleet over 20 years starting in the 1970s. We do not have that runway for the near-term climate targets. We need to be building renewables as fast as we can now AND investing in the nuclear pipeline for the 2035+ grid.
The word "centerpiece" in the debate topic is doing a lot of work. If you mean centerpiece of a long-run decarbonized grid, I am closer to your position than my opening suggested. If you mean centerpiece of what we build in the next decade, I still think you are wrong.
I am willing to propose closing.
Accepted. Let's close. But one thing first.
The word centerpiece was in the original topic and I want to stand by it, not equivocate. My argument is that we should be building nuclear aggressively right now, in parallel with renewables, because the 2035+ grid cannot be decarbonized without it. Every year of delay on permitting and financing nuclear that comes online in 2035 is a year of fossil fuel backup after 2035. The climate window is not just 2025-2035. It is ongoing.
I think we actually agree that nuclear belongs in the future grid and disagree about urgency and sequencing. That is a real disagreement but it is a narrower one than it looked from the opening positions.
Nuclear power has a capacity factor above 90%, a 60-80 year operational life, zero carbon emissions, and a track record of providing reliable baseload power on every continent. The United States has 93 operating reactors that currently provide about 20% of our electricity. The case for expanding that fleet is the same as the case for keeping those reactors running: you cannot run a modern economy on sources that produce power only when the weather cooperates.
My opponent made honest concessions over the course of this debate. Germany's nuclear shutdown is a lesson. France's nuclear build-out is a template. The ADVANCE Act creates a better regulatory pathway. None of those things were in the opening. They emerged from argument.
Nuclear should be the centerpiece of US climate policy because it is the only technology that provides firm, dispatchable, zero-carbon power at scale. Renewables are essential — I have said so throughout. But without nuclear as the backbone, you are building a grid that depends on either fossil fuel backup or storage technology that does not yet exist at the required scale. That is not a climate strategy. That is a hope.
Climate is an emergency and emergencies require deploying the fastest available tools first.
In the next decade, that means solar, wind, storage, and grid modernization. Not because nuclear is bad — it is not — but because we cannot wait 12-15 years for new reactors while emissions continue to accumulate. Every megawatt of solar we build in 2025 starts displacing carbon in 2027. Every reactor we permit in 2025 starts displacing carbon in 2037 at the earliest. In a climate emergency, that difference is not marginal. It is decisive.
My opponent is right that nuclear belongs in the long-run decarbonized grid. I came around to that position more clearly over this debate. France's experience is genuinely compelling and the Germany comparison is a real lesson for the left.
But centerpiece means the thing you build your strategy around. And you cannot build a 2030 climate strategy around a technology with a 2037 earliest-deployment timeline. You build the 2030 strategy around wind and solar, and you build the 2040 strategy around nuclear plus storage plus whatever else has matured. That sequencing is not anti-nuclear. It is arithmetic.