If you think Nuclear is the answer, you're stuck in the past

"If only we'd build nuclear!" is one of those common tropes in techno-optimist circles that sounds right on the surface but is pretty clearly wrong. I'm not anti-nuclear in the long run, but if we want to decarbonize quickly, solar and wind are our best options.

Solar and wind are so cheap that in many cases building them is cheaper than the marginal (read fuel) costs of running nuclear, colar, and gas plants. This is before a friendly Russian dictator sent gas prices through the roof so you can imagine what this looks like now.

The first and most obvious reason is cost. I don't think most people appreciate how absurdly cheap solar and wind have gotten. 3 years ago they were 1/4 the cost of nuclear, prices have dropped since then so it's closer to 1/8th now.

This might have been different if we had invested more in nuclear 50 years ago. We'd likely be in a very different spot if our grid looked more like France's. We didn't and now we have to decide how we move forward, not dwell on yesterday's spilled milk. The clock is running out to decarbonize. Solar and wind have shown they're up to the task. Nuclear is at best a slower more painful way of getting there and at worst a red herring meant to keep us from trying. Let's not get distracted.

With all this said, I'm not anti-nuclear. There's a lot of great potential emerging nuclear technologies and we should invest in them. I also strongly agree with the nuclear stans that closing existing nuclear plants is foolish when we're still dependent on fossil fuels. Nuclear has a role in a carbon free future, but let's not pretend it will solve all our problems if only we'd let it. While it has some strengths, it also has some glaring weaknesses.

On 3, if your solution relies on convincing people we just need to make nuclear plants less safe then your solution is dead on arrival. If we're going to live in a hypothetical world, can it at least be one where we've solved cold fusion instead?

On 2, maybe. But if we continue to invest in solar, wind, and batteries they'll also see additional economies of scale there. There's no reason to believe the drop will be faster for nuclear than it will be for the technologies already winning the race.

On 1, new reactor technologies are great but they're still in the development stage. We don't have 10 years to refine the technology and another 10 to figure out how to scale it. We need technologies we can deploy massively now, not R&D stage experiments.

Now, to steel man the nuclear stans, I think there's three counter-arguments they'd make:

1. I'm not accounting for new reactor technologies
2. If we build more nuclear, it will experience economies of scale
3. Nuclear is expensive because of unnecessary regulatory burdens

If we're talking about disadvantages, we also can't ignore nuclear's biggest one - people are afraid of it. That's not something you can just wave away with a wand. A technology that people will gladly put on their roof has an inherent advantage over one they'll protest to avoid.

You could overbuild, but then you're overbuilding an already expensive power source. You could rely on the same solutions renewables will need to smooth supply and demand, but in that case you might as well go with the far cheaper power source.

Reliability is also a knife that cuts both ways. It's an advantage for nuclear but also a weakness. The thing is, the goal is not consistency of generation, it's being able to match supply and demand. While nuclear generation is incredibly consistent, it isn't flexible. It takes a nuclear reactor 24 hours to fully ramp up. That's why nuclear plants operate at close to full capacity. They're great for base load, but face the same supply / demand challenges as renewables just in a different way.

There are other pieces to the puzzle like demand management, hydro, larger more interconnected grids, and yes even some nuclear and gas to help make a renewable grid more resilient. However, the fact remains that reliability is not the achilles heel it's made out to be.

This isn't even accounting for the potential impact of EVs. The typical EV packs has a 60-120 kWh battery, enough to power the average American household for 2-4 days. As EVs go mainstream, a huge new storage source will come online with minimal marginal cost to the grid.

Costs have come down so rapidly that LA was able to sign a contract for a solar + storage project at just 3.3 cents per kWh (equivalent to $33 / MWh or about 1/5 the cost of a new Nuclear plant) and those costs will only drop over time. https://pv-magazine-usa.com/2019/09/10/los-angeles-commission-says-yes-to-cheapest-solar-plus-storage-in-the-usa/

Like solar and wind, the massive drop in battery costs makes a renewable grid feasible. The cost of battery storage is falling even more steeply than solar and wind, but even without further declines the cost of stored power is already lower than new nuclear.

There is however, one gap that anyone who lives in California is likely familiar with - the evening. As the sun sets and before wind generation picks up, you can see power generation dip in the chart above. That's why peak pricing in California runs from 4-9 pm. This is why the one key remaining piece for the renewable transition is storage.

But what about reliability? The sun sets, the wind doesn't always blow, a nuclear plant can run rain or shine. It's true that this is one place where nuclear has the upper hand, but that advantage is smaller than you'd think. While solar + wind each have their inconsistencies, their generation profiles are complimentary. While solar peaks during the day, wind, conveniently, peaks at night. Combined, the two start to look much closer to base load, especially at grid scales.

Investing in solar and wind brings down the cost for everyone for a technology that is easier to deploy globally. You could easily build solar generation in any country in the world today. How many countries is that true for with existing nuclear technology?

A second, under-appreciated advantage of solar and wind is how scalable they are. A single wind turbine costs $2 - $4m. The cheapest solar panel I could find on Amazon was $15. A new Nuclear plant will cost on the order of $2-4b. The reason that matters is because the US is only 5% of global carbon emissions. The entire developed world represents less than 1/3 of global emissions. Nuclear might solve the problem for developed markets, but we need global solutions not regional ones.