Does California Need New Nuclear Power Plants

There's been a lot of talk about nuclear power recently. Plant owners have been making plans to restart their retired nuclear power plants, Google signed the first corporate deal to purchase energy from small modular nuclear reactors, and the US pledged to triple its nuclear generating capacity by 2050. As one energy news outlet put it, "2024 was a breakout year for the US nuclear power sector - at least on paper."

At least on paper…

In reality, the picture isn't nearly as promising. For conventional nuclear reactors, only three units have come online over the past two decades, and two of those three units at the Vogtle nuclear power plant in Georgia completely blew their budget. The picture is even more uncertain for small modular reactors (SMRs), which differ from conventional reactors in that they are significantly smaller, manufactured at a centralized facility, and deployed in a modular fashion. NuScale is the first company to get regulatory approval for an SMR design. But after making major changes to its approved design, projecting significant cost increases, and failing to attract enough customers, NuScale cancelled its first big SMR project.

In short, there's a lot of hype around nuclear, but there isn't much tangible progress.

That hasn't deterred some California officials from taking up the cause and trying to open the door to a nuclear power resurgence in California. At present, California effectively has a ban on new nuclear power plants, but some California legislators are interested in rolling that back ostensibly to advance California towards its clean energy goals.

So the question is, does California need new nuclear power plants? Or is all this nuclear hype a distraction?

California's ban on new nuclear power plants

Before I get to the question at hand, it'll be helpful to review some history to understand how California ended up with its current ban on new nuclear power plants.

At present, Diablo Canyon is the only nuclear power plant operating in California, but the state has a long and storied history with nuclear power (including the tale of missing nuclear fuel rods that PG&E never could locate). California had two other large reactors: Rancho Seco and San Onofre, but both suffered serious equipment failures and were shut down in 1989 and 2013, respectively. California also had three much smaller nuclear power plants that only operated in the 1950s through 1970s, and there were a handful of proposed nuclear power plants that never came to pass.

In 1976, the California legislature passed a law that effectively bans new nuclear power plants. To be more specific, the law forbids the California Energy Commission from issuing a permit to any new nuclear fission power plants until there's a way to dispose of toxic and long-lived nuclear waste. With no permanent repository for nuclear waste and with federal policymakers at an impasse over how to handle the spent fuel that's been piling up, this California law precludes any new nuclear power plants from getting built in the state.

This Diablo Canyon containment structure was under construction in 1972, and the plant began operation in the mid-1980s. Diablo Canyon is now California's last operational nuclear power plant, but it's currently slated to shut down by 2030.

Recent interest in nuclear power in California

In the past couple of decades, California legislators have put forth different proposals to repeal the ban. For example, a 2008 bill would have lifted the moratorium entirely and replaced it with a less stringent set of requirements. More recently, a 2023 bill would have exempted SMRs from the ban on new nuclear power plants, and a 2024 bill would have required a study on the feasibility and benefits of deploying SMRs. None of those bills passed, though some legislators are clearly intrigued by the prospect of opening the door to SMRs in California.

But should they be?

The nuclear bros make all sorts of claims extolling the merits of SMRs; however, there's good reason to be skeptical. For example, as I'll discuss later, SMR cost estimates vary widely, and we still don't have a good sense how much costs will really come down and whether SMRs will be cost-competitive with other resource types. Furthermore, SMRs won't necessarily be safer than conventional reactors, and some so-called "advanced" reactor designs may actually be less safe. Lastly, SMRs won't meaningfully reduce the amount of nuclear waste that's created, and they won't use fuel more efficiently than conventional reactors. In fact, the fuel required for some new SMR reactor designs could be used directly to create nuclear weapons, greatly increasing the risk of nuclear proliferation.

Those are significant downsides that cannot be ignored. But just for a moment, let's focus solely on the costs of SMRs and how those costs influence the potential role of SMRs on the grid.

Would new nuclear help with cost-effective decarbonization?

When planning for the future of California's grid, the goal is generally to build the mix of resources that meet clean energy, reliability, and other state policy goals at least cost to consumers. To develop these plans, decisionmakers use grid modeling tools to help identify that least-cost mix of resources.

So when we consider whether new nuclear power plants could help California decarbonize, the question is really whether new nuclear power plants are part of that least-cost mix. (Again, we're temporarily setting aside the critical safety and security concerns.) And the answer to that question depends heavily on the costs of new nuclear power plants.

It depends on the costs

The levelized cost of energy (LCOE) represents the average cost to build and operate a generating resource per unit of electricity generated. LCOE is not a perfect metric, but it's still useful to get a sense for how much new nuclear power plants might cost. But to be frank, the LCOE estimates are all over the place, which isn't too surprising given that they're highly speculative due to a lack of real-world data from actual projects.

For example, Lazard puts the LCOE of new conventional nuclear power plants between $142 and $222 per MWh (and they put the LCOE of the new Vogtle units mentioned earlier at $190/MWh). While estimates for SMRs vary even more widely, a literature review of cost estimates for advanced reactors found an LCOE range of $60-$100 per MWh for large reactors and SMRs. However, this study was conducted by researchers at a national lab with a mission to advance nuclear energy, so we should take these findings with a grain of salt. In addition, it's important to note that the range in that literature review represents the LCOE after initial deployments have started to bring down costs. For example, the LCOE for NuScale's first SMR project mentioned earlier rose to $119 per MWh (without federal subsidies) before it was cancelled, so the LCOE for SMRs is likely much higher than that range at least in the near-term.

Last, I should note that, when determining whether nuclear power plants could be part of that least-cost mix of resources, the cost of nuclear isn't the only thing that matters. The cost of other energy technologies matters as well. So if solar, wind, energy storage, geothermal, etc. were to come down in price more than expected, that could change the calculus as well.

What the studies tell us

With those caveats and ballpark LCOE numbers in hand, there are a handful of studies we can examine to gauge the extent to which new nuclear power plants could advance California's decarbonization efforts at least cost. Let's review a few of them before drawing some conclusions.

• The California Energy Commission's SB 100 report examined various scenarios that reach 100% clean electricity in California. One scenario included "generic zero-carbon firm resources," a category that nuclear power plants fall into, with an LCOE of $60 per MWh. In this scenario, the lowest-cost mix of resources included 20 GW of those "generic zero-carbon firm resources." Those resources dramatically reduced the scale of solar and storage buildout by more than 50 GW and 30 GW respectively, and they led to a modest six percent cost reduction.
  
• The Environmental Defense Fund and the Clean Air Task Force's Clean Firm Power study examined what it would take to completely eliminate emissions from California's electricity sector. The study included nuclear power resources with an LCOE just shy of $60 per MWh. Similar to the SB100 study, it found that including nuclear power as an option leads to the buildout of tens of gigawatts of nuclear capacity, which dramatically reduces the scale of the renewable and energy storage buildout. And in this case, it also significantly reduces the overall costs.
  
• My colleagues at the Union of Concerned Scientists conducted a national decarbonization analysis, Accelerating Clean Energy Ambition. While not California-specific, this analysis examined what it would take for the entire U.S. economy to reach net-zero emissions by 2050. The analysis included new nuclear power plants as an option, with an LCOE starting at $89 per MWh (with federal subsidies, or $119 per MWh without) in 2030 and declining slightly over time. And in this analysis, no new nuclear power plants were selected. In other words, new nuclear was not part of the least-cost mix of resources.
  
• Evolved Energy Research's 2024 Annual Decarbonization Perspective is also a national decarbonization study, which examined various scenarios for reaching net-zero emissions across the entire U.S. economy. This study assumed a nuclear LCOE of $104 per MWh in 2030, decreasing to $63 per MWh in 2050. In its central decarbonization scenario, no new nuclear plants were built, indicating that it's more cost-effective to utilize other zero-emission technologies. Furthermore, this study included additional analysis that found nuclear would need to achieve an LCOE of $47-$59 per MWh for it to play a large-scale role in economy-wide decarbonization.

Without turning this blog post into a full-blown literature review, we can start to identify some takeaways: nuclear power could play a big role in the least-cost transition to clean electricity, but it depends heavily on achieving major cost reductions which have yet to be realized.

If nuclear power were only $60 per MWh, it would look like a good deal, and it would become part of the least-cost mix of resources. But when nuclear costs are higher, it doesn't make sense to invest in the technology since there are lower-cost alternatives available.

Rancho Seco was California's third largest nuclear power plant, but it was closed by public referendum in 1989. Hajhouse/Wikipedia

Does California need new nuclear power plants?

In short, the evidence suggests that it is neither necessary nor wise for California to pursue new nuclear technologies right now. There are three reasons.

First, for nuclear to play a big role in decarbonizing California's electricity sector at least cost, new nuclear technologies would need to come down in price dramatically. No one knows to what extent nuclear costs will go down, but the nuclear industry has a long history of cost overruns, and right now SMR projects are trending in the wrong direction. It'll be no small feat to cut costs to get the LCOE down below $60 per MWh, at which point nuclear becomes more cost competitive.

Second, as I mentioned earlier, the cost of other clean energy technologies matters too. For example, enhanced geothermal could compete directly with nuclear since it also has the capability to operate continuously over long periods, a valuable characteristic for grid reliability. With vast geothermal potential in addition to encouraging recent trends, geothermal could very well beat out nuclear. Other emerging clean energy technologies could also aid decarbonization efforts without turning to nuclear. For example, long-duration energy storage or green hydrogen could both play a role in squeezing out the last tons of emissions from the electricity sector. I'm not going to try to predict which of these technologies will come down in cost quickly enough to play a significant role on the grid. But California has already signaled its intention to make big investments in geothermal, long-duration storage, and other clean energy technologies, and the recent trends in progress with these solutions lead me to be much more optimistic about their prospects.

Third, safety, security, and waste concerns matter greatly with nuclear power plants, and those must be considered as well. Even if new nuclear technologies come down in cost enough to compete with other clean electricity technologies, all nuclear power plants have a unique set of serious risks and associated costs attached to them. Any nuclear power plant comes with a risk of a catastrophic accident that is difficult to estimate accurately, and the U.S. Nuclear Regulatory Commission (NRC) doesn't have the greatest track record of taking those risks seriously. The NRC is under significant pressure to weaken safety standards, and many new nuclear reactor designs may be less safe than current designs. With that reality, California's moratorium on new nuclear power plants, which the state fought hard to defend in court, serves as an important line of defense against potentially unsafe nuclear facilities being sited in the state. On top of the safety concerns, some new reactor designs would use fuel that increases the risk of nuclear proliferation, and all nuclear reactors will still produce high-level nuclear waste with no long-term storage solution.

There's a lot of hype around new nuclear power technologies, but they're still very expensive, and they come part and parcel with unique risks. California has been leading the nation in the clean energy transition, investing in new clean energy technologies and bringing down their costs enough for them to outcompete fossil-fueled energy sources. To achieve the state's clean energy goals, California should continue to invest in new clean technologies that will ease the transition, and reopening the door to new nuclear power plants now is neither necessary nor prudent.

Public link

Please use the above public link if you want to share this noodl on another website.