One of the most common claims from anti-nuclear people aimed at nuclear power is that it is too expensive, usually implying that there is something fundamentally endemic to nuclear power that makes it expensive. They always exclusively cite Levelized Cost of Energy (LCOE), and they almost exclusively focus on costs in the USA. This approach is a combination of confirmation bias, cherry-picking, egocentrism, and intellectual dishonesty. The high cost of nuclear power is to some degree unique to the USA and its politics regarding nuclear energy and energy production more generally.
The US Is a Nuclear Cost Outlier for Many Avoidable and Political Reasons
As you may know, the massive bulk of the cost of a nuclear power plant is the upfront costs of planning and construction. This expense is between 60 and 70 percent of overall nuclear costs (Lovering et al., 2016; U.S. Energy Information Administration, 2020). Normal operation and maintenance and spent fuel storage is very small proportionally.
Very little nuclear construction has occurred in the US in the last several decades to even give us a good consistent idea of how much it costs to build here. But the comprehensive worldwide data we have, as reported by Lovering et al. (2016), indicates that most countries maintained a relatively steady cost regarding nuclear plant construction. In particular, costs actually decreased in South Korea. The highest behind the US, Japan, is still much cheaper than ours. And their costs may be a result of building in their uniquely geographically unstable and tsunami-prone area.
Before going on I want to emphasize that the facts reported in this above picture fully debunk the claim of nuclear opponents that nuclear power is intrinsically expensive.
But this begs the question, why does nuclear cost so much more in America? Undoubtedly many factors are to blame, but a partial answer can be gleaned in a paper going back to the early nineties. MacKerron (1992) concluded that “It is clear that the pressures to higher capital costs…have been powerful… The pressures for increases in cost have mostly been connected to the growing complexity of broadly similar basic design… The growing complexity has itself been driven primarily by persistently more stringent safety requirements, themselves driven by…growing public environmental consciousness…and increasingly sophisticated anti-nuclear campaigning.”
The number of engineering standards applied to nuclear equipment rose from 400 in 1970 to 1800 in 1978; the of number guides and positions from the US NRC increased from 4 in 1970 to 304 in 1978; there was essentially a doubling in the required amounts of concrete, steel, cable, and conduits for nuclear plants from 1973 to 1978 to protect from things like seismic events.
Many of these increases in safety requirements were warranted. But because of the political anti-nuclear hysteria in the US, particularly following the overblown coverage of the Three Mile Island (TMI) incident that had no casualties and no meaningful negative environmental effects, US nuclear regulations have transmogrified into prohibitively costly and unnecessary. Just obtaining the very licenses and permits required to build new reactor units on an existing plant can take north of 5 years a piece. Reasons like this are drastically increasing the time and costs of completing projects, unnecessarily.
Lovering et al. (2016) looks specifically at Overnight Constructions Costs (OCC):
The “overnight” attribute refers to the construction cost of a nuclear reactor as if the reactor construction process were completed instantaneously, without incurring the financing charges accrued before commercial operation, known as Interest During Construction (IDC). The OCC metric is meant to isolate the cost invariant to construction duration and interest rate, in order to capture the cost intrinsic to the reactor technology.
The authors however note that an average of 46% of a nuclear plant’s cost is IDC. What affects IDC? Among a few factors, duration of construction is massively important. Of course, regulations have a huge impact on duration of construction. Let us look at duration of construction and OCC following the TMI incident in America.
The graphic basically tells us that both construction duration of nuclear plants and OCC rise nearly in lockstep after TMI, which are both very likely a factors of increased regulations. Lovering et al. (2016) state, “reactors that received their operating licenses before the TMI accident experience mild cost escalation. But for reactors that were under construction during Three Mile Island and eventually completed afterwards, shown in red, median costs are 2.8 times higher than pre-TMI costs and median durations are 2.2 times higher than pre-TMI durations.” They conclude that “issues such as licensing, regulatory delays, or back-fit requirements are a significant contributor to the rising OCC trend.”
It’s no surprise then that states which have become politically hostile against nuclear are the ones where nuclear is disappearing from—overregulation and draconian laws being a primary factor.
Diablo Canyon Power Plant (DCPP), California
California for example, just scheduled the shutdown of its last nuclear power plant, the Diablo Canyon Power Plant (DCPP). This shutdown had been vigorously encouraged by supposed environmentalists—in and outside the local government—for a long time, with the final nail in the coffin being the looming process of obtaining a Federal license renewal for operation. Looming large in this event is California’s Senate Bill No. 100 which requires that electric utility providers provide 60% of their electricity from renewables by 2030. Nuclear is zero/low-carbon, but not renewable, so this policy blatantly disadvantages nuclear. As recently as 2013 California subsidized solar double what it subsidized nuclear (The Breakthrough Institute, 2013). Most of that was direct solar subsidies, making solar artificially cheaper for consumers. The subsidies nuclear received was primarily for R&D. Given such preferential policies, of course Pacific Gas & Electric, the utility in charge of DCPP, did not consider it economical to fight for a nuclear plant that State law would require them to limit the scope of regardless.
That DCPP’s supposed economic infeasibility is political can be seen by just looking at its cost-benefit basics. It cost $14 billion in 2019 dollars to complete in 1986 and it has operated for 35 years. Its positive yearly statewide economic benefit is $1.1 billion, with a nationwide economic benefit of almost $2 billion (Riener & Mayeda, 2013). That means after 35 years it is overall $56 billion in the green—an economic no-brainer. This is admittedly a fairly crude estimation, but deeper analysis is unlikely to yield markedly different results. It is interesting how anti-nuclear people claim DCPP closed not because of laws and regulations but because of economic factors, even though those economic factors resulted from laws and regulations. Mental gymnastics at its finest.
Vogtle Reactors 3 and 4 in Georgia
Let’s compare some prices. Topaz Solar Farm, 241 MW (after accounting for 26.6% capacity factor), $2.4 billion, completed 2014. Votgle reactors 3 and 4, 2,002 MW (after accounting for a likely 91% capacity factor), $27.5 billion. Reactors 3 and 4 cost 11.45 times as much as topaz, while delivering only 8.3 times the power. This makes Votgle come out the loser in comparison. Anti-nuclear people would stop here and declare nuclear failed, but there is good reason to believe Votgle is not a reasonable extrapolation for the cost of typical nuclear construction in the US, and it is most definitely not a reasonable assumption for costs endemic to nuclear construction period.
Vogtle was plagued by the unnecessary blunders by the contractors who installed rebar in a design not in accordance with US Nuclear Regulatory Commission requirements, along with supposedly doing the “wrong type of welds”. These issues caused great delays which caused increases in cost, and these didn’t have to happen. Additionally, as we know the US has increased its nuclear regulation to stifling levels following the fear-mongering of TMI, the original rebar design and weld may realistically have been more than sufficient.
The Vogtle project also had to defend itself against multiple lawsuits, including one started in 2012 by anti-nuclear activists claiming that it was too dangerous, and that Georgia already made way more electricity than it needed. I suppose they paid no attention to the fact that the power company, Georgia Power, as of 2011 supplied 62% of the power with coal plants that it intends to close for environmental reasons. Thus, the Vogtle expansion was necessary to offset those closures with low-emission sources. But anti-nuclear activists weren’t looking to understand the reasoning, good or bad. They were simply grasping at whatever they could to justify their anti-nuclear sentiments. Had Georgia Power wasted $27.5 billion on a botched solar farm these groups likely wouldn’t have made a peep. They certainly wouldn’t have suggested such a botched solar project indicated the generalized nonviability of solar power. Of course, solar faces monumentally fewer regulations than nuclear anyway.
And remember, the two new units at Vogtle are almost the only added nuclear capacity added since the early 90’s, which means we have extremely little to compare to in order to determine how much the average cost of a nuclear project in the US would be. Vogtle is likely the upper extreme in cost scenarios, considering the installation of extremely similar Westinghouse units in other countries.
For an example of how a nearly identical construction project can be done right, and perfectly cheap and cost-effective, look at China’s Sanmen Nuclear Power Station. Built from scratch construction started in 2009 and both units were successfully commissioned in 2018. The units installed were the exact same make and model as those installed at Vogtle; Westinghouse (American company) AP 1000 reactors. The cost was $7.3 billion dollars, or about 1/4th the cost of the two Vogtle reactors.
Even the pricey Barakah nuclear power plant in the United Arab Emerates blows the Vogtle project and basically every completed US solar project out of the water when it comes to cost compared to power output. Barakah cost $24.4 billion, is nearly complete, will use four South Korean APR-1400 reactors, and has a 5600 MW nameplate capacity. Assuming the world average capacity factor for nuclear plants of 82%, we’re looking at an average actualized capacity of 4,592 MW (World Nuclear Association, 2020). Compared to Topaz solar, Barakah is 10 times the cost, but will produce at least 19 times the power. The reactors could even be much higher of a capacity factor. The average capacity factor of Finland’s nuclear plants is 95%. If Barakah ran at those levels, it would output 5,320MW, or 22 times Topaz.
For a more solar-friendly scenario we could compare Barakah to a newer, larger solar park; Pavagada Solar Park in India. It is a 2,050 MW solar park commissioned in 2018. It cost over ₹14,425 crore ($2.1 billion). Accounting for a likely 26.6% capacity factor, that will look more like 545.3 MW. So Barakah is 11.6 times as expensive, and produces 8.4 times as much electricity (9.75 times if 95% capacity factor). Solar does inch ahead of nuclear in this specific match-up. Though, nuclear opponents should refrain from rejoicing just yet. Nuclear is also much cheaper in India. The large Kudankulam Nuclear Power Plant reactors 1 and 2 (completed and commissioned 2013 and 2016 respectively) cost a scant $2.6 billion and put out about 1,865 MW. So, those nuclear reactors are barley more expensive, and produce almost four 4 times the power of the comparable solar farm. This puts nuclear way back on top. Why is India cheaper for both solar and nuclear? Labor cost is probably a big factor. Typical salary in India is $7,000 but $27,00 in the UAE. The moral of the story is that one must be extremely careful when comparing solar in one country to nuclear in another. They are not apples-to-apples comparisons. Solar is much more expensive in America as well, not just nuclear.
Barakah, Sanmen, and Kudankulam are just examples of what the graphic at the beginning of this article clearly told us; most countries do nuclear far cheaper than the US, and while being just as safe. Regarding the Barakah plant, the UAE website states they “adopted best practices from operators around the world and from industry organizations, including the International Atomic Energy Agency (IAEA) and the Institute of Nuclear Power Operations (INPO).”
As a last example of nuclear being more than affordable enough to beat solar hands down, let us consider the Palo Verde Nuclear Generating Station. Completed in 1988 it took 12 years and $11.9 billion 2019-dollars to make. It’s currently the largest nuclear plant in the US. It cost 4.9 times as much as Topaz solar farm, but it generates 25.22 times the electricity. After feeding $2 billion into the Arizona economy for the last 33 years it is around 54 billion in the green (Arizona Public Service, 2018). Like most US nuclear projects, it went well past initial expected costs, and like most US nuclear projects, its net economic benefit blew way past its final costs regardless.
Change in Governmental Economic Approach to Utilities
Another major factor in the bloated nuclear costs in the US is that “utilities are almost everywhere being encouraged by governments to behave in a more market oriented and less public service mode” (MacKerron, 1992). This means that energy sources like natural gas that pollute more and are more expensive over the life of the plant are favored because they can be made more quickly and cheaply to deliver profit in the short term. This focus on short-term economics rather than long-term economics is childish, irresponsible, and not something that should be championed by supposed environmentalists.
Solar PV and the Problem with Variable Output Sources
And all this is just looking at nuclear power in comparison to solar PV at its best. Because of the variable nature of solar PV and all other variable output sources such as wind, the cost of integrating them into a power grid skyrockets as they make up a larger and larger percentage of said grid.
Reichenberg et al. (2018) found that going from 0% to 80% of an electric grid the LCOE cost of VRE sources nearly doubles! Past 80% their price increases even faster! LCOE is a metric biased against nuclear (in fact it is terribly flawed as I have previously proved). But just going by LCOE, solar PV passes nuclear in cost before it even makes up 50% of the grid! At about 50% of the grid, the LCOE of VRE sources reaches about 65 euros/78 dollars; more than nuclear, which was $69.39 in 2020 (U.S. Energy Information Administration, 2020). So much for being an affordable alternative.
These findings aren’t a one-off result either. Capellán-Pérez et al. (2019) found that if we try to get to 100% renewable energy (primarily focusing on VRE sources) by 2060, renewables fall from an average Energy Return on Investment (EROI) of 12 today to 3 by mid-century, then level out at 5 after that. EROI is how much energy we get back for the amount we put into building and maintaining an energy source. Results like this show 100% reliance on VRE sources is completely infeasible. Dispatchable sources like nuclear don’t have these VRE problems. The EROI of nuclear is 75, or 6.25 times solar PV (Weißbach et al., 2013; Raugei et al., 2017).
Looking from the context of the engineering problems a grid runs into trying to incorporate variable sources of energy, the rest of the economic superiority nuclear opponents claim about VRE sources is exposed as hollow. With all these things in mind, it is not surprising that a study by Emblemsvåg (2020) in the International Journal of Sustainable Energy concluded that VREs such as “windfarms are not cost effective when a certain output must be guaranteed as major opportunity costs are introduced… New technologies do not provide enough productivity gains to offset the advantage that existing power plants have concerning lower fixed costs.”
To summarize, the US is unique in how bloated the price of nuclear is. Virtually every other country does it cheaper, and among the main reasons the US is so expensive is because of prohibitively strict overregulation resulting from irrational fears stemming from Three Mile Island incident. This doesn’t mean other countries don’t regulate nuclear, they do. They just do it in a much more rational less fear-based way. Additionally, explicitly anti-nuclear policies from law-makers are leading to plants closing for supposedly “economic reasons” which in reality are the result of anti-nuclear policies. However, even with the highly regulated nature of nuclear in America most nuclear projects would still likely deliver a much better cost-benefit ratio than solar PV farms. We simply start nuclear projects so rarely that we have few examples to even look at. Lastly, when looking at alternatives like solar and wind power in the context of integrating them into a power grid, the lead of nuclear power over such sources widens considerably.
Don’t let any of this give the impression that any VRE use is bad, or that I oppose adding involving them in the energy mix. That is not the case. I think VRE energy sources should make about around 25 to 35 percent of the energy mix, with nuclear making up most of the rest. Unlike nuclear energy opponents, my position isn’t motivated by a fear or hate of other energy sources, it is motivated by my best interpretation of the facts. And based on these facts, I am not an opponent of solar and wind. Advocacy of solar and wind can and should coexist with advocacy of nuclear power if we are serious about solving climate change.
- Arizona Public Service. (2018). Palo Verde Generating Station. Retrieved from https://www.aps.com/-/media/APS/APSCOM-PDFs/About/Our-Company/Energy-Resources/PV_FactSheet.ashx?la=en
- The Breakthrough Institute. (2013). Subsidies for solar two times higher than for nuclear in California. https://thebreakthrough.org/issues/energy/subsidies-for-solar-two-times-higher-than-for-nuclear-in-california
- Capellán-Pérez, I., de Castro, C., & Miguel González, L. J. (2019). Dynamic energy return on energy investment (EROI) and material requirements in scenarios of global transition to renewable energies. Energy Strategy Reviews, 26(September 2018), 100399. https://doi.org/10.1016/j.esr.2019.100399
- Emblemsvåg, J. (2020). On the levelised cost of energy of windfarms. International Journal of Sustainable Energy, 0(0), 1–19. https://doi.org/10.1080/14786451.2020.1753742
- Lovering, J. R., Yip, A., & Nordhaus, T. (2016). Historical construction costs of global nuclear power reactors. Energy Policy, 91, 371–382. https://doi.org/10.1016/j.enpol.2016.01.011
- MacKerron, G. (1992). Nuclear costs. Why do they keep rising? Energy Policy, 20(7), 641–652. https://doi.org/10.1016/0301-4215(92)90006-N
- Raugei, M., Sgouridis, S., Murphy, D., Fthenakis, V., Frischknecht, R., Breyer, C., … Stolz, P. (2017). Energy Return on Energy Invested (ERoEI) for photovoltaic solar systems in regions of moderate insolation: A comprehensive response. Energy Policy, 102(January), 377–384.
- Reichenberg, L., Hedenus, F., Odenberger, M., & Johnsson, F. (2018). The marginal system LCOE of variable renewables – Evaluating high penetration levels of wind and solar in Europe. Energy, 152, 914–924. https://doi.org/10.1016/j.energy.2018.02.061
- Riener D., K., & Mayeda, P. (2013). Economic benefits of Diablo Canyon Power Plant: An economic impact study (Issue June). http://large.stanford.edu/courses/2016/ph241/orozco2/docs/mayeda.pdf
- U.S. Energy Information Administration. (2020). Levelized cost and levelized avoided cost of new generation resources in the annual energy outlook 2016. US Eia LCOE, February, 1–20. https://www.eia.gov/outlooks/aeo/pdf/electricity_generation.pdf
- Weißbach, D., Ruprecht, G., Huke, A., Czerski, K., Gottlieb, S., & Hussein, A. (2013). Energy intensities, EROIs (energy returned on invested), and energy payback times of electricity generating power plants. Energy, 52, 210–221. https://doi.org/10.1016/j.energy.2013.01.029
- World Nuclear Association. (2020). World Nuclear Performance Report 2020. https://www.world-nuclear.org/getmedia/3418bf4a-5891-4ba1-b6c2-d83d8907264d/performance-report-2020-v1.pdf.aspx