Joshua Antonini, Mackinac Center for Public Policy
“Dying of an incurable attack of market forces.” This diagnosis of the nuclear energy industry comes from environmental activist Amory Lovins. Reuters agrees that nuclear power is financially untenable, claiming that bringing reactors online is “too slow, too expensive” as the market for non-carbon-emitting energy heats up.
You might be inclined to agree with these skeptics. After all, what do the benefits of powerful, clean, safe, and efficient nuclear energy matter if the cost is prohibitively high?
As it turns out, the economic case against nuclear energy is faulty. An analysis of the metrics used reveals serious flaws with those methods, misleading conclusions about nuclear energy, and unrealistic assumptions about potential alternatives.
Lazard, a leading investment and asset management firm, uses Levelized Cost of Energy (LCOE) to estimate the average cost of various forms of energy. Lazard found that utility-scale solar and wind is around $40 per megawatt-hour, while nuclear plants average around $175. Because LCOE is often used to argue for renewables and against nuclear (Lovins and Reuters both use LCOE in the articles referenced above), it requires closer examination.
Mark Nelson, environmentalist and managing director of Radiant Energy Fund, explains that LCOE was developed as a tool to describe “the cost of energy for power plants of a given nature.” But this tool fails when it attempts to compare the different energy sources needed to provide reliable, 24/7 electricity supply.
“[T]he cost and performance of an electricity grid is dominated by the ‘extremes’ and the worst case,” Nelson says. “And what are the extremes? Extreme shortages of supply. Extreme difficulties with combining the right generators at the right time at the right user load.”
Nelson uses the following example to illustrate the inability of LCOE to take into account the inadequacy of solar and wind: Imagine you are standing in Manhattan and need to get to London in the most cost-effective way. We would find that swimming is the cheapest! By the cost per mile of swimming, it is far cheaper than building a boat, and the infrastructure needed to use a plane would be very expensive; swimming is clearly the cheapest way to get to London. Furthermore, you can have a reasoned debate with the top experts in ocean-crossing and you can all agree that you’re using the same metric. Of course, none of you have any plans on swimming there. After all, it’s not physically possible. That doesn’t stop the experts from advocating that other people be required by government mandate to swim because it’s cheap.
Another factor that cost analyses like levelized cost of energy miss is the energy density of each form of electricity and the subsequent environmental impact of the facilities themselves. A wind facility would require more than 140,000 acres — 170 times the land needed for a nuclear reactor — “to generate the same amount of electricity as a 1,000 megawatt reactor,” according to the Nuclear Energy Institute. The institute notes that while nuclear requires 103 acres per million megawatt-hours, solar needs 3,200 acres, and wind uses up 17,800 acres.
Considering the LCOE of new sources also misses the comparatively low cost of existing generation, according to a 2019 report by the Institute for Energy Research.
“The average LCOEs for existing coal ($41/megawatt-hour), CC [combined-cycle] gas ($36/MWh), nuclear ($33/MWh) and hydro ($38/MWh) resources are less than half the cost of new wind resources ($90/MWh) or new PV solar resources ($88.7/MWh) with imposed costs included,” the report states. Imposed costs include the need to keep baseload energy like coal or natural gas idling in case the wind or solar are not producing enough energy to meet demand; such costs are often ignored by advocates of wind and solar.
Thus, levelized cost of energy misrepresents the cost of solar and wind as too low, puts nuclear energy’s costs as too high, and misses key parts of the picture.
However, the cost of nuclear power itself doesn’t need to be as high as it is in the United States. Japanese nuclear power plants only take an average of three to four years to build, from pouring concrete foundation to grid connection. French power plants mostly took between five and eight years to build.
American plants used to be built at a similar pace, before the Nuclear Regulatory Commission began to regulate the most minute aspects of construction. Contemporary American nuclear plants commonly take over a decade to build (assuming the construction plan is not simply abandoned). The NRC has a 32-step construction licensing process, and many of those steps require approval from other regulatory agencies that impose their own multi-step approval processes.
While federal, state and local agencies are legally obligated to draw up their reports within set timeframes, they routinely take significantly longer. For example, the NRC is required by law to create an Environmental Impact Statement (EIS) within two years. “However,” as author and nuclear engineer Robert Zubrin notes, “the NRC operates as if without constraint by law and actually takes an average of four years, sometimes as long as six, to write the EIS.”
What do licensing, approval, and construction time have to do with costs? Zubrin explains:
“Experience has shown that the cost of building a nuclear power plant increases roughly in proportion to the construction time squared. This is because the longer the project goes on, the more requirements, technical changes, and legal actions are levied on it… By multiplying the time it takes to complete a nuclear power plant, the antinuclear regulatory process has inflated the cost of nuclear power by two orders of magnitude.”
The Institute for Energy Research reports that it takes the NRC “an average of 80 months to approve the most recent combined construction and operation licenses. This contrasts to regulatory approval in the United Kingdom, which can be completed in about 54 months.” Furthermore, “the NRC does not provide the early feedback that would let companies properly assess regulatory risk before investing hundreds of millions of dollars in further design and development.” Meanwhile, regulatory regimes in Canada and China are able to quickly approve new, state-of-the-art projects for molten salt reactors, attracting reactor companies and leaving the U.S. in the dust.
Given that solar and wind receive almost five times the subsidies that nuclear receives and more than 50 times the subsidies (when considered in terms of dollars of subsidy received per unit of energy produced), the competition is hardly slanted in nuclear’s favor.
The problem of cost is therefore one that is both exaggerated by critics and exacerbated by overzealous regulation. In other words, not only is the problem not as bad as it is often portrayed, but there’s far more significant room for improvement.
—————
Joshua Antonini is an environmental policy intern at the Mackinac Center for Public Policy.
