Send this article to a friend:


Are Small Scale Modular Reactors Becoming Too Expensive?
Nick Lavars

According to industry reports the builders of the NuScale small modular reactor (SMR) project recently submitted revised cost estimates to their muni and co-op partners. Initial cost estimates were for power to be produced at about $58/MWh. This figure was recently revised upwards to roughly $90-$100/Mwh, a projected price increase of 60-70%. The causes cited by management for these price increases were twofold: inflation (in material costs, i.e. steel) and higher interest rates. This initial NuScale project located at the federal government’s Idaho National Laboratories in Idaho Falls would consist of six 77 MW reactors with the units slated to enter commercial service in 2029-2030. These estimates of per KWH cost are significantly above those we have seen recently for renewables plus storage.  This process occurs in this uniquely US fashion because the power buyers, the members of  UAMPS (Utah Associated Municipal Power Systems), a rural electric generation and transmission co-op, get to vote whether or not to continue with this project at predetermined points of price discovery, like the present. UAMPS is comprised of 50 small cities and electric co-ops spread across seven states in the Intermountain West. According to co-op members they will have a subsequent opportunity in about a year to again abandon their participation in this project after which the remaining participants will be financially obligated until completion. Thus far the UAMPS members have agreed to purchase 116 MWs of the 462 MW project with the remainder to be marketed to other utilities in the region. Several co-op members have already exited this project in the face of rising costs. We suspect that the remaining participants may be worried because of the high cost overruns at the Vogtle and V.C. Summer nuclear projects, even though those projects are dissimilar to NuScale’s. Their position is a bit like cruise ship financiers asked to finance a new boat shortly after the Titanic.

There is an elegant simplicity to the utility co-op structure. Raise money primarily through the sale of debt.  (Co-ops have a  thin layer of equity capital, called the “net position” and UAMPS had a negative net position at the end of its last fiscal year.) Then build a power plant. And the electricity they sell services the plant’s debt and interest expense. But if runaway construction costs lead to uncompetitive power prices, i.e. electricity that no co-op member wants or can afford, the result can be a quick march to chapter 11 and insolvency. In a way, co-ops are financially fragile by design. They can fail quickly but only if the financial ventures they support are not viable. This leads to perhaps an unintentional form of financial conservatism and a need for much quicker acknowledgement and resolution of financial error— a feature we think is highly desirable. And since co-ops like UAMPS have no equity investment to speak of, once bondholders lose interest in financing a failing or uncertain venture, the co-ops' access to vital new capital vanishes and suddenly they become financially impaired. We should add that in South Carolina, it was the co-ops led by Santee Cooper and not their investor-owned partner, SCANA, that forced cancellation of the V.C. Summer nuclear project deeming it financially unviable.

The US Department of Energy in 2020 agreed to support the NuScale project with an authorization of $1.4 billion. However, these funds must be re-approved by Congress annually. More important is that this is R&D-type funding and as the company gets closer to actual plant construction it becomes less meaningful.

The green energy-focused think tank, the Institute for Energy Economics and Financial Analysis  (IEEFA)) issued a critical report on SMR technology in February. Using historical data, the report showed that literally every recently completed nuclear project promised about 48-month completions with zero successes. Most of the units took about twice as long to complete. NuScale officials publicly responded to their IEFFA  critics suggesting their modular construction techniques and similar advances would ultimately alleviate these concerns. That’s what they said. What they really meant to our cynical ears was, “Don’t worry, this time it’s different”.

But much more interesting than a critical think tank assessment was a study of the export potential of SMR technology done by Korean officials with an eye to replicating the country’s success with gigawatt-scale reactors. As our readers know, the Koreans have sold multiple unit, large nuclear power stations to both the UAE and recently the Polish government. The report essentially asked at what price a Korean SMRs would be globally price competitive. Their answer was that longer-term --meaning after a few units were built and efficiencies realized— reactor prices for export would have to approximate an installed cost of $3,000 per kwe. For us, the most interesting thing about the Korean analysis is their conclusion that to be truly viable, SMRs will have to produce power more cheaply than their own gigawatt scale reactors designed for export. Interestingly, several years ago NuScale’s management also suggested the identical figure of $3,000/kwe as a long term target. . (We estimate that the NUScale plant in Idaho  will cost substantially more, not surprising considering that it is the first plant produced with the  new technology.)

And this goes to the heart of the SMRs dilemma from a price perspective. We all understand the concept of economies of scale, that bigger typically equals cheaper. This has certainly been true for electric power plants including nuclear. However, by reversing this long standing trend and making reactors much smaller, even though they are sold in groups, means they produce power more expensively on a per unit basis. This is like a comparison between a $200,000 bus that transports 50 versus a $40,000 car with a four passenger capacity. The capital costs for the bus are $4,000 per rider vs $10,000 per rider for the car. The smaller vehicle, despite its diminutive size, still has to have almost all the same components (engine, brakes, tires etc.)  as the larger one. It’s the same thing for SMR economics vs gigawatt scale reactors at present. The attraction of SMRs is that their smaller size, NuScale offers reactor modules in groups of 4, 6, and 12, resulting in a less expensive overall project cost because the facilities are smaller. This is of course in addition to their rapid, promised build time of three years. In a recent analyst presentation management stated that they expected to receive final US NRC approval for their designs by 2024 and could ship their first reactor module in 2027.

One caveat about economies of scale. The large plant’s estimate of economies of scale is valid only if we can accurately predict cost of construction, cost of capital, duration of construction, and state of the market upon completion.   Without knowing that information, the cost estimate verges on fiction, and the risk to the builder is largely indeterminable. With a small, modular unit, the risks are limited. The builder can stop after a small investment, if need be. A builder that wants a non-fossil base load power plant might prudently choose to build an SMR despite its supposed cost disadvantage,  and due to the risks involved, not build a full-size nuclear plant under any circumstances other than a government underwriting of the risks.  

The advantages of the NuScale project are several. In the SMR industry, NuScale will have first mover advantage and is receiving what appears to be ample government support. It has its first customer, UAMPS, and is actively seeking others both in the US and Europe. Management noted that Russia’s military aggression in Ukraine has spurred additional interest in Eastern Europe, particularly in  Poland and Romania.

From a broader perspective, NuScale is attempting to offer utilities a non-CO2 emitting, base load replacement for coal fired power generation. This poses two issues for us. The location of most coal fired power plants is determined by proximity to coal mines and rail lines for transport. Many coal fired stations are literally adjacent to mines for greater efficiency. The power produced is then transported via an extensive transmission network to load centers often hundreds of miles away. Locating new nuclear plants on these coal sites that may soon be shuttered increases regional dependence on an increasingly fragile transmission system owing to wildfires, hurricanes, and the like. This serves to reinforce a kind of status quo bias to the existing grid that to us is self defeating. If these power plants are relatively safe then locate them much closer to urban and industrial load centers and avoid potential transmission disruptions.

There is also a rural welfare aspect to new nuclear construction such that the politics of siting a new power plant resembles that of the siting for new prisons. The economic loss of a coal plant in rural areas means the elimination of well paying jobs plus the loss of tax payments to state and local governments. Re-using a brownfield site of this type has enormous local appeal, especially for rural communities where the possibility of attracting new industries is remote. This groundswell of support creates very favorable politics for new nuclear development. But again, do we want to site our newest and most advanced energy technologies the same way we locate prisons, i.e. as rural welfare? To us, this makes zero sense.

Lastly, NuScale received considerable economic benefits from the recently passed Inflation Reduction Act. On the company’s recent post earnings analyst call, the CFO stated several times that the IRA act reduced costs by 30-50% via generous investment tax credits and subsidies for hydrogen production among others. This is a substantial economic benefit.

The nuclear power industry in the US began with the promise of “too cheap to meter.” It has evolved in recent years to a situation of being too expensive to matter. SMRs like those from NuScale are trying to reverse this trend. But we think this somewhat obscures a broader issue. New nuclear can provide reliable base load power, that is, round the clock power, 24/7–unlike much cheaper but intermittent sources of power such as wind and solar. The reason we think this issue is interesting is because it focuses energy planners attention on a newer, more difficult question. What is the right balance between expensive base load versus much cheaper intermittent sources of power generation? The monopoly providers of electricity favor base load power sources for their reliability. Whether their customers ultimately agree remains to be seen. 

By Leonard Hyman and William Tilles for


Leonard S. Hyman is an economist and financial analyst specializing in the energy sector. He headed utility equity research at a major brokerage house and has provided advice on industry organization, regulation, privatization, risk management  and finance to  investment bankers,  governments and private firms, including one effort to place nuclear fusion reactors on the moon. He is a Chartered Financial Analyst and author, co-author or editor of six books including  America’s Electric Utilities: Past, Present and Future and  Energy Risk Management: A Primer for the Utility Industry. 

You can find Leonard Hyman's lastest book Electricity Acts on Amazon 



 William I. Tilles is a senior industry advisor and speaker on energy and finance. After starting his career at a bond rating agency, he turned to equities and headed utility equity research at two major brokerage houses and then became a portfolio manager investing in long/short global utility equities. For a time he ran the largest long/short utilities equity book in the world. Before going into finance, Mr. Tilles taught political science .


Send this article to a friend: