The New Nuclear Energy Revolution

Throwing off excessive regulatory precaution will bend the nuclear cost curve down, argues new study



Escalating costs for nuclear power plant construction in the U.S. were not inevitable, according a new study in Energy Policy. They were largely the result of the increased regulation that followed the Three Mile Island reactor meltdown in 1979.

To get a better handle on nuclear power construction cost trends, the study compares the experience of building plants in seven different countries over the past 50 years. The researchers analyze power plant construction in the U.S., France, Canada, West Germany, Japan, South Korea, and India. But before we get to that, let's take a quick back-to-the-future look at nuclear power.

In 1966, Alvin Weinberg and Gale Young of the Oak Ridge National Laboratory proclaimed that the Nuclear Energy Revolution was about to take off. And why not? After all, the cost of building nuclear power plants was declining sharply. "Nuclear reactors now appear to be the cheapest of all sources of energy," they declared. "We believe, and this belief is shared by many others working in nuclear energy, that we are only at the beginning, and that nuclear energy will become cheap enough to influence drastically the many industrial processes that use energy." Such optimism seemed more than justified. The Atomic Energy Commission—the lead nuclear regulatory agency at the time—was predicting that more than 1,000 reactors would be operating in the United States by the year 2000.

Had that transformation actually taken place, the entire U.S. 1,000-gigawatt electricity generation sector could notionally have been replaced by 1,000 nuclear power plants, each with a generating capacity of 1,000 megawatts. Among other things, this would have made the country's greenhouse gas emissions more than 30 percent lower than they currently are.

Weinberg and Young also estimated that the capital costs to build a new coal-fired power plant or new nuclear plant were both around $110 per kilowatt of generating capacity ($804 in 2015 dollars) and falling. The latest figures from the Energy Information Administration report that the overnight capital cost for a modern coal plant is $2,917 per kilowatt; for an advanced nuclear plant, it is $5,366 per kilowatt. In other words, the costs for coal plant have escalated 3.6-fold faster than the inflation rate, 6.7-fold faster for nuclear. In comparison, the capital costs for a conventional natural gas power plant stands now at $912 per kilowatt.

Interestingly, Congress in 1978 passed the Powerplant and Industrial Fuel Use Act, which actually prohibited the use of natural gas or petroleum as energy sources in any new electric power plants; it also mandated that all new electric power plants have the capability to burn coal or any alternate fuel as a primary energy source. With copious supplies of cheap natural gas made possible by fracking shale, scores of coal-fired plants have been converted to burning natural gas, and utilities have announced the closure of 200 out of the nearly 600 coal-fired generation plants in the U.S. Absent a price on carbon dioxide emissions, cheap natural gas poses a big challenge to any efforts to build new nuclear power plants.

But set that aside for the moment, and consider the new Energy Policy study.

The authors—Jessica Lovering, Arthur Yip, and Ted Nordhaus, all associated with the Breakthrough Institute—focus chiefly on overnight construction cost trends. The overnight construction costs are the dominant component for the lifetime costs of nuclear power; they include money spent on such things as land, site preparation, engineering, procurement, and construction services. Like Weinberg and Young, the Breakthrough Institute writers find that in the mid-1960s, the capital costs for nuclear construction had been steeply declining, falling to around $600 to $900 per kilowatt of generation capacity in current dollars. But that proved to be the nadir in U.S. nuclear construction costs.

The late '60s and early '70s saw a plethora of new environmental regulations. For example, in 1971 the District of Columbia Federal Court of Appeals ruled in a case brought by the Sierra Club and the National Wildlife Federation that the Atomic Energy Commission must change its rules to conform to the new National Environmental Policy Act's requirement to consider the environmental impact of each new power plant. In addition, accidents such as a fire at the Brown's Ferry reactor and a rapid reactor cool-down due to an electrical short at the Rancho Seco plant led to tightened, and more expensive, safety requirements and equipment back-fits. As a result, construction costs more than doubled during the 1970s to a range of $1,800 to $2,500 per kilowatt of generating capacity. Some enhanced safety measures were clearly called for, but federal regulators arguably overdid it.

The analysis of the construction costs for the 51 U.S. reactors completed after the Three Mile Island meltdown is divided into three groups. For plants that had received their operating licenses before the accident, costs mildly escalated to a range of $1,800 and $3,000 per kilowatt of generating capacity. Construction costs for 38 of the plants averaged between $3,000 and $6,000 per kilowatt, and 10 reactors eventually cost between $6,000 and $11,000 per kilowatt. Overall, the researchers find that after Three Mile Island, increased regulatory uncertainty and mounting technology back-fit requirements substantially boosted construction costs. As utilities saw costs climbing, orders for 120 reactors were canceled between 1972 and 1982, and there were no nuclear plant construction starts in the U.S. from 1978 until 2013.

Things proceeded differently abroad. As in the U.S., nuclear construction costs steeply declined in France, West Germany, and Canada during the 1960s. In the 1970s, their costs began to rise but not nearly at the American rate. Basically, their inflation-adjusted costs doubled whereas U.S. costs have more than sextupled. In other words, the trend in nuclear construction costs in these countries is similar to the cost increases for building new coal-fired plants in the United States.

In Japan, nuclear plant construction costs doubled in the 1970s and then remained flat for the next 30 years. In India, nuclear construction costs rose 150 percent between 1976 and 1990 but have since fallen by 10 percent. And in South Korea, the cost of building nuclear plants has fallen by 50 percent since 1972. America's steeply rising cost trend for nuclear construction is clearly the outlier.

As an aside, the researchers note that the investment prices for solar and wind in the U.S. rose steeply between 2002 and 2008. For example, the capacity-adjusted price of a wind turbine rose during that period by 100 percent. "Rather than predict that wind power cost would continue to rise or that the technology was inherently expensive," they tartly observe, "most analysis focused on understanding the drivers of these cost increases." In this case, price increases were driven by rising steel costs, labor expenses, diseconomies of scale, and currency fluctuations.

So what accounts for the difference between the U.S. experience and those of the other countries? For one thing, countries that had stable or falling costs generally standardized their reactor designs and often built two plants at the same site. "How costs evolve over time appears to be dependent on different regional, historical, and institutional factors at play," the authors argue. "Factors such as utility structure, reactor size, regulatory regime, and international collaboration may play a larger effect. Therefore, drawing any strong conclusions about future nuclear power costs based on one country's experience—especially the U.S. experience in the 1970s and 1980s—would be ill-advised."

Can the U.S. nuclear construction cost curve be bent downward again? And more importantly, can new nuclear power generation cost less than current fossil fuel alternatives, especially natural gas? Presentations at the recent Advanced Nuclear Summit in Washington, D.C., suggest that this might be possible. At the very optimistic end, the nuclear start-up ThorCon claims that the capital costs for its molten salt reactor would amount to $700 per kilowatt of capacity. Less optimistic analyses for new reactor designs put the costs at around $2,000 to $3,400 per kilowatt. This is comparable to building a coal-fired plant, but considerably more pricey than natural gas plant construction. On the other hand, Terrestrial Energy estimates that, if fuel costs are taken into account, its molten salt reactor would produce electricity at a lower cost than do natural gas plants.

In any case, throwing off excessive regulatory precaution to speed up the approval of new advanced nuclear power plant designs would go a long way toward finding out which energy sources are ultimately cheaper and safer for people and the planet.