Nuclear Electricity Makes No Sense.
By Russell Lowes, 11/18/2014
The Obama administration is already doing all it can realistically do. Despite its “all-of-the-above” façade, it favors nuclear power. To start with, the Energy Department is essentially a nuclear department. Professor Moniz is [was] Secretary because of his nuclear ties. DOE’s national laboratories are basically nuclear labs. It organizes international nuclear R&D groupings to encourage worldwide commitment to nuclear power. The Obama administration has created an inter-departmental Team USA, including State and Commerce, specifically to encourage domestic nuclear industry by promoting nuclear exports. The White House dedicates a staffer to this task. Secretary Moniz emphasizes his commitment to “jumpstart” the U.S. nuclear power industry. DOE subsidizes new domestic nuclear plants through loan guarantees. The nuclear Navy provides government-trained operating personnel. And to facilitate the licensing of new plants, and extend licenses for existing ones, the administration’s appointments to the Nuclear Regulatory Commission have ensured that it remains industry-friendly.
--Victor Galinsky, ex-NRC Commissioner, National Journal, February 2014
We keep hearing from certain people that nukes are essential to solve energy and global warming problems. They say that nuclear energy is carbon-free, or some say low-carbon. They are neither. They say that nuclear is low-cost. They say building another round of nuclear reactors is essential for the U.S. and the world. It is neither low-cost nor essential. To build more megawatts of nuclear energy would be a mega-distraction.
Such an emphasis would weaken our response and ability to stem future climate chaos. I will take on the mission here of showing how the horrendous costs of nuclear energy makes this source an unpractical one. It is especially unpractical now, during our quest to truly course-correct on climate change.
The bottom line is that electricity generated from new nuclear reactors is about 24 cents per kilowatt-hour. About this 24 cents per kilowatt-hour:
1) This is double the electricity price for the U.S. on average .
2) The cost of 24¢ for nuclear electricity is more than twice the 10¢ cost of solar electricity in Arizona, about twice the national average for solar.
3) It is more than twice the cost of wind-generated and delivered electricity.
4) Most important, nuclear electricity is 8 times the 3¢ national average cost of energy efficiency.
5) It is about twice the cost of new coal and gas-generated electricity.
You might ask, well how do we know how expensive a reactor will be? We have nuclear plants scattered across the nation, so how much did these plants cost in the last round?
First, I have been using empirical analysis of the cost of nuclear energy since 1977. We used regression analysis in a book released in 1979. This book was instrumental in convincing investors to pull out of the Palo Verde Generating Station Units 4 & 5, America's largest nuclear plant, west of Phoenix. Our analysis projected the cost of the Palo Verde to be $6.1 billion in 1986 actual completion dollars. The managing utility company, Arizona Public Service Co. (APS), projected $2.8 billion at the same time, and they never waivering on its projection until construction was well under way.
That down-graded plant of 3 reactors was finished for $5.9 billion. The APS projection was overrun in costs by 111%, while our projection was slightly over the final cost by less than 4%. Of all the reactor projections done across the land that we could find, ours was the most accurate nuclear reactor projection in the nation.
We used empirical approach to costing reactors, with regression and other modeling techniques. Apparently APS used the tried and true method of sales pitch estimation.
So how do we jump from then, when the final reactor at PVNGS was completed in 1986 to now? The method I use is four-fold.
1) First, find out what the average cost of the last rush of reactors, which happened around 1987;
2) Then apply general inflation to that cost to bring it up to today’s cost;
3) Third, apply a projected inflation to the year that a new reactor might be completed; and
4) Finally, weigh a series of factors that might increase or decrease this figure.
For step 1, a low/conservative estimate on reactor average cost for 1988 was $3100 per kilowatt of net plant size.
Putting that $3100 into 1987 dollars at the U.S. Bureau of Labor Standards inflation calculator yields $6105 per kilowatt of electrical capacity in 2013 dollars.
For Step 3, I project a common 4% inflation rate through 2022, the first year it is likely for the next small group of reactors in the U.S. to be completed. This yields a completion cost in 2022 of $8689/kWe.
For Step 4, I have come up with a survey of 27 reactor construction cost factors. This is the most varied and numerous list of items I have seen, so far, from all my reading on reactor costs. I estimate that the reactors of the early 2020s will cost about 20% more than the reactors finished in the last big wave of the mid-late 1980s.
In this 4th step, I have considered factors that would make nukes cheaper than in the real (inflation adjusted) dollars of the past, like labor cost declines in America. I have also taken into consideration factors that would increase the costs like certain material cost increases, and increases in plant robustness requiring more cement, copper, steel, etc.
After comparing the changing conditions since the time the last reactors were completed, I have come to what I consider a fairly accurate projection. It probably won’t be as accurate as our PVNGS <4% accuracy level, but I am fairly sure it will be in the ball park.
After going through this process, the final figure I project for the next round of nukes built in 2022 is $9149/kilowatt of plant size. This is in sharp contrast to most sales pitches from utilities today, where they project more like $4000 per kWe. It would be good to remember that the average overrun was 220% in the last round. They sell these plants by unrealistically lowballing the construction cost.
What does that come out to in cost per kilowatt-hour? Just like with solar and wind, you can break this down to the kilowatt-hour of electrical capacity (kWe) level, and then apply production time (hours) to it to get kilowatt-hours of electricity delivered (kWhe). You can also multiply these kWe units to the typical sizes of the wind turbines, solar panels, or coal or nuclear plants.
Here are the calculations.
This is what it would cost roughly, to install 100 reactors in the U.S., a figure being brought up from time to time by members of Congress.
X 1,350,000 kWe plant size
= $12.351 billion
X 100 reactors occasionally proposed
= $1.2351 trillion total construction cost for 100 reactors
X 14% loan payback per year (capitalization rate)
= $172.9 billion per year for 30 years
X 30 years
= $5.187 trillion paid just for construction and loan and tax expenses, not counting fuel or operation & maintenance, nor transmission and distribution.
That $172.9 billion/year will cost the average person in the U.S. (assuming an average of 350 million people into the future):
$494/person/year for 30 years if we have a 350 million population, or
$988/taxpayer/year if we have 175 million taxpayers.
So, how do we get to cost per kilowatt-hour? For each kilowatt of plant capacity, you can calculate the cost to construct, the capital cost and then calculate the electricity the plant produces over a typical 40 years (before major costs of renovation add to the equation). Then simply divide the capitalization cost by the kWhe. Here we go (simply). . .
Cost Portion of the Equation:
X 14% capitalization rate =
$1,281 in capital cost/year
X 30 years
= $38,426 capital payback over 30 years for each kWe of size – This is just the total capital cost over 30 years.
Electrical Output Portion of the Equation:
X 8766 hours/year on average
X 85% average capacity factor (electrical performance) over the life of the reactor
X 40 years
= 298,044 kWhe over 40 years – THIS is the e output over 40 years. Note that the capital payback is 30 years and the plant runs for a projected 40 years (before major capital upgrade, if it runs longer).
The Final Capital Cost/kWhe Calculation:
$38,426 Capital cost over 30 years per kilowatt of installed electrical capacity
/ 298,044 kWhe e output over 40 years
= 12.9¢ per kilowatt-hour of electricity.
There was a multi-disciplinary report put together by the nuclear industry, along with governmental and non-governmental entities called the Keystone Report.
This report projected fuel and operations and maintenance costs at:
4.3¢ per kWhe for fuel and O&M. That, plus. . .
+ 12.9¢ capitalization cost
= 17.2¢ production cost (pre transmission & distribution)
+ 7.0¢ per kWhe for transmission & distribution
= 24.2¢ per kilowatt-hour to your meter
What are the implications of such a high cost to your household, and to the larger society, the U.S. in this case?
I’ll leave that up to your imagination, as you ponder that solar is currently less than half the cost, while it continues its cost plunge, energy efficiency is about one eighth the cost and wind is also about half the cost. Getting back to Victor Galinsky’s quote from the beginning, the only way in which nuclear energy can compete in the market is in a skewed way, with the U.S. Government favoring it all the way along. That in fact is how nukes have gotten as far as they have. It’s time to nuke the nuclear option!