Friday, March 28, 2014

Did Three Mile Island produce the most expensive power in history?

Three Mile Island
Today is the 35th anniversary of the accident at Three Mile Island, the worst nuclear mishap in U.S. history. In world consciousness it ranks with Chernobyl and Fukushima, and it happened about 25 miles west of where I'm sitting right now.

TMI Unit 2 began commercial operations on Dec. 30, 1978. That means it was producing power for consumers for just under 90 days before catastrophe struck on March 28, 1979. Basically, it was the nuclear power equivalent of a teenager buying a Maserati then crashing it the next day.

How bad was it? Unit 2's total cost was $700 million, according to the "TMI Facts & Figures" (PDF) put out by GPU Nuclear, the utility that built and operated it. Spread that over 90 days, you're looking at $7.78 million a day. Even for a power utility, that's not chump change.

We can also spread that construction cost factors over the amount of power Unit 2 produced in its short life. According to my back-of-the-envelope calculations, Unit 2's power cost $2,640 per megawatt-hour, or $2.64 per kilowatt-hour. That's about 30 times more than my local utility PPL's current price to compare of 8.75 cents.

To be honest, I expected that result to be far worse than it is. Still, 30 times is a lot. If you're an average homeowner in PPL's service area, paying $80 a month (U.S. households typically use about 900 kilowatt-hours a month, says the U.S. Energy Information Administration), you wouldn't want that to shoot up by a factor of 30 to $2,400, would you? Hawaii's electricity costs are considered pretty outrageous, but even there, the stuff is only about 35 cents per kilowatt-hour.

Plus, the "price to compare" includes not only generation, but transmission. According to the EIA (it's in a spreadsheet, so no link), the cost ratio is about 85/15. So the generation portion of the price to compare is 85 percent of 8.75 cents, or 7.44 cents. Apples to apples, then, the cost of Unit 2's power was about 35 times PPL's present generation price.

In short, I think there's a pretty good chance that, yes, the electricity produced at TMI's Unit 2 was the most expensive ever generated.

The math that yielded that $2.64 figure is a little tedious, so you can stop reading now if you don't care how I reached it. If you do, take a deep breath, and let's carry on.

The calculation

Unit 1's capacity was about 900 megawatts, according to GPU. Operate a 900-megawatt plant for one hour, and you've generated 900 megawatt-hours. There are 24 hours in a day, and Unit 2 operated for roughly 90 days. So we have 900 x 24 x 90 = 1,944,000 megawatt-hours, max.

Divide $700 million by 1.944 million and you end up with $360 per megawatt-hour. But that was 1979. Scaling up with the Bureau of Labor Statistics handy inflation calculator, that works out to $1,164 per megawatt-hour in 2014 dollars.

Add in operation, maintenance or fuel costs, which according to the U.S. Energy Information Administration, were $25.48 per megawatt-hour in 2012, and you get a total of $1,189.48. One megawatt is 1,000 kilowatts, so that works out to $1.19 per kilowatt-hour, or nearly 14 times PPL's current "price to compare" of 8.75 cents.

But wait, there's more! The initial cleanup of Unit 2 after the accident took more than a decade and cost $973 million. Treating those as 1988 dollars for inflation-adjustment purposes (the cleanup stretched from 1979 to 1993, so using 1988 yields a conservative figure), that's another $1.9 billion in 2014 dollars. Plus, there's more cleanup and the decommissioning yet to come, estimated at $837 million as of 2009. Chalk up another $915 million after inflation. So we have $1.9 billion + $915 million = $2.815 billion. Divide by the 1,944,000 megawatt-hours that Unit 1 produced, and we get $1,448 per megawatt-hour. So our final cost is $1,448 + $1,189 = $2,637 per megawatt-hour, or $2.64 per kilowatt-hour, which is 30 times PPL's price to compare.

Let me stress,  by the way, the "back-of-the-envelope" nature of all of this. I'm seeking a plausible general order of magnitude here, not a result that's exact down to the penny.

No comments:

Post a Comment