Total cost projections for new nuclear power plants

This post originated in the middle of a great discussion on the Oil Drum. On March 1, 2007, that blog published a story titled Is Nuclear Power a Viable Option for Our Energy Needs? The story generated a lot of interest with more than 300 comments and about 678 “Diggs”. Comments are now closed on that story.

However, one of the recent visitors to that story posted a comment on a previous entry here and asked a good question about something I said during that conversation. (Sometimes putting context together on a blog can be a tortuous experience, but bear with me.)

Someone with the Internet alias of “Valuethinker” made a statement about how the 2003 study from MIT titled the Future of Nuclear Power showed that nuclear power would cost 6.75 cents per kilowatt hour with the possibility of a 25% reduction, putting it in the same cost category as wind. He went on to say that the only technology that makes sense in a market where carbon dioxide emissions are free is supercritical steam fired by coal. He even stated that the TXU announcement of their plan to build 11 coal fired power stations in Texas proved his point. I disagreed with his analysis with the following comment:

Like all economic projections, the MIT study had to make a lot of assumptions in order to compute the price of electricity from various power sources. Their methods were reasonable, but only as accurate as their ability to predict the future. Here are a couple of the assumptions that were made that have a big impact on the computed cost of nuclear power:

Capital cost of the plant – $2000 per kilowatt capacity
Construction duration – 5 years
Capacity factor – 85%
Plant lifetime – 40 years
Required return on equity – 15%
Interest rate – 8%
debt to equity ratio – 50/50

Small variations in each of the above changes the predicted number, so I dislike the idea that it is accurate to 3 decimal places.

On page 7 of the study, for example, there is an example of the kinds of sensitivity analysis that can be performed along with the resulting cost per kilowatt-hour:

Reduce construction cost 25% – 5.5
Reduce construction time 5 to 4 years – 5.3
Further reduce O&M to 13 mills/kWe-hr – 5.1
Reduce cost of capital to gas/coal – 4.2

I have reconstructed the economic model that MIT used based on the information in the report. You would be impressed by the effects of some other changes that are pretty well supported by actual experience like improving capacity factor to 90%, increasing plant life to 60 years, and shifting the debt to equity ratio to 80/20.

I am not worried about whether or not nuclear power can compete with fossil fuel economically. I do have a number of investments in various companies that should benefit by new nuclear plant construction and operation.

Rod Adams
Editor, Atomic Insights

(Interesting side note: In the intervening time since Valuethinker made his comment, TXU has backed away from their coal plant construction plans and placed an order for two large nuclear power reactors to be built by Mitsubishi Heavy Industries (MHI) of Japan in cooperation with Washington Group International.)

I had almost forgotten about that discussion until today when I found the following comment on an unrelated post:

OFF Topic:

looking at the nuclear primer on The Oil Drum, I saw your response to the MIT study estimating the cost of the nuclear kwh.
Comments are closed there, so I came here.

Assumptions are :
“Capital cost of the plant – $2000 per kilowatt capacity
Construction duration – 5 years
Capacity factor – 85%
Plant lifetime – 40 years
Required return on equity – 15%
Interest rate – 8%
debt to equity ratio – 50/50″

You speak about tweaking them. It’s fine with me. BUt had you ever try to think out of the american box?

The state as investor instead of private capital there, would translate more or less like that:
Required return on equity – 5%
Interest rate – 5%
debt to equity ratio – 95/5

It doesn’t forbid to use private contractors for some managing tasks you better have a whiff of competition in, you just need to pay the manpower it needs with the normal margin in the service industry on year to year basis.
BUt the captial????
Who said that the private sector need to reap 15% return on equity on a baseload ressource like a nuke plant, and why couldn’t the state (we the people) use its comparative advantage in long term borrowing, for a system which will be there as long as the insterstate highways?

Funny, that is more or less the way the french built their nuclear park if you think about it…
French Accounting | 03.18.07 – 6:38 am |