Quantifying the Price-Anderson "subsidy"

It is difficult to get into an internet conversation about the economic advantage of nuclear power without having someone make the claim that nuclear plant operators could not be economically competitive without the Price-Anderson insurance “subsidy” from the federal government.

My general response has always been that it is hard to call Price-Anderson a subsidy when the US government has never had to spend any money paying a claim under that act. However, I always supposed that there was some way to quantify the value provided by having a federally arranged insurance pool that allowed plant owners and their financial sponsors to determine the general amount of their liability in the case of an accident.

Yesterday, Eric McErlain at NEI Nuclear Notes provided a link to a rather typical anti-nuclear comment titled Nukes can never be the answer posted on a blog called Shakesville. Of course, I cannot resist taking a look at a post with that kind of provocative title, so if you want to add to your site visits, there is a hint. (Eric’s post and the link can be found under the title of Another Debate Over Nuclear Energy).

Though the author of the post catalogued a familiar list of complaints about nuclear power, he also provided a valuable service by pointing to an article from the Stanford Institute for Economic Policy Research titled Does the US Subsidize Nuclear Power Insurance. Here is how the “quixote” described the Price-Anderson subsidy.

Nukes are dangerous. We all know that. What fewer people are aware of is that they are so dangerous they are uninsurable. If companies actually had to pay the amount of money it would take to induce an insurance company to take on the risk, the power generated would be eyepoppingly expensive.

The problem is solved by a neat hat trick called the Price-Anderson Act. First passed in 1957, it limited liability for reactor-owning companies. Currently, the limit that any one company must pay out is $95.8 million. (The direct costs of cleaning up Three Mile Island in 1979 were near one billion. That does not include cleaning up the dead plant itself, which is now a mothballed no-man’s land. It does not include the cost of storing the mountains of extra waste. It does not include any indirect medical, real estate, or other costs.) The government maintains a fund that companies must pay into in the event of an accident, which is capped at 9.5 billion dollars. Any costs above that belong to the federal government, i.e. taxpayers. Any costs the government can avoid belong to the people who’ve been harmed. You are the insurer of last resort for the nuclear industry. That subsidy has been valued at about $2.3 million per reactor per year (pdf). Unfortunately, nobody’s paying you your share.

As I have frequently complained, it is disheartening how many people either cannot “do the math” or who refuse to do it. Even if I assumed that the link would lead me to an article with an anti-nuclear slant, I could tell that there was no way that getting a computed “subsidy” of $2.3 million per plant per year would have much impact on the cost or profitability of a typical US reactor plant. It would certainly not change nuclear generated electricity from the bargain that it is into something that is “eyepoppingly expensive”.

After all, a 1000 MWe nuclear power plant operating at an average capacity factor (90%) and selling its power for an average price (8.5 cents per kilowatt hour in the US in 2006, based on information from the Energy Information Agency) produces about $744 million in revenue each year. If the plant had to pay an extra $2.3 million per year for insurance, its cost per kilowatt hour would increase by only .02 US dollar cents per kilowatt hour, raising the total average cost of power from a US nuclear plant in 2006 to 1.74 US dollar cents per kilowatt-hour ($138 million per year instead of $135.6 million per year.)

Maybe there is a reason that anti-nuclear commentators avoid math – it makes them look pretty foolish.

After running those numbers, I decided to investigate the link provided to the analysis to determine its value. After reading the article (highly recommended for pro-nuclear debaters) I realized that “quixote” apparently avoids reading as well. (Perhaps he has been “left behind” by America’s educational system.)

Here is the conclusion Geoffrey Rothwell, the Stanford economics professor who wrote Does the US Subsidize Nuclear Power Insurance, reached at the end of the article.

While there is the potential of federal payments to nuclear accident victims, there has been no direct subsidy of the nuclear power industry through the PAA.

Instead, his analysis shows that the public is better protected by having Price Anderson in place since it provides more compensation that would be available by the normal liability practice for companies engaged in “hazardous” enterprises. Normally, what those companies do is to purchase as much insurance as they can and self-insure for the rest. The limit to that liability is the complete value of the firm – if damages exceed the company’s ability to pay, it becomes bankrupt. For many nuclear operating companies, their full market capitalization is less than the current Price-Anderson limit.