6 ms·

> The reason is that the costs of a failure of a nuclear power station are such that no insurance company would take on such a risk on commercial basis and even

by cbsmith 14y ago

> The reason is that the costs of a failure of a nuclear power station are such that no insurance company would take on such a risk on commercial basis and even if they did, no energy company would be able/willing to pay the premiums. I'm not sure how you can assert this to be true. Because it is a rare event, the premiums won't be that obscene, particularly if the carrier isn't carrying related policies. The expense of a nuclear plant failure is most certainly high but, for example, can't match the impact of say Katrina (and while no insurance company was carrying all the impact of Katrina, several of them bore far more than the costs of pretty of completely destroying an entire major city, which is more than any nuclear power plant failure). Certainly the money to be made from producing that much energy that cheaply outweighs the insurance costs. Ironically, I'd argue this is part of the problem with nuclear energy. If left completely unregulated, I could see that a typical nuclear power station would be built with quite spare safety precautions, with the expectation that it likely would fail within a certain period of time, with an insurance policy in place to cover the expenses when that occurs.

SagelyGuru 14y ago
https://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disaster#Insurance https://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disa... It turns out that Fukushima was 'insured' but the insurers won't be paying anything. It is all excluded, see? To prove your assertions, you need to come up with an example where a major nuclear disaster was actually cleaned up at the insurers expense. Till then, I stand by my statements.
- cbsmith 14y ago
  How many nuclear power plants are their in the world? How many have had failures from "earthquake shock, fire following earthquake and tsunami"? How many failures have occurred for other reasons? From that standpoint it is hard to argue that the premiums would have been significantly different without that clause. I'd also point out that such a clause is hardly unusual in the insurance business. You'll see the like on all kinds of property and liability policies. To prove your assertions, you'd need to come up with a nuclear reactor which was constructed in an unregulated environment. ;-) Look, I'm not making claims one way or another. I'm simply suggesting that your claim doesn't seem very credible once one has applied some very simple logic. I could be missing something, but I think you ought to have some evidence to back up such an extraordinary claim. As someone else pointed out, your typical coal operation ought to have a higher incident rate and cost, which means it ought to be even more expensive... Realistically, you can calculate the costs without coming up with your silly example. It is sufficient to simply add up the costs needed to cover the disaster, and then look at the premiums paid for covering similar sized disasters with similar rates of occurrence, making your best effort to extrapolate over any gaps. Alternatively, speak with an actuarial who works in the business and they can extrapolate what the premiums would be. In a truly unregulated market, I'd imagine the incident rate would be higher and the recovery costs a bit lower, so it might be hard to get a real cost, but go for the costs in the existing regulated market. I have a hard time imagining they'd be prohibitive, but I'd be intrigued if you proved otherwise.
  - moe 14y ago
    How many nuclear power plants are their in the world? 436 plus 63 in construction (as of 2010). It is sufficient to simply add up the costs needed to cover the disaster The total cost for the Chernobyl incident was estimated at $235 billion[1] in 2006 and the figure only keeps growing. Since 2007 they're building a new sarcophagus (because the old one is falling apart). The costs were estimated at another $1.4 billion for that alone. But they're late already, and well, you know how it goes. In contrast Hurricane Katrina was a bargain at $150 billion dollars [2]. The Fukushima incident is estimated to cost $257 billion dollars [3] and that's probably a little optimistic. and then look at the premiums paid for covering similar sized disasters with similar rates of occurrence See, here is your problem. At these scales there is no coverage and no reference; these are disasters of national scale. For Katrina the insurance industry paid $41 billion dollars [2], guess who carried the rest. And the real question is: What happens when some confused individuals somehow manage to smuggle nail-clippers onto passenger planes and then fly them into multiple reactors, in densely populated areas, at the same time? What is the insurance premium on that? [1] http://www.greenfacts.org/en/chernobyl/l-3/5-social-economic-impacts.htm http://www.greenfacts.org/en/chernobyl/l-3/5-social-economic... [2] http://en.wikipedia.org/wiki/Economic_effects_of_Hurricane_Katrina http://en.wikipedia.org/wiki/Economic_effects_of_Hurricane_K... [3] http://www.reuters.com/article/2011/12/06/japan-nuclear-cost-idUSL3E7N60MR20111206 http://www.reuters.com/article/2011/12/06/japan-nuclear-cost...
    - poutine 14y ago
      "Probably not much" to answer your real question: http://www.youtube.com/watch?v=--_RGM4Abv8 http://www.youtube.com/watch?v=--_RGM4Abv8 The US government has done the test, granted an F4 isn't a passenger jet but it probably has similar kinetic energy to a smaller passenger plane as it was going rather fast.
      - moe 14y ago
        probably has similar kinetic energy to a smaller passenger plane The most widely used passenger plane is the Boeing 737. F4 Phantom empty weight: 29,500 lbs Boeing 737 empty weight: 62,000 lbs The average operating weight difference is probably quite a bit larger. You know, fuel, baggage, passengers. You may want to review those videos from 9/11 to see what a passenger plane does to a building. You may also want to listen to the narrative of your video: The wall in your video is a specially hardened wall. Your reactors are not equipped with such walls because that would be cost prohibitive. And even these hardened walls are unlikely to withstand a passenger plane (ever wondered why the US government selected such a small plane for your video?).
caf 14y ago
It is demonstrably true that no insurance company is willing to take on the full risk of insuring a nuclear power plant against disaster. The reason for this is that insurance companies aren't gamblers - they don't like insuring against minutely-likely but stratospherically-high-cost events, even if they can charge a premium that gives them a positive expected value. That's because they have to factor in that with such a high payout, there is a risk of destroying their business altogether. You can't capitalise on the positive expected value over 1000 years if you went bankrupt in year 3, even if that was just due to "bad luck". This is just like the understanding that poker players have, that you need a considerably larger bankroll than the stakes you are playing in order to ride out "variance", as it's known. Insurance companies understand risk.
- cbsmith 14y ago
  It's worth noting that it is common practice for potential large claims capable of sinking a single insurer to be distributed over the wider market. Other than that, you make a good point, and history bares it out. Insurance companies aren't gamblers, they aren't willing to support minutely-likely but stratospherically-high-cost events, even if they can see high profit margines. Bottom line: they really understand risk. http://en.wikipedia.org/wiki/American_International_Group#Financial_crisis http://en.wikipedia.org/wiki/American_International_Group#Fi... Note that the $441 billion was just for a small division inside of AIG. The bigger divisions are obviously capable of handling more.
justincormack 14y ago
The problem is that being insured reduces incentives to not make a mistake. Nuclear companies are already often better at covering things up (look at Tepco, and Chernobyl for extreme cases, but also the ongoing poor disclosure in the uk). Thats why insurance wont happen, not some actuarial issues about probability.

by cbsmith 14y ago

SagelyGuru 14y ago

https://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disaster#Insurance https://en.wikipedia.org/wiki/Fukushima_Daiichi_nuclear_disa... It turns out that Fukushima was 'insured' but the insurers won't be paying anything. It is all excluded, see? To prove your assertions, you need to come up with an example where a major nuclear disaster was actually cleaned up at the insurers expense. Till then, I stand by my statements.

cbsmith 14y ago
How many nuclear power plants are their in the world? How many have had failures from "earthquake shock, fire following earthquake and tsunami"? How many failures have occurred for other reasons? From that standpoint it is hard to argue that the premiums would have been significantly different without that clause. I'd also point out that such a clause is hardly unusual in the insurance business. You'll see the like on all kinds of property and liability policies. To prove your assertions, you'd need to come up with a nuclear reactor which was constructed in an unregulated environment. ;-) Look, I'm not making claims one way or another. I'm simply suggesting that your claim doesn't seem very credible once one has applied some very simple logic. I could be missing something, but I think you ought to have some evidence to back up such an extraordinary claim. As someone else pointed out, your typical coal operation ought to have a higher incident rate and cost, which means it ought to be even more expensive... Realistically, you can calculate the costs without coming up with your silly example. It is sufficient to simply add up the costs needed to cover the disaster, and then look at the premiums paid for covering similar sized disasters with similar rates of occurrence, making your best effort to extrapolate over any gaps. Alternatively, speak with an actuarial who works in the business and they can extrapolate what the premiums would be. In a truly unregulated market, I'd imagine the incident rate would be higher and the recovery costs a bit lower, so it might be hard to get a real cost, but go for the costs in the existing regulated market. I have a hard time imagining they'd be prohibitive, but I'd be intrigued if you proved otherwise.
- moe 14y ago
  How many nuclear power plants are their in the world? 436 plus 63 in construction (as of 2010). It is sufficient to simply add up the costs needed to cover the disaster The total cost for the Chernobyl incident was estimated at $235 billion[1] in 2006 and the figure only keeps growing. Since 2007 they're building a new sarcophagus (because the old one is falling apart). The costs were estimated at another $1.4 billion for that alone. But they're late already, and well, you know how it goes. In contrast Hurricane Katrina was a bargain at $150 billion dollars [2]. The Fukushima incident is estimated to cost $257 billion dollars [3] and that's probably a little optimistic. and then look at the premiums paid for covering similar sized disasters with similar rates of occurrence See, here is your problem. At these scales there is no coverage and no reference; these are disasters of national scale. For Katrina the insurance industry paid $41 billion dollars [2], guess who carried the rest. And the real question is: What happens when some confused individuals somehow manage to smuggle nail-clippers onto passenger planes and then fly them into multiple reactors, in densely populated areas, at the same time? What is the insurance premium on that? [1] http://www.greenfacts.org/en/chernobyl/l-3/5-social-economic-impacts.htm http://www.greenfacts.org/en/chernobyl/l-3/5-social-economic... [2] http://en.wikipedia.org/wiki/Economic_effects_of_Hurricane_Katrina http://en.wikipedia.org/wiki/Economic_effects_of_Hurricane_K... [3] http://www.reuters.com/article/2011/12/06/japan-nuclear-cost-idUSL3E7N60MR20111206 http://www.reuters.com/article/2011/12/06/japan-nuclear-cost...
  - poutine 14y ago
    "Probably not much" to answer your real question: http://www.youtube.com/watch?v=--_RGM4Abv8 http://www.youtube.com/watch?v=--_RGM4Abv8 The US government has done the test, granted an F4 isn't a passenger jet but it probably has similar kinetic energy to a smaller passenger plane as it was going rather fast.
    - moe 14y ago
      probably has similar kinetic energy to a smaller passenger plane The most widely used passenger plane is the Boeing 737. F4 Phantom empty weight: 29,500 lbs Boeing 737 empty weight: 62,000 lbs The average operating weight difference is probably quite a bit larger. You know, fuel, baggage, passengers. You may want to review those videos from 9/11 to see what a passenger plane does to a building. You may also want to listen to the narrative of your video: The wall in your video is a specially hardened wall. Your reactors are not equipped with such walls because that would be cost prohibitive. And even these hardened walls are unlikely to withstand a passenger plane (ever wondered why the US government selected such a small plane for your video?).

caf 14y ago

It is demonstrably true that no insurance company is willing to take on the full risk of insuring a nuclear power plant against disaster. The reason for this is that insurance companies aren't gamblers - they don't like insuring against minutely-likely but stratospherically-high-cost events, even if they can charge a premium that gives them a positive expected value. That's because they have to factor in that with such a high payout, there is a risk of destroying their business altogether. You can't capitalise on the positive expected value over 1000 years if you went bankrupt in year 3, even if that was just due to "bad luck". This is just like the understanding that poker players have, that you need a considerably larger bankroll than the stakes you are playing in order to ride out "variance", as it's known. Insurance companies understand risk.

cbsmith 14y ago
It's worth noting that it is common practice for potential large claims capable of sinking a single insurer to be distributed over the wider market. Other than that, you make a good point, and history bares it out. Insurance companies aren't gamblers, they aren't willing to support minutely-likely but stratospherically-high-cost events, even if they can see high profit margines. Bottom line: they really understand risk. http://en.wikipedia.org/wiki/American_International_Group#Financial_crisis http://en.wikipedia.org/wiki/American_International_Group#Fi... Note that the $441 billion was just for a small division inside of AIG. The bigger divisions are obviously capable of handling more.

justincormack 14y ago

The problem is that being insured reduces incentives to not make a mistake. Nuclear companies are already often better at covering things up (look at Tepco, and Chernobyl for extreme cases, but also the ongoing poor disclosure in the uk). Thats why insurance wont happen, not some actuarial issues about probability.