It's not that it's a lousy idea. It's been tried in Ontario, and the Ontario government just had to throw in $500 million. It's been tried in the U.S. and it's insolvent. It's been tried in the U.K., where it's sort of insolvent, and that's a more recent experiment.
The problem is that it's not clear that you can insure this risk. Because the main problem with this insurance is that if you try to charge the ones with the biggest risks, you find they're the ones who can least afford to pay, and this is a real problem when you have an insurance system where the big bill goes to the organization least able to pay the big bill. Nobody has yet found a way to do this in a way that is totally self-supporting.
Now, I have no objection if governments want to say, look, we're going to do this, we know it's a money-loser, and we know it will, from time to time, go insolvent and need bailing out; the taxpayers will bail it out and we think that's a good use of public money. That's fine. The problem with it is obvious, though, and it is that the taxpayers include many people who don't have pensions, and what they're told is that from time to time money will be used to bail out the ones with pensions.
If we want to go there, let's just be open. Most governments so far have really not been open about it. They say that they're doing this, it's going to be self-supporting, and the government won't have to put any money in. But when the hard times come, the government has to put money in.