3 ms·

> And most people won't buy deposit insurance at any price higher than the "free" one they get from their government. It's a product that can't really be sold,

> And most people won't buy deposit insurance at any price higher than the "free" one they get from their government. It's a product that can't really be sold, because the only people buying would be fraudsters. Suppose a new bitcoin "bank" came to an insurer and said "we want to buy insurance against losing any of our customers' money". The insurer would, quite rightly, assume that the business plan here was to steal all the money and put the loss on the insurer. The insurer would assume that to even consider writing such a policy, they'd need to verify and stand behind every single aspect of the "bank"'s business operations, website, and so forth. The insurer would quickly realize that the premiums for such a policy would have to be extravagant, like 100% of the amount insured. And the insurer would decline the business. The FDIC insurance "works" because of the large body of regulation and the mandatory enrollment requirement, neither of which exist in bitcoin-land. There isn't insurance for bitcoin banking, and can't be. (Well, I suppose someone could offer fraudulent insurance - take premiums but have no intention of paying out. But that's just fraud, not insurance.)

fauigerzigerk 13y ago
I think a deposit insurance scheme could work very well if all (remaining) big players came together and decided collectively to create one. It would be widely publicised and it would be made very clear that no one should trust a bitcoin bank that is not part of the scheme. Such a scheme would have to include mandatory auditing of finances, QA and security. So what I'm proposing is industry self regulation. That said, I doubt that any deposit insurance scheme could ever be large enough to cover losses of such enormous proportions as we have seen in recent weeks. The hope would be that utterly incompetent adventurers like Mt. Gox would never pass the audit.
- jellicle 13y ago
  If you think a bit about it, it can't work. A bank can verify, on a daily basis, that they still have "most" of their assets. And a lot of the assets can't be stolen in any real way. Banks owns a house - house can't be stolen. The legal process will get it back. Bank owns treasury bonds - can't be stolen, we know who is "supposed" to own them. Etc. Occasionally there may be an unexpected asset leak (rogue trader or something). But the damage is still limited, it's still small compared to the assets of the bank. Barings lost something like 20% of all their assets - obviously disastrous, but they still had 80% left. And a failure that large is intended to be impossible to occur. A bitcoin "bank", well, you audit it at 11AM and at 11:02AM it can have nothing, 0, zero. No insurer is going to sign on for that sort of risk at any price.
  - fauigerzigerk 13y ago
    >No insurer is going to sign on for that sort of risk at any price. One already has: http://www.bbc.com/news/technology-25680016 http://www.bbc.com/news/technology-25680016 But I'm not suggesting getting insurance from insurance companies. I'm suggesting that every bitcoin bank should set aside some small share of its revenues to fund the industry's own deposit insurance scheme. That's how deposit insurance works in many countries. If the losses exceed the fund's resources then it's too bad for the depositors. Insuring bitcoin banks (or rather storage companies) is not actually that problematic because losses at different banks are uncorrelated. That's a much simpler situation than insuring regular banks, which have loan loss risks that are strongly correlated. Recession -> unemployment and bankruptcies -> non performing loans at many banks at the same time. You are right that regular banks cannot lose everything in a single event. But that's not true for many other types of insured assets. Fire insurance being one example. As long as there are many insured and the losses are not strongly correlated it's not a problem for insurers.

fauigerzigerk 13y ago

I think a deposit insurance scheme could work very well if all (remaining) big players came together and decided collectively to create one. It would be widely publicised and it would be made very clear that no one should trust a bitcoin bank that is not part of the scheme. Such a scheme would have to include mandatory auditing of finances, QA and security. So what I'm proposing is industry self regulation. That said, I doubt that any deposit insurance scheme could ever be large enough to cover losses of such enormous proportions as we have seen in recent weeks. The hope would be that utterly incompetent adventurers like Mt. Gox would never pass the audit.

jellicle 13y ago
If you think a bit about it, it can't work. A bank can verify, on a daily basis, that they still have "most" of their assets. And a lot of the assets can't be stolen in any real way. Banks owns a house - house can't be stolen. The legal process will get it back. Bank owns treasury bonds - can't be stolen, we know who is "supposed" to own them. Etc. Occasionally there may be an unexpected asset leak (rogue trader or something). But the damage is still limited, it's still small compared to the assets of the bank. Barings lost something like 20% of all their assets - obviously disastrous, but they still had 80% left. And a failure that large is intended to be impossible to occur. A bitcoin "bank", well, you audit it at 11AM and at 11:02AM it can have nothing, 0, zero. No insurer is going to sign on for that sort of risk at any price.
- fauigerzigerk 13y ago
  >No insurer is going to sign on for that sort of risk at any price. One already has: http://www.bbc.com/news/technology-25680016 http://www.bbc.com/news/technology-25680016 But I'm not suggesting getting insurance from insurance companies. I'm suggesting that every bitcoin bank should set aside some small share of its revenues to fund the industry's own deposit insurance scheme. That's how deposit insurance works in many countries. If the losses exceed the fund's resources then it's too bad for the depositors. Insuring bitcoin banks (or rather storage companies) is not actually that problematic because losses at different banks are uncorrelated. That's a much simpler situation than insuring regular banks, which have loan loss risks that are strongly correlated. Recession -> unemployment and bankruptcies -> non performing loans at many banks at the same time. You are right that regular banks cannot lose everything in a single event. But that's not true for many other types of insured assets. Fire insurance being one example. As long as there are many insured and the losses are not strongly correlated it's not a problem for insurers.