There's a lot of debate on whether or not we will be able to remove the new moral hazard that has been introduced as a result of the bailouts. Can financial engineering help? From Nick Schulz, an AEI Shadow Financial Regulatory Committee panel came up with two financial instruments. Let's look at each of them.
Failure Prediction Contract
In the first, Treasury:
create a prediction market for bailouts of financial institutions... The "Failure Prediction Contract" (FPC) would be issued for individual large financial institutions, perhaps in proportion to their capital, and sold by the Treasury. It would pay out a pre-specified amount to investors in the event a bank fails, is bailed out or taken over by the government or the regulator, or receives any form of emergency credit support. The creation of such contracts would create incentives for financial regulators to enforce capital requirements in the first place.
We have several markets that predict a failure of a company already, the most notable being the credit default swap market. Credit default swaps (CDS) are insurance for asset losses. So the buyer of the CDS makes regular payments to the seller and gets a big payoff if the loan, or security goes into default. A higher chance of default raises the price of the CDS. The graph of Lehman's CDS prices is below.
I imagine the FPC market price would have looked very similar. Otherwise you could arbitrage the difference. It would be easy: If FPC was predicting a lower default probability, you'd buy the FPC and short a CDS contract. It wouldn't be an exact match, and I'd have to see specifics on the two contracts to get a sense of what else would have to be held, but in general the FPC price couldn't wander too far away from the CDS contract.
Now let's assume that the CDS spread, for now, is a coherent measure of credit risk, against insider opinion that it is rather a bet against the spread rising or falling. Again, looking at that Lehman CDS contract, it's not necessarily clear that it would give regulators a lot of heads-up that a crisis was coming - we might want regulators to be most on the ball at the height of the business cycle, when the CDS/FPC contract would be predicting the least chance of default.
The bigger question is making Treasury pay out this CDS-like contract. I'm not sure if it would change any regulator's behaviors - it was clear that Fannie/Freddie would have to be backstopped by the government, and yet there was no due-diligence there. At the end of the day, that is still passed on to taxpayers. You could make the government regulators lose their life savings if there is a bailout, but they aren't getting any upside on the finance sector's profits - you'd immediately lose any talented people you could hire by forcing them into a labor contract option that is entirely downside.
The next one?
is to require large financial institutions to issue a bond that does not have to be repaid in the event a bank fails, is bailed out or taken over by the government or the regulator, or receives any form of emergency credit support. This security is analogous to a catastrophe bond, except that its trigger event is a bank failure or government intervention. Such a security would have the added benefit of absorbing losses in a time of crisis.
I'm not sure how this is different than just requiring banks to hold more capital/increasing the leverage requirements. Again, this is the equivalent of a bank writing a CDS contract on itself. If it isn't fully funded, nobody would want it (it would be like writing a put option on yourself). If it is, there's just an extra layer of capital that is sitting around in reserves inside the bank. However in the case of distress, it isn't deployed to help handle losses but instead goes off to some lucky counterparty. So it may even be subpar in terms of stabilizing capital reserves.
In terms of helping with moral hazard, it's the equivalent of a gambler saying "here take this chip, if I go bust you can keep it." I'm not sure if that changes very much.
We have all the tools we are going to get for predicting catastrophes. Credit risk is a fundamental type of risk, and it is as much of a 'prediction market' as we are going to get for credit risk (and if you ask insiders, it's not much of one). In terms of penalizing institutions extra if they go bankrupt, remember that institutions are legal constructs, and we need to penalize individuals - make it so that their downside is equal to their upside. I am doubtful we can create that from the outside, and simpler and blunter regulations - capital reservings, silo-ing of functionality -- might be the smart move for future regulation.