Fortunately, the Treasury and Fed are looking for a “comprehensive approach to address the illiquid assets on bank balance sheets.” Their idea is to swap up to $700 billion in Treasuries for the “toxic” assets, putting a floor on bank losses and leaving the government to hold the risky assets until conditions improve. The big question is the swap rate; i.e., how to price these thousands upon thousands of illiquid securities so that both taxpayers and bank shareowners are fairly treated.
Treasury Secretary Paulson wants to run auctions to determine prices, but this will take time to set up and may be impractical given the highly complex nature of many of the securities involved. Furthermore, those holding the worst securities will be the most eager to sell.
Our answer is to rely on the same pricing mechanism, but not the same prices, used before things hit the fan. In the good old days, mortgage derivatives were priced by reference to the cost of buying default insurance on five tranches of a bundle of 20 standard subprime mortgages. The top tranche (AAA) provided the highest probability of full repayment. The lowest tranche (BBB) had the lowest probability.In between were another 3 tranches. The market for these insurance contacts is called the ABX.
Avant le deluge, you could use ABX prices to figure out what a given tranche of a standard mortgage-backed security was worth. It would imply be the value of a safe bond — a Treasury with the same coupon — minus the ABX-determined price for default insurance for that tranche. The reason is that once you had purchased the insurance, youare guaranteed a safe income stream; i.e., you were guaranteed the equivalent of a Treasury bond. More exotic mortgage-backed securities, with now scary initials, like CDO2s, could also be priced using the ABX.
So here’s our specific idea. Rather than ask Hank Paulson to determine the price of each and every toxic asset, let’s have him simply set prices for the ABX insurance policies (or credit default swaps, as they are called). Right now these insurance policies are selling for crazy prices because nobody can insure against systemic risk. Nobody, that is, except the government. The government is in a unique position to insure against system-wide risk because its own decisions determine, to a very large degree, the extent of this risk.
Were the government to start selling the ABX insurance policies at reasonable prices, our Cinderella mortgage-derivatives market would suddenly wake up and start pricing every mortgage-related security in sight based on these ABX prices. If Hank does this, the market will do essentially all the pricing; Hank will have only a handful of prices to set, not thousands.
But how does Hank set those prices? What’s fair? What’s fair are insurance policy prices that assumes no system collapse, a modest additional decline in house prices, a mild recession, and modest additional increases in default rates. Yes, these are optimistic assumptions, but that’s the economic outcome the government is arranging and it needs to signal its resolve.
Once the five prices are set, Hank can keep the $700 billion in his pocket. What will happen to the banks? With reasonable ABX insurance prices, their toxic assets will take on reasonable values. This will restore their balance sheets and allow them to keep operating. Yes, this will help bank shareholders, but they will still end up far worse off than at the beginning of this crisis. Taxpayers will keep their $700 billion pistol dry for another day, including the potential need to cover possible losses on these insurance polices.
The Treasury will be receiving premium payments as it sells the ABX policies — premiums it could receive in the form of equity claims, rather than cash, which would further help capitalize the banks.
Playing the ABX market-maker is step one for the government. Step two is reorganizing banks whose capital is still too low even when its “toxic” assets are revalued. This means helping such troubled banks finding a marriage partner. Step three is reregulating the entire financial sector. This includes establishing a Federal Financial Authority that stamps a seal of approval on consumer financial products that it deems to be safe, that rates individual securities, and that audits the books and rates the performance of each and every one of our nation’s major companies.
The final step, and the most important, is to require financial institutions to report on line and in fine detail everything they know about the assets they hold. The principle here is simple enough even for Wall Street “geniuses” to understand. If you want to sell the public a product, including your stock, you need to explain what it is.
Laurence J. Kotlikoff is Professor of Economics at Boston University
and Perry Mehrling is Professor of Economics, Barnard College.
15 Responses to “The Right Financial Fix by Laurence J. Kotlikoff and Perry Mehrling”
Simply elegant … I do not have expertise in this area, so I am curious as to the validity of ratings of these tranches in your proposal using the ABX.Is AAA really AAA ?
The underlying MBS for the AAA tranche of ABX are all AAA at the moment the index is created. These ratings can and do change over time, also some rules for replacing one MBS with another. Find full documentation and details of the ABX index on http://www.markit.com.
Since the ABX index references subprime specifically and at this point even S&P considers 25% default rates on some vintages of subprime reasonable (http://ftalphaville.ft.com/blog/2008/09/18/16080/a-rising-tide-of-writedowns/). Of course, expected default rates are far more likely to worsen than improve. Gary Gorton (http://www.kc.frb.org/publicat/sympos/2008/Gorton.08.25.08.pdf), an expert in these matters, pretty much says that these mortgages were designed to default. How is anybody supposed to help borrowers who were never expected to be able to repay their loans in the first place?We’re very likely to reach 50% default rates on the worst vintages of sub-prime, and your plan is designed to ensure that the taxpayer ends up transferring huge sums of money to wealthiest people on Wall St.
That first sentence should read:Since the ABX index references subprime specifically and at this point even S&P considers 25% default rates on some vintages of subprime reasonable, this insurance policy is sure to cost US taxpayers a fortune.
Actually the best advice for you is to go read Daniel Gros: http://www.rgemonitor.com/globalmacro-monitor/253756/no_recourse_and_put_options_estimating_the_fair_value_of_us_mortgage_assets
Not quite. The BBB tranche is quoted at 5. That means (loosely) that insurance on the BBB tranche would cost the banks 95, which is a transfer from the banks to the government, not the other way around. How would they pay. One way is to pay with preferred shares which would then be the reserve asset held by the government against its insurance policies.
Thanks for the heads-up. It is a pretty interesting article, pointing out that US mortgages are no-recourse, so the borrower can just walk away. European mortgages are not like that, so it is possible that European insurance companies systematically erred in pricing, but that is beside the point we want to make. The point that Gros makes is that the mortgages probably have very little value. That means that it should cost a lot to buy insurance that promises to make all payments–the premium had better cover all the payments that we already know are not going to be made, plus the ones that are likely not to be made. As I wrote above, if the fair price of the asset is 5, then the fair price of insurance is 95.
When I read this paragraph:”But how does Hank set those prices? What’s fair? What’s fair are insurance policy prices that assumes no system collapse, a modest additional decline in house prices, a mild recession, and modest additional increases in default rates. Yes, these are optimistic assumptions, but that’s the economic outcome the government is arranging and it needs to signal its resolve. … With reasonable ABX insurance prices, …”I interpreted it to mean that in your judgement market prices on the ABX are incorrect and that Paulson’s job is to reduce the cost of insurance to the banks. I’m all for the preferred shares idea.Also, did you think about the fact that you will be insuring synthetic assets that never actually contributed to any real economic activity and thus you are proposing that the federal government insure the gambling activity of financial institutions?
Your interpretation of the paragraph is fair. My point is that the virtues of the pricing mechanism we propose, namely working through the index swap market, are independent of the price we choose. We could use liquidation prices, and I think there are some arguments in favor (see http://www.rgemonitor.com/financemarkets-monitor/253753/the_right_financial_fix). Even if we choose liquidation prices, it may be that simply by putting a floor under the market, it will find its own level somewhat higher. We could also choose somewhat higher, and that would help the banks out by itself. We could also choose a different pricing policy for each tranche. Myself, I favor pricing the AAA tranche nearer to “normal” and the BBB tranche nearer to “distress”, but we can argue about that separate from the pricing mechanism itself.We propose writing default contracts on an index. What people use that insurance for is their own concern. Unlike Paulson, we will not be buying CDOs, or CDO2, or CPDO, or synthetic CDO, or anything like that.
How is anybody supposed to help borrowerswho were never expected to be able to repay their loansin the first place? We’re very likely to reach 50% defaultrates on the worst vintages of sub-prime, and your plan isdesigned to ensure that the taxpayer ends up transferring huge sumsof money to wealthiest people on Wall St.
IMHO, the clarity of and substance in your last 2-sentences …is exactly what the general public does not know but needs to.
What about the adverse selection problem: You’re sure to sell plenty of underpriced insurance and no overpriced insurance. Don’t forget that any bank can create an infinite quantity of synthetic assets.
We’re clearly in a situation of multiple equilibria. The current pricing of the ABX insurance policies seems to place a fair amount of weight on the U.S. economy ending up in a really bad equilibrium. But if Uncle Sam can take actions that coordinate private behavior to expect and, thus, achieve a good equilibrium, Uncle Sam can, in effect, set the market prices for these policies. I.e., by setting ABX insurance prices below their current cost, Uncle Sam can, we argue, move the economy to a point where lower prices for insurance are justified and equal market prices.
That’s where we differ: I look at the ABX and see an index that reflects the subprime mortgage fairly accurately given what we know about the market.
And I should add that I am not assuming economic conditions any worse than a longer than normal recession.
Where we set the prices is clearly important, but I don’t want to lose sight of the importance simply of setting the prices somewhere, in order to stabilize things. If it makes people happier to have an auction, we could have an auction to establish the initial prices, and then make a market at a spread around those prices. The auction could be in the underlying MBS if you want, or in index CDS themselves.