Thursday, October 2, 2008

Bailout Can Work and At No Cost to Taxpayers

While I do think there are better alternatives than the current plan, something is needed. HOWEVER, I absolutely think this plan can help unfreeze the credit markets and recapitalize the financial institutions at a small or no cost to taxpayers. How? Let’s first rewind to see how we got into this mess using a very simple bank I'll call Bank A.


In mid 2007, Bank A was doing well with $100 in assets and $60 in liabilities, thus had equity of $40. Unfortunately all the assets bank A owned were subprime mortgages, which now have a market value of $50. Which means Bank A currently has negative equity of $10 and is insolvent. As such, no other bank will provide lending to Bank A as they will fold with no government intervention. This would be fine (and a natural part of capitalism), except Bank A is intertwined with Banks B, C, D, and E and if Bank A goes bankrupt, all these banks go bankrupt. Thus, credit markets are frozen (i.e. NOBODY can get a loan).

Under the bailout the Treasury buys the assets currently priced at $50 for $90 (I will explain later why this won’t cost taxpayers money), Bank A uses the proceeds to make $90 of loans to small businesses / individuals that need credit (for things such as payroll), Bank A’s equity rebounds to $30, Bank A is now solvent (and other Banks are once again willing to lend them money), and the credit markets are now functional (i.e. if you have decent credit, now you can get a loan - forgot the "old" days of easy money).


Won’t this cost taxpayers money you ask? After all, they did buy assets worth $50 from the bank for $90? Nope... fortunately for the U.S., the Treasury can borrow money on the cheap.

Value of Assets to Bank A
Assuming Bank A's subprime assets pay $4 coupons per year over 8 years and returns the $100 principal in year 8 (I understand this is not how subprime deals work, but lets keep this simple) and requires a return on capital of 15% (bank capital ain’t cheap these days), the assets are worth the $50 shown above (the NPV calculation is shown in the chart below).

Value of Assets to Treasury
However, the Treasury currently has the ability to borrow at less than 4% / year for 10 years (the current yield on the ten year bond is ~3.7%). Assuming the same cash flows as above, but discounting them at 4% per year instead of 15%, the assets are worth $100 (again the NPV is shown below). Lets assume the Treasury pays Bank A $90 for these subprime assets "worth" $100. In this case, it doesn't only not cost taxpayers a cent, but they “receive” $10.

The bailout will not solve all the economic problems we are currently facing. In fact, not even close. We still have a massive amount of leverage in the system that needs to be unwound. However, if this bailout is done right, it should help unfreeze credit markets (which are currently non-functioning) at little or no cost to taxpayers.


  1. What if you don't get the principle back because the mortgage payments stop?

    What if the collateral will never ever see those bubble prices again?

    What if the collateral has to be sold into a market that is both oversupplied and where valuations are depressed?

    What if the cost of money increases for the US Treasury as is likely to happen as they flood the market with massive debt offerings...

  2. 1) Principal - My example was extremely simplified (I understand they don't have cash flows in the way I presented). That being said, conservative expected returns on non-Agency mortgages are well above 10% at current valuations (Bill knows a lot more than I do about this stuff: My point was just that the Treasury can finance itself at a much cheaper rate than banks can. Either way, they are worth more to the Treasury that has a much lower cost of capital.

    2/3) Collateral - collateral will not EVER see those bubble prices again and may be dumped on the market. Again... using conservative assumptions (much higher defaults, slow paydowns, etc...), these should return 10%+. They are priced to the point that expectations are for practically EVERYONE to default.

    4) Treasuries are EXTREMELY overpriced considering what will likely happen. That being said, rates are currently below 4% for a 10 year loan. If this increases to 5%, they still can offer prices where expected return = 5% and if met, this will not cost taxpayers a cent.

  3. As leverage increases so does borrowing costs. Adding 10% to the national debt could easily add 1% to the nation's TOTAL borrowing rate on $10 trillion, adding $100B in ANNUAL debt interest payments (could add 5% or 10%!). This isn't even considering the gov't is propping up home prices, and increasing borrowing costs, both of which are negative for taxpayers (but good for bankers). Nouriel Roubini saw all this coming, and him and 400 other economists say this won't work. "Follow the money", not the spin. Money is going from taxpayers to a few bankers. Remember that historically bankers have bought and sold nations (France, House of Morgan, etc). We should be buying equity in banks, not overpaying them for bad assets. Not good for America or Americans.

  4. I agree with you postpeak, not ideal, but better than nothing.

    If this bailout puts the U.S. at less credit risk (without a bailout, the U.S. is very shaky) then this offsets the increase in financing rate that is due to increased supply.

  5. Jake - thanks very much for posting this. One of the few bloggers and other pundits trying to take a balanced view. Unfortunately grasp on contagion and it's consequences appears to be widely limited.

    One might wonder, as you sorta do, what happens to our credit costs if our ratings drop thru the floor. Or, more likely, what happens when foreign funds start going elsewhere if they were able.

    More importantly this is an investment - running alternative scenarios on different GDP growth paths what the rescue package is trying to do is avoide $17-25T in lost GDP growth over the next ten years.
    Marketing Elephant Pills: Struggling to Explain the Package