Fed Follies: TARP - How it should work

Dugan Says Pricing Assets ‘Key’ Issue in Bank Rescue

2009-02-02 16:55 - By Margaret Chadbourn and Alison Vekshin

Robert Rubin Says ‘Mark-to-Market’ Accounting has Done ‘Damage’

2009-01-28 18:14 - By Josh Fineman and Ian Katz

Price and value are two different things -- three if you include mark-to-market (MTM). (See below) If regulators were able to validate fair or economic values, then maybe prices wouldn't be so depressed and they wouldn't need to buy so much. Price is relevant for leveraged institutions like banks & brokers but, ultimately, if value -- fair value or economic value -- is perceived then maybe they can fund themselves. (FASB has corrupted the term "fair value".) In buying assets the Fed 'gives up some of their height', as Teddy Atlas might say (ESPN-2's boxing analyst), and as I've said before.

Here's the sketch of an idea how the Fed could be more parsimonious with their capital -- our tax dollars! First, regulators validate fundamental or fair values (not the corrupted FASB definition) by modeling the expected cashflows. Then they validate market value (FASB's fair values, levels 1 - 3) and translate or 'calibrate' that value/price into the implied cashflows. The Fed then puts a guaranty 'collar' on those cashflows: the Fed pays the shortfall if actual cashflows are less than the market-implied, and the Fed receives the excess if cashflows are greater than the fundamental forecast. As with conventional bond insurance, the bonds receive a new CUSIP and trade as one with the Fed's wrap. Perhaps it could be a derivative that could trade separately (although forever tied to the original bond's performance). Obviously, this is a massive undertaking given the number of banks, securities, and their complexities. But things like this have been done before and so can be done again. Sometimes there is no shortcut.

Example of price v. value v. MTM: The negative basis trade -- if I bought an investment grade bond at T+200 and matching CDS protection at T+75 then I've locked-in a 125 bps p.a. return. When spreads blew-out, let's say doubled to T+400 & T+150, respectively, then I've a larger MTM loss on my bond than gain on my CDS so I report a huge MTM loss. But, assuming my CDS ctpy is now the Federal Reserve (d/b/a BofA or JPM) and not Lehman, I still have a locked-in 125 bps pa return which has gone up, not down, in value because the discount rate is now lower! (ceteris paribus -- legal risk, etc.)

JRB

2/2/09