Simple Facts

This morning I read two articles side-by-side in yesterday's Financial Times.

One was by James Carville, who argued that Obama has not lost his abilities as a "great communicator," but it was the fault of the topics he had to speak of as President.

It is not that Mr. Obama is not communicating as well; it is that what he is communicating is too complex to reduce to simple words, especially when in the last 40 years, the length of a TV soundbite has dropped by 40 seconds.

In short, we've become dumbed down, so just trust him.

However, the other article presented a succinct, yet rich explanation of the risk to the taxpayers inherent in Geithners' plan for the government to back toxic assets. The noted economist, Jeffrey Sachs, writes:

The plan’s essence is to use government off-budget money to overpay for banks’ toxic assets, perhaps by a factor of two or more. This is done by creating a one-way bet for private-sector bidders for the toxic assets, then cynically calling it “private sector price discovery”. Consider a simple example: a toxic asset with face value of $1m pays off fully with probability of 20 per cent and pays off $200,000 with probability of 80 per cent. A risk-neutral investor would pay $360,000 for this asset.

Along comes the government and says it will finance 90 per cent of the investor’s purchase and, moreover, do so as a non-recourse loan. Non-recourse means the government’s loan is backed only by the collateral value of the toxic asset itself. If the pay-out is low, the loan is defaulted and the government ends up with the low pay-out rather than full repayment of the loan.

Now the investor is prepared to bid $714,000 (with rounding) for the same asset. The investor uses $71,000 of his/her own money and $643,000 of the government loan. If the asset pays off in full, the investor repays the loan, with a profit of $357,000. This happens 20 per cent of the time, so brings an expected profit of $71,000. The other 80 per cent of the time the investor defaults on the loan, and the government ends up with $200,000. The investor just breaks even by bidding $714,000, as we would expect in a competitive auction.

Of course, the investor has systematically overpaid by $354,000 (the bid price of $714,000 minus the market value of $360,000), reflecting the investor’s right to default on the loan in the event of a poor pay-out of the toxic asset. The overpayment equals the expected loss of the government loan. After all, 80 per cent of the time (in this example) the government loses $443,000 (the $643,000 loan minus the $200,000 repayment). The expected loss is 80 per cent of $443,000, equal to $354,000.

The idea of “private sector price discovery” is therefore flim-flam. There would be price discovery if the government’s loan had to be repaid whether or not the asset paid off in full. In that case, the investor would bid $360,000. But under the Geithner-Summers plan the loan is precisely designed to be a one-way bet, for the purpose of overpricing the toxic asset in order to bail out the bank’s shareholders at hidden cost to the taxpayers.

I think I can find better odds at the racetrack. The government is not assuming all of the risk - we are, through the government.