Why Geithner’s plan is the taxpayers’ cursePeyton Young,
http://www.ft.com/cms/s/0/3e985de0-1ee7-11de-a748-00144feabdc0,Authorised=true.html">Financial Times
People who outbid others in auctions sometimes pay too much, a phenomenon known as the winner’s curse. Yet the plan outlined last week by Tim Geithner, US Treasury secretary, for pricing the toxic assets clogging up the financial system provides private investors with an unusually strong incentive to overpay: the government is proposing to pick up most of the tab if the assets turn out to be worth much less than was spent on them. Indeed, the more aggressively investors compete in bidding for these assets, the worse off the taxpayers will be. I call this the taxpayers’ curse.
A simple example will illustrate the problem. Suppose that a given bundle of mortgage-backed securities would be worth $20m (€15m, £14m) if you could be sure that all the mortgages will be repaid in full, but they might also turn out to be worthless. No matter how much you pay for them, the US government agrees to absorb any losses beyond approximately 15 per cent, while you get to keep half of any gains. In return, you only have to put up about 7.5 per cent of the purchase price. How much will the assets sell for? That depends on two things: how aggressively others bid and how much uncertainty there is about their ultimate value.
For simplicity, assume the assets could be worth $20m or zero with equal probability. Assume that yours is the winning bid at a price of $10m. Under Mr Geithner’s plan, you put up $750,000 for an equity stake and the government puts up the remaining $9,250,000: a loan for $8,500,000 and $750,000 for an equal share of the equity. There is a 50 per cent chance that you will get your money back in full and make a profit of $5m (in which case the other $5m in profit goes to the Treasury).
Of course, it is equally likely that the assets will turn out to be worthless, but in that case all you lose is your initial payment of $750,000, and the Feds are on the hook for the rest. That works out to an expected profit of $2,125,000 for an investment of $750,000, a return of 283 per cent.
If this seems too good to be true, it is: competition from other bidders will probably drive the bid price much higher. This would be unfortunate, however, because $10m is already the expected value of the asset. For example, a bid price of $14m would still be a bargain, because the investor’s expected profit would be approximately $1m on an initial investment of approximately $1m, which represents a 100 per cent return. Meanwhile, the taxpayers can expect to lose nearly 40 per cent of their money.
This is the singularly perverse feature of the Treasury proposal: the greater the competition among the bidders, the worse off the taxpayers and the more distorted the so-called “market” prices that result. More generally, one can work out the amount of price distortion and the expected returns to the taxpayers as a function of the variance in the realised values of the asset and the expected returns demanded by investors. For example, if there are two equally probable outcomes, one 50 per cent above the mean and the other 50 per cent below the mean, taxpayers can expect to lose money unless private investors make more than 180 per cent in expectation.