Wednesday, March 4, 2009

A strategy of contingent nationalisation

By Jeffrey Sachs One of the vexing problems of the US banking crisis is the difficulty of valuing the toxic assets on the banks’ balance sheets. The government is proposing to remove these assets in return for taxpayer equity in the banks, but at what terms of exchange? It seems that if the government pays too much it bails out the banks, while if it pays too little it de-capitalises them. There is way, however, to be both fair and efficient, by settling the taxpayers’ ownership at a later date, after the toxic assets have been monetised. Consider a bank balance sheet with 100 in assets at face value, 90 in liabilities, and 10 in shareholder equity. For simplicity, suppose that the 90 in liabilities are in government-insured deposits. The assets are worth less than face value. Suppose that 80 of the assets are actually worth face value, while 20 are at a deep discount. If the true value of the 20 is 15, the true shareholder value is 5, while if the true value is only 5, the bank is insolvent, with zero true shareholder value and a government net liability of 5 to honour the deposit guarantee (once the bad assets are realised). The problem is that the market value of the assets is not known now, because the credit squeeze has temporarily eliminated the liquidity in the markets of the toxic assets. There is an added problem. If the government pays fair value for the 20 of toxic assets, it willy-nilly forces a severe write down on the balance sheets even if it pays fair price. This in turn can lead to a further squeeze of lending. If the toxic assets are indeed worth 15, and these are swapped for 15 of government bonds, the recognised bank capital falls to 5, and capital adequacy standards would induce a further retrenchment of loans. The bank can be recapitalised at fair value to taxpayers and without inducing a squeeze on bank capital and lending. The government can swap 20 in government bonds for the 20 in toxic assets plus contingent warrants on bank capital, the value of which depends on the eventual sale price of the toxic assets. The government would then dispose of the 20 in toxic assets at a market price over the course of the next year or two and exercise its contingent warrants at that time. Warrants are securities that entitle the holder to buy shares of common stock at a specific price. Contingent warrants would entitle the holder, in this case the US government, to buy shares of common stock at a price contingent on the eventual sale price of the toxic assets. The cleaned bank now has true market capital of 10, equal to 100 in good assets minus 90 in deposit liabilities. The 10 in equity would then be shared between the original shareholders and the taxpayers, with the division to be determined by the eventual sale price of the toxic assets. If the 20 in toxic assets end up selling at their face value, the taxpayers end up with zero ownership of the bank. If the toxic assets end up selling for less than 10, the bank ends up wholly owned by the taxpayers, because the bank is in fact insolvent at the time of the swap. If the toxic assets end up selling for over 10 and less than 20, the taxpayers get an equity stake equal to 20 minus the eventual sale price. A sale price of 12 (of the 20 face value) leaves the taxpayers with 8 in stock ownership (20 minus 12) and the original shareholders with the remaining 2 (10 minus 8). The taxpayers are thereby fully compensated as long as the bank is solvent, and is left to cover the deposit liability in the event that the bank is insolvent. The shareholders also get the true value of their claims. In this process, there are no taxpayer bailouts, and there is also no squeeze on bank capital resulting from the exchange of toxic assets at less than face value. In practice, the conversion of taxpayer warrants into bank equity would proceed step by step. Suppose that the taxpayers already own 2 and the shareholders own 8 as a result of partial liquidation of the toxic assets. Now, at a later date, additional sales of the toxic assets cause a further realized loss of 1. The bank would then issue further equity at that date equal to 1 (at the contemporaneous market value), further diluting the existing shareholder claims by 1 and raising the taxpayer stake to 3. During the period of liquidating the toxic assets, the government would exercise a kind of receivership over the banks, preventing the stripping of remaining assets through bonuses, balance sheet transactions, or “Hail-Mary” lending (in which shareholders of zombie banks gamble recklessly because they have nothing to lose and possibly something to gain). In practice, this could be exercised in the form of a “golden share” which gives the government the right of refusal over major bank decisions, including executive compensation. Once the warrants are exercised, the government would sell its ownership stake to private investors. The TARP has been stymied to this point over the valuation conundrum, stuck between paying an unduly high price for the toxic assets, and thereby bailing out the shareholders, and paying a low price, and thereby expropriating them while inducing a further credit squeeze. A “contingent warrant,” as recommended here, can combine bank recapitalisation with a fair value divided between the taxpayers and original bank shareholders.

No comments: