One can disagree with the particulars of this comment by George Soros, but his main point is sound. The financier argues that the Treasury Department’s $700 billion Troubled Asset Repurchase Facility should be used to recapitalize banks. This blog and most economists have argued that restoring depleted bank equity is the top priority for shoring up the financial system, and the original TARP program was a costly and inefficient means to that end.
The provision that will likely strike some readers as odd is Soros’ call for banks to have their assets written down, then get new equity infusions to bring the ratio of equity to total assets to 8%, but then permit banks to operate at lower (unspecified) equity to total balance sheet level. The reasoning behind that is if equity is brought to a prudent level based on accurate valuations, banks will not be able to make new loans, and those balance sheet constraints would hamper growth. It is typical for bank regulators to let banks run at lower equity levels in downturns (although they usually do this in ways that are not obvious to the public at large). This is a realistic approach, but one would also hope that regulators would apply pro-cyclical capital rules in good times (ie, requiring banks to hold higher equity levels when the economy is robust. There have been proposals advanced for how that would work operationally that appear sound).
From the Financial Times:
Now that Hank Paulson has recognised that the troubled asset relief programme is best used to recapitalise the banking system, it is important to spell out exactly how it should be done. Since it was not part of the Treasury secretary’s original approach, there is a real danger that the scheme will not be properly structured and will not achieve its objective. With financial markets on the brink of meltdown it is vital to make the prospects of a successful recapitalisation clearly visible.
This is how Tarp ought to work. The Treasury secretary should begin by asking the banking supervisors to produce an estimate for each bank, how much additional capital they would need to meet the statutory requirement of 8 per cent. The supervisors are familiar with the banks and are aggressively examining and gathering information. They would be able to come up with an estimate in short order provided they are given clear instructions on what assumptions to use. The estimates would be reasonably reliable for the smaller, simpler institutions, but the likes of Citibank and Goldman Sachs would require some guesswork.
Managements of solvent banks would then have the option of raising additional capital themselves or turning to Tarp, which would state the terms on which it is willing to underwrite a new issue of convertible preferred shares. (Convertibles are better than warrants because the banks should not need additional capital infusions later.) The preferred shares would carry a low coupon, say 5 per cent, so as not to impair banks’ profitability. The new issues would dilute existing shareholders but they would be given preferential rights to subscribe on the same terms as Tarp and if they were willing and able to put up additional capital they would not be diluted. The rights would be transferable and if the terms were set right, other investors would take them up.
Using this approach, $700bn should be more than sufficient to recapitalise the entire banking system and funds would be available to buy and hold to maturity mortgage related securities. Since insolvent banks would not be eligible for recapitalisation, the Federal Deposit Insurance Corporation would certainly require topping up.
Concurrent to the recapitalisation scheme the authorities would lower minimum capital requirements so that banks would compete for new business. The Fed would also guarantee interbank borrowing by banks eligible for recapitalisation. This would reactivate the interbank market and return the spread of Libor over Fed funds to normal and reduce the abnormally high interest rates on business and mortgage loans linked to Libor.
The success of the bank recapitalisation programme could be undermined by a downward overshoot in housing prices. A separate set of measures is needed to keep foreclosures to a minimum and to fundamentally restructure the deeply flawed US system of mortgage finance. Taken together the two sets of measures would not prevent a recession – too much damage has been done to the financial system and the general public has been traumatised by the events of the past few days – but they would reduce its duration and severity. Once the economy returns to normal, the minimum capital requirements of banks would be raised again.
The international financial system also needs repairing but there are grounds for optimism. Europe has realised that it needs to complement the euro with a government safety net for interbank credit. And the International Monetary Fund is finding a new mission in protecting countries at the periphery from the storm at the centre.
The recapitalisation scheme outlined here would suffer from none of the difficulties of reverse auctions for hard-to-price securities. It would help restart the economy and likely produce returns for taxpayers comparable to my fund’s. But time is of the essence. The authorities have lost control of the situation because they were constantly lagging behind events. By the time they acted, measures that could have stabilised markets were ineffective. Only by promptly announcing a comprehensive set of measures and executing them vigorously can the situation be brought under control.
Actions speak louder than words. Specifically, Morgan Stanley urgently needs rescue. The Treasury should offer to match Mitsubishi’s investment with preferred shares whose conversion price is higher than Mitsubishi’s purchase price. This will save the Mitsubishi deal and buy time for successfully implementing the recapitalisation and mortgage reform programmes.