British Chancellor of the Exchequer Alistair Darling has proposed a new banking regulation regime under which very large “too big to fail” institutions should be compelled to carry more capital than smaller banks. At first sight, this looks sensible, but on further examination the change may go in the wrong direction, having the perverse effect of making the “too big to fail” problem worse.
The idea that a very large bank is “too big to fail” is a relatively modern one in the United States, created by the mass bank amalgamation movement of the 1980s and 1990s. The first time I remember hearing it was at the time of the Continental Illinois collapse in 1984, when the Federal Deposit Insurance Corporation infused $4.5 billion to rescue the bank. However the precedent was actually set four years earlier, in the 1980 collapse of First Pennsylvania Bank, which had lost money through an incredibly dozy policy of buying fixed rate Treasury bonds and funding the purchase with short term deposits – the sharp rise in interest rates in 1979-80 made it effectively insolvent, even though the accounting of that period did not require it to admit this. First Pennsylvania was not allowed to fail, as had previously been the practice, nor sold to another bank (in those days prior to interstate banking, only Mellon Bank of Pittsburgh would have been large enough to buy it) but was granted a $300 million loan from the FDIC and $1.2 billion in other loans because it was “essential.” Since its “essential” quality consisted only in its size, the “too big to fail” doctrine was born.
Looking back in history, the “too big to fail” doctrine could have had application earlier. In the United States, the Second Bank of the United States should probably have been considered “too big to fail” in 1841 – it had after all been the country’s central bank. On the other hand Jay Cooke & Co. (1873), Knickerbocker Trust (1907) and the Bank of United States (1930) could not reasonably have been considered “too big to fail” in terms of size, although Cooke was certainly central to the US financial system.
In Britain, Pole Thornton, which crashed in December 1825, was relatively small, although its network of correspondent banks was extensive and its failure brought down many other banks. Overend, Gurney, which fell in 1866, was certainly large enough to apply a “too big to fail” doctrine, but had been known to be in trouble for five years, so the smart money had got out. Barings was not considered “too big to fail” either when it was rescued (1890) or when it wasn’t (1995), although on the former occasion it was deemed too good a name to be allowed to collapse. Finally Northern Rock, which was rescued at enormous taxpayer expense in 2007-08 was a gimcrack empire that should never have been allowed to grow so large, and whose collapse should have been welcomed.
In Continental Europe, the most important precedent of all was that of 1931 when Austria’s Creditanstalt, which had more than 50% of Austrian deposits, was rescued by the Rothschilds and the Austrian National Bank. Fears of the inflationary effect of the rescue caused a run on the Austrian currency that spread the panic first to Germany then to Europe as a whole, turning a moderate recession into the Great Depression. In that case the policy decision to consider Creditanstalt too big to fail brought down the entire European economy, suggesting that “too big to fail” designations are not without risk.
Darling’s proposal to make it more difficult for “too big to fail” banks to require bailing out thus addresses a real problem. However it seems unlikely to provide a solution. Increasing the capital base required of large banks will make them uncompetitive with their smaller brethren on simple, low-risk businesses, because their capital costs of, for example, additional lending business will be greater. It will therefore force them into higher-risk businesses, where smaller banks are less able to compete because of their lack of specialist staff.
In particular, such a requirement is likely to drive the largest banks towards businesses whose capital requirements are modest (often because regulations have not caught up with reality) and whose risks are large. That would include securitization, through which they will be able to take assets off their balance sheets altogether and credit default swaps, where they will be able to assume gigantic credit risks while reporting only the modest “mark to market” value of all CDS save those relating to companies near default. Public and private equity business will also become relatively more attractive for such banks, because their capital requirements on equity holdings will not be increased further and their cost of leverage will be extremely competitive.
The Darling proposals would thus turn the largest banks into hedge funds, their losses guaranteed by the taxpayer. Not a move in the right direction. A tax on liabilities above a certain absolute level would work better, since it would encourage the largest banks to divest assets so as to remain below the threshold, but would also tend to force their operations towards riskier and more profitable businesses. The regulation of “too big to fail” banks should restrict them to the less risky portions of the financial services business, leaving the highest-risk businesses to be carried out by smaller entities whose failure would not endanger the banking system.
This is not something that can be left to fester. On Friday it was announced that Goldman Sachs’ profitability and potential bonuses are running ahead of 2007 levels. This cannot be through carrying out low risk business, nor through sophisticated advisory work, the market for which has been distinctly depressed. It can only be through “principal trading” and the like – the activity that since 1990 has turned Goldman Sachs into the world’s largest insider-trading hedge fund, profiting from its access to corporate decision making, its deep knowledge through its trading desk of financial flows on a day-to-day basis and its superb “crony capitalism” network of contacts to carve out superior returns at the expense of the market as a whole.
Goldman Sachs, which currently operates under a banking license, is the most egregious example of the moral hazard that the “too big to fail” doctrine can cause. Its activities distort markets, because it is able to profit as principal from transactions in which its prestige and standing are deemed essential. Doubtless a substantial portion of its returns have come from the field of credit default swaps, in which it was bailed out by taxpayers after the AIG failure to the tune of $13 billion. In this area in particular, its operations are almost entirely to the detriment of the US economy as a whole, since it is able to profit by manipulating markets into creating bankruptcies – the ability to profit from CDS on AIG while at the same time receiving a taxpayer bailout of losses on CDS written by AIG was truly egregious.
Hence any effective “too big to fail” regulation needs to cut Goldman Sachs down to size, before taxpayers’ pocketbooks are hit with yet another gigantic loss. The advisory role of traditional investment banking is essential; it is extremely inefficient to outsource high-level market and securities design expertise to every borrower and investor that needs it. It is also necessary to have an underwriting mechanism for new issues, although the London merchant banks demonstrated that this function could equally well be assumed by investment institutions, with the bank acting as mere arranger and broker. The hedge fund function of a proprietary trading desk is also valuable in moderation, but should be separated from the advisory and underwriting functions, because of the gigantic conflicts of interest involved. None of these operations should be guaranteed by taxpayers.
The solution is thus clear. Institutions that benefit from a “too big to fail” guarantee should be sharply restricted in their operations, becoming modeled on the pre-1986 UK clearing banks. They would be able to advise, underwrite in moderation, lend and take deposits, but their activities beyond the commoditized sectors of financial services would be sharply restricted by regulations specific to their “too big to fail” position. In turn, they would have lower funding costs than their competitors because of their low risks and effective government guarantee, thus being highly competitive in low risk product areas.
Principal trading, credit default swaps (if legal) securitization and other high-risk operations would be assumed by institutions whose size was limited by statute, both in terms of assets and capital. Essentially, most of the high-risk business would be done by hedge funds, whose size and leverage would be restricted. That way, the high risk businesses that had true profit potential would be done by low-regulation institutions, which would in their turn have relatively high funding costs (lending of “too big to fail” banks to such entities would be tightly restricted.) By separating out the guaranteed institutions by business line, rather than by capital requirements or tax, the principal functions of a free market would be preserved.
It’s government regulation, yes. But propping up “too big to fail” institutions with taxpayer money is also government intervention, and without wholesale reform we have lost the possibility of allowing the free market to reign supreme through providing deposit insurance, over-expanding the money supply, regulating and guaranteeing home mortgages and bailing out the banking system.
A root and branch reform of the financial system, including a “Volckerization” of the Fed to prevent it over-expanding credit would be ideal, but is presumably politically impossible. As a second-best solution, if we must have “too big to fail” banks, they must be made safe and boring. Taxpayers deserve no less.
-0-
(The Bear’s Lair is a weekly column that is intended to appear each Monday, an appropriately gloomy day of the week. Its rationale is that the proportion of “sell” recommendations put out by Wall Street houses remains far below that of “buy” recommendations. Accordingly, investors have an excess of positive information and very little negative information. The column thus takes the ursine view of life and the market, in the hope that it may be usefully different from what investors see elsewhere.)