Friday, August 03, 2007

When will we find a financial institution small enough and systemically insignificant enough to fail?

Why bail out IKB? A smallish German specialised lender to small and medium-sized businesses, IKB, has been bailed out. Apparently it considered lending to small businesses not sufficiently rewarding, because directly or indirectly it ended up with a serious exposure to the US subprime residential mortgage market. The rescue was a Pan-German affair. It involved the good offices of the regulator (Bafin); a € 3.5bn rescue fund for IKB Deutsche Industriebank, put together by a cross-section of German banks ranging from the largest (private) bank, Deutsche Bank to the publicly owned saving banks – Sparkassen; and a further €8.1 liquidity line from IKB to Rhineland Funding, guaranteed by IKW’s main shareholder, KfW, a state-owned German development bank, that holds a 38% stake in IKW.

Why was IKB bailed out? Bankruptcy and related forms of corporate restructuring are an integral and essential part of the capitalist system; it gives the market system its ultimate Darwinian edge. Without bankruptcy there is no hard budget constraint. Without hard budget constraints a market economy cannot function. This is true for a ball bearings company in Wuppertal. Why is it not true for IKB?

The argument that banks are different because a bank failure may involve material systemic externalities that can cause a system-wide crisis and significant economic dislocation is often alluded to but not often challenged. In fact, the ‘too systemically important to fail’ argument has been extended from (deposit-taking) banks to cover a whole range of financial institutions whose failure would have no direct or intrinsic systemic significance whatsoever, but that acquire indirect or derived significance through their relationships with other financial institutions that are generally considered to be inherently systemically important. Hedge funds, (large) private equity funds and investment banks are examples of institutions that are sometimes argued to have derived or indirect systemic significance because other institutions that are intrinsically or directly systemically important are exposed to them.

Why bail out LTCM? The rescue of Long-Term Capital Management organised in September 1998 by the New York Fed, was justified on the grounds that bankruptcy of LTCM and a quick liquidation of its portfolio would have created serious problems for some of its creditors (which included leading commercial banks and investment banks) and could have meant fire-sale prices for some of the assets to be sold as part of LTCM’s liquidation. Before addressing these points, it is useful to underline the systemic insignificance of LTCM. LTCM was a hedge fund, a betting shop, taking highly leveraged bets on interest rate convergence plays for the prospective Eurozone members. When the Russian crisis intervened in August 1998, the spreads that the LTCM geniuses (including Noble Laureates Merton and Scholes) had bet would narrow, in anticipation of the January 1, 1999 start of EMU, instead widened. Then key financial markets and institutions, including creditors and counterparties of LTCM, ceased to act like the passive price-taking drudges of text-book fully rational optimising finance (the kind of finance theory created and taught by Scholes and Merton) and began to act strategically and/or according to the teachings of behavioural finance.

LTCM was at serious risk of going under and should have been allowed to go under. Its creditors, including commercial banks, might have taken a hit, although it is not at all clear that an orderly bankruptcy process would necessarily have involved the excessively hasty liquidations of LTCM assets that those favouring a rescue invoked. Since no commercial bank was exposed to LTCM an extent that would have undermined its financial viability even if its entire LTCM exposure were to have been written off, there was no reason for the Fed to be involved in the rescue.

Rescues often involve conflicts of interest, even when there is no public money involved. In the case of LTCM, it would, for instance, clearly have been improper for a Chairman/CEO of a financial institution involved in the rescue, to put his institution’s money at risk in the bail out if he had a personal exposure to LTCM. Yet such conflict was present and was implicitly condoned by the Fed, through its involvement in the rescue.

The reasoning that resulted in some of the original LTCM shareholders, including the two boffins, being allowed to keep a small amount of their equity, would be hilarious if it were not so outrageous. Only the unbounded (from above) IQs of the original team were capable of unwinding the infinitely complex and delicate structures that LTCM had on its books. Sure. Half a dozen third-year PhD students in Finance from any top-twenty Economics Department or Business School could have done the job at least as well and for a lot less money. Why give the job to people that had just demonstrated their lack of understanding of both the substance of macro risk and of micro-market structure – the two areas that conventional finance theory has indeed nothing to say about?

This relates to a common argument for preventing bankruptcy – that a corporate entity may be more than the sum of its separately realisable parts, that is, that it has ‘going concern’ value that gets destroyed when the existing legal/organisational entity gets disbanded. I recognise that there may be such forms of ‘organisational social capital’, but I don’t think they played any role in LTCM. Its assets were supposed to be the geniuses that founded it. When the geniuses turned out to be dunderheads instead, the value of the enterprise fell to zero.

Regulatory capture by the financial sector? Why did the Fed (through the New York Fed) offer its good offices by arranging the rescue, even if no public money was put at risk? I believe there are two reasons – manifestations of the two classical regulatory failures: regulatory capture and excessive prudence motivated by a desire not to have any shipwreck on one’s watch.

We have known since the classic work by Stigler that regulators tend to be captured by the industry they are mandated to regulate. The financial sector and its regulators are no exception. The Fed, like every central bank, and especially like every central bank that also has a supervisory and regulatory function, has to be close to the key financial markets and the key players in it, if they are to do their job properly. The central bank operates on a daily basis in the same domestic and financial markets as the financial institutions it regulates and supervises. It also encounters on a daily basis representatives of institutions which, although not themselves supervised or regulated by the central bank, are the counterparties of the central bank directly or of the institutions that are supervised and regulated by the central bank. It is therefore all but unavoidable that the central bank becomes too close to the financial institutions and markets to remain objective and impartial. Even when this does not lead to technical or legal improprieties, it does mean that too much weight is given to the often self-serving arguments of the markets and private financial institutions. This rather incestuous closeness of most central banks, including the Fed, and the private financial institutions may well contribute to the excessive readiness of central banks to bail out failing private financial institutions. It could even extend to the readiness with which former Fed Chairman Greenspan in particular flooded the markets with liquidity to mitigate any actual or threatening collapse of the stock market. The ‘Greenspan put’ could be the expression of ‘psychological regulatory capture’ of the Fed in its monetary policy capacity (not in its regulatory capacity) by the American stock market investing public at large.

The second argument – excessive aversion to a major financial institution failure on one’s watch – is based on the asymmetry of the regulator’s payoff function. If, thanks to central bank intervention, an institution is bailed out that really ought not to have been bailed out, there is some loss of reputation and perhaps some financial penalty if, unlike what happened with LTCM, the central bank puts in some of its own or the government’s money. If a major commercial bank which is regulated by the central bank were to fail and be liquidated, the head of the Chairman of the Federal Reserve Board would roll.

That was then – this is now: no private financial institution is systemically significant The lessons of LTCM were learnt later, with the failures of the Amaranth hedgefund in 2006, the two Bear Stearns hedge funds that found themselves on the wrong side of the subprime mortgage blow-out in 2007.

Hopefully, no hedgefund, however large and well-connected, will be bailed out ever gain with the assistance of the state or through conflicted private sector arrangements. The same ought to hold for private equity funds, investment funds of any kind, pension funds, insurance companies and investment banks.

Are commercial banks uniquely fragile and/or of unique systemic significance? Commercial banks used to have mainly deposits on the liability side of their balance sheets and loans and government debt on the asset side. Deposits, especially sight deposits, were an important member of the set of media of exchange/means of payments. Loans tended to be held to maturity and were an important part of the credit channel of monetary policy. Since sight deposits could be withdrawn on demand and were subject to a ‘sequential service constraint’ – first come, first served – and since loans were illiquid, banks were always potentially vulnerable to runs. With commercial banks both systemically important because of their role in the payment mechanism and vulnerable because of their strange combination of assets and liabilities, a case for considering a bank failure more important that the failure of any other company with comparable net worth could probably be made up till the 1960 or thereabouts.

But that was then and this is now. Today no individual commercial bank is of systemic significance and even collectively, the commercial banking sector is no longer something without which households and non-financial enterprises could not function effectively. Sight deposits are much less important as a medium of exchange/means of payment than they used to be, thanks to electronic funds transfer at the wholesale and retail levels and the use of credit cards, debt cards, travellers cheques, cash-on-a-chip and myriad other means of payment at the retail level. The large-scale clearing, payments and settlement mechanisms, both for inter-bank transactions and for securities not dependent on the survival of any private financial entity.

Deposits are increasingly just one liquid asset among many held for precautionary and portfolio investment reasons by households, non-financial enterprises and financial enterprises. Bank loans have been securitised and sold off, and government bonds have been joined on the asset side of banks’ balance sheets by equity, alternative investments. The asset side of commercial banks is looking increasingly like that of investment banks. On the liability side, deposits remain a unique feature of commercial banks, and the sequential service constraint means that they continue to be vulnerable to runs. Runs imply bankruptcy risk only if assets are illiquid however, and commercial banks assets are much more liquid than they used to be, because of securitisation than they used to be. In addition, bank failure or even failure of a large chunk of the entire commercial banking has lost most of its power to disrupt real economic activity because of the ‘demonetisation’ of bank deposits.

So, no Virginia, there are no private systemically important financial institutions any longer. Central banks are systemically important, as the only providers, when the chips are down, of unquestioned liquidity at no cost and without notice. The ‘plumbing’ of the payment, clearing and settlement systems has to be safeguarded against both liquidity crunches and operational risk. The central bank must ensure, through its discount window(s) (which must be open to all those capable of meeting the discount window collateral requirements), that those who need to borrow and possess collateral that would be good in normal times, can indeed borrow in abnormal times when disorderly markets and illiquidity threaten transactions between private parties.

Hedge funds, private equity funds, investment banks and commercial banks can, provided the central bank plays by the rules of the previous paragraph, be treated like the ball-bearings company in Wuppertal: there is no need for public sector involvement in the prevention of bankruptcy or for the condoning by the regulators of conflicted behaviour in privately orchestrated bail outs.

And, yes, IKB should have been left to sink or swim on its own. To see the main shareholder, itself a publicly-owned German development bank, put at risk public money to try and salvage its investment is not a pretty sight. Let's hope the EU Commission will come down on this weirdly perverse form of state aid (from the tax payers, organised by one agent of the state, via a second agent of the state to a private party in whom this second agent of the state has a large ownership stake) like the proverbial ton of bricks.

No comments: