spirals of death

Close observers of the US housing finance disaster in recent months will have noted a curious phenomenon. Companies such as Countrywide that were in late August regarded as rock solid have recently passed clearly into the danger zone while those like Fannie Mae and Freddie Mac that were regarded as potential market saviors have come under a cloud. In Britain Northern Rock, whose September bailout was said to be modest, involving little risk to the taxpayer has now turned into an immense 25 billion pound ($51 billion) potential black hole – real money even in the US economy let alone in the much smaller British one. This illustrates a deeply troubling quality of the largest downturns: the tendency for the free market to turn into a death spiral, in which even sound well-run institutions are engulfed.

Death spirals are fairly rare in financial history. The Wall Street Crash of 1929 was perhaps the most virulent example. After the first downturn, the market recovered for several months. Then the collapse of the Bank of the United States in December 1930, together with the further economic damage from the Smoot-Hawley Tariff caused a further collapse in confidence and activity that was concentrated in the banking sector, as relatively solid institutions followed the Bank of the United States into bankruptcy. The Federal Reserve failed to correct for the money supply contraction caused by the bank bankruptcies, leading the US economy further into the pit. The additional shove given by President Herbert Hoover’s 1932 tax increase was almost unnecessary; only the confidence brought by a new president (albeit with equally counterproductive economic policies) brought recovery from 1933. By the time the spiral was over, more than one fourth of the banks in the United States had gone bankrupt and the stock market had bottomed out at one tenth of its peak.

A second death spiral, with somewhat less dire economic consequences, occurred in Britain in 1973-74. Edward Heath’s government had removed the quantitative controls on bank lending in 1971, which resulted in an orgy of high risk lending against real estate, very similar to the recent episode in the US except that most of the loans were made against commercial real estate rather than housing. When the first major real estate lender, London and County Bank, collapsed in November 1973 another more conservative house, First National Finance (FNFC), was used as the epicenter of the “lifeboat” rescue organized by the Bank of England. However, the decline in confidence and real estate values quickly sucked FNFC into the maelstrom.

The lifeboat rescue fund grew larger and larger for more than a year as the stock market declined to record low levels, 70% below its 1972 high. Homebuilders such as Northern Developments, in no way involved in the original crash but dependent on bank lending, were dragged down. So were the two most important entrepreneurial finance houses, both internationally diversified and neither significantly involved in commercial real estate lending – Jessel Securities, founded by Oliver Jessel and Slater Walker, founded by Jim Slater.

Neither Jessel nor Slater had been aggressively run – indeed Jim Slater had begun de-leveraging a year before the crash, as he saw trouble coming – and no wrongdoing was proved against the head of either organization, yet by the end of 1975 both very substantial companies had gone bankrupt and neither founder played a significant further role in the British financial sector. This was a great pity: in losing Jessel and Slater Britain had lost not only their very able founders but the most aggressive entrepreneurial teams in the City of London, who might have been best able to compete against the foreign invasion when Britain deregulated the financial services sector in 1986.

The British experience of 1973-74 seems more like the current position in the United States. National policy is currently reasonably neutral, so far avoiding the twin dangers of protectionism and tax increases which caused the medium sized downturn of 1929-30 to turn into the Great Depression. The problem is concentrated in the property sector. However there are already worrying signs that the magic alchemy of modern finance, though such mechanisms as securitization vehicles whose funding falls apart and complex derivative securities that prove to be unsalable in a crisis, is causing the problem to metastasize. In the consumer sector, GMAC has reported problems with its automobile loan portfolio, while it appears that credit card debt quality is rapidly deteriorating. In the corporate loan sector, loans to aggressive leveraged buyouts have got in trouble, and loans to hedge funds and private equity funds have been sharply cut back. (The latter effect can be seen in the movement of the yen/dollar exchange rate from 120 to 108, as the hedge funds’ ”carry trade” positions have been de-leveraged.)

The “death spiral” characteristics of the current market are pretty clear. If Fed Chairman Ben Bernanke’s original estimate of subprime loan losses of $50-100 billion had been anywhere close to accurate, there would have been no problem. The market deals with difficulties of that size all the time, without significant effect on surrounding sectors. A few fringe operators go bankrupt, a few large houses show unexpected losses, and the overall market continues without a tremor. The collapse of the Amaranth hedge fund in September 2006 or that of Refco a year earlier were substantial events, causing losses to a number of those institutions’ business partners, but there was no question of any general market disturbance.

When the subprime problem first emerged in February, it appeared that it would also be limited. A number of subprime lenders, relatively insignificant institutions, were forced to shut down. However the general market appeared unaffected; its view appeared to be that the problem was localized and should have no effect on the real economy, nor even any great effect on the broader housing finance market.

August’s widening in Libor spreads, at which banks lend money to each other, should have told us that this problem would be different, and altogether more important. If leading banks were unable to assess each other’s credit quality for short term transactions then something much more serious was wrong than the collapse of a modest fringe sector of the housing finance market. The Fed’s chosen solution, dropping interest rates and pumping more money into the system, did not address the real problem and was thus useless, as it has since proved. It has only postponed the denouement for a few months and stored up further trouble with inflation.

Two factors are at play here. The first is sheer size. If as now appears likely the eventual losses in the home mortgage market do not total only $100 billion, but a figure much closer to $1 trillion, then the subprime debacle becomes something much more than a localized meltdown. $1 trillion of losses is 7% of US Gross Domestic Product. The market cannot absorb losses of that size without some major institutional bankruptcies or a lengthy recession. The closest equivalent problem is the savings and loan collapse of 1989-92; that caused a major housing downturn but only a minor recession. However its cost (mostly borne by the US taxpayer) of $176 billion was about 3% of 1990 US GDP, only half the size of the likely current losses on mortgage loans.

The second is lack of transparency, and the blow to confidence that comes from the dawning suspicion that a large portion of the derivatives and securitization mechanisms designed in the last quarter century are faulty. The unluckily timed implementation for years beginning after November 15 of FAS Rule 157, requiring banks to divide their assets into three levels according to their degree of marketability, has thrown an unwelcome spotlight on the problem. If Level 3 assets can be valued only by reference to an internal valuation model, and have been allowed to accrue value in banks’ financial statements for a decade or more (enabling hefty bonuses to their progenitors) then how do we know they are really worth anything close to what the model says, and how do we go about realizing them, in a market where confidence has vanished?

To ask those questions is to answer them. Since every incentive led bank mathematicians to devise models that maximized the reported value of the bank’s holdings, and since little or no market existed by which those values could be checked, it is likely that today those assets’ book values are highly overstated. Moreover, even in banks where the mathematicians and their bosses were scrupulously, even impossibly disinterested and intelligent, there still remains the problem that those assets are worth far less in a downturn, because their illiquidity makes them intrinsically unattractive in a market where liquidity has become once more important. Anyone who has attempted to sell venture capital positions in a bear market can attest to how rapidly and completely the value of such assets can disappear. It is thus perfectly possible that the true realizable value of “Level 3” holdings in a bear market is no more than 10% of their book value.

This immediately demonstrates the problem. Goldman Sachs, generally regarded as insulated from the subprime mortgage problem, has $72 billion of Level 3 assets; its capital is only $36 billion. If anything like 90% of the Level 3 assets’ value has to be written off, Goldman Sachs is insolvent. They do not have the option of acting like Nomura Securities did recently, selling everything possible and writing the remainder down to zero, because they would be without capital. Instead they are likely to be dragged kicking and screaming, quarter by quarter, to a gradual writedown and sale of their Level 3 assets, with their true position remaining undisclosed and obfuscated by meaninglessly optimistic statements by top management. Only the bonuses will survive, paid in cash and draining liquidity from the struggling company.

That’s what a death spiral looks like. The US survived the Great Depression, eventually, and Britain survived the 1973-74 debacle. However the market recovered only after it had plumbed depths previously thought impossible, at which even the soundest investments were trading far below their true value. After normality returned, the financial services landscape was very different, with many large and apparently solid houses having disappeared, a generation of participants reduced to driving taxis or selling apples and a generation of investors scarred by their losses and unwilling to return to the market. Emergency infusions of money, from the Fed or the taxpayers, generally do no good, only postponing the denouement and delaying the arrival of truly bargain price levels.

Such spirals of death represent the final definitive triumph of the bears.