Between 12 and 21 December 2006 the data for asset-backed securities index futures (ABX 2006-1 AAA spread) showed a small but significant departure from its normal daily pattern of price fluctuations. This coincided with rumours that a trading division of a major US investment bank had expertly quit its collateralised debt obligation (CDO) portfolio. Over the ensuing months, analyst reports began to circulate that led some banks to delve into their loan portfolios to see if they had exposures to these securities. What sparked this activity? The US housing prices had begun to trend down, many âlow startâ loans were approaching anniversaries when their interest payments would âstep upâ. In February 2007, HSBC recognised the problem in its New York branch with a $10.5 billion charge. Market makers reacted by forcing a 350% increase in volatility.
Hitherto, the originating banks had bundled up these higher interest loans and securitised them. The resulting pools of securitised debt were re-packaged into CDOs, where equity and debt investors could participate in a preferred tranche and trade off a levered return against reduced exposure to losses on default relative to the overall pool of loans. New products were dreamt up, such as funds that borrowed to lever further the promised returns that would rid the originating investment banks of the inventories of higher risk equity tranches that were accumulating in their books.
Underpinning the market were assumptions about average rates of default that these loans would see. The US government had encouraged products like low-start loans as a means of providing home finance to borrowers who had until then been excluded from home ownership. With little credit history to draw on, analysts had to strike an educated guess as to the likely number of defaults. From the beginning of rapid growth in âno-documentâ loans in 2004, it took until early 2007, when low interest payment inducements were due to expire, for the divergent trend of high defaults to emerge.
Possibly dismissed as âtoo soon to callâ, the âlow-docâ CDO machine continued to revolve at a breakneck pace. Documentation processes for these instruments, and indeed many other credit derivative contracts, fell way behind. The NY Fed called several round table meetings to gauge the depth of the problem and set an agreed remediation path. Yet new tickets1 continued, seemingly unabated, to be written.
Re-engineering the higher expected default frequency into securitisation and CDO models showed that these structures were over-promising returns. Higher frequency of defaults, like peeling away the layers of an onion, meant for CDO structures not only that regular interest payments to the investors were jeopardised, but that the expected loss of all capital increased. Curiously to some, the valuation effect was often most damaging in the so-called âsuper seniorâ tranches, rated âAAAâ.
By mid-2007, the approaching savage repricing had yet to occur but the market place seemed to have reached saturation, and the new issues market showed signs of imminent closure. Some structured credit funds managed by the US investment house Bear Stearns imploded. As if to preserve capital, banks began to enforce tighter underwriting standards across their businesses. There was little they could do for loans already âbaked into their bookâ, so the emphasis turned to re-financings and holdings of traded securities.
This hardening of underwriting standards saw hitherto high-quality securitisations or revolving finance facilities supported by even low-risk assets, such as portfolios of shopping centres, falter (and borrowers forced to liquidation). Banks also withdrew their support for loan notes issued by securitised vehicles.
Little did they at first realise, but to get the securitised assets off their balance sheets in the first place, the banks were required to provide a standby line of liquidity that would purchase the loan notes should a market buyer not be found. These requirements were, for the most part, an unplanned call on their balance sheets. Liquidity started to become scarcer.
Organisations such as Northern Rock, whose liability management was reliant on an âopenâ securitisation market, fell to a run by its retail depositors. A Canadian money market fund broke the buck. A CDO fund had to close. Around this time, with sponsorship by the Fed, Bank of America purchased Countrywide, a major West Coast provider of âlow docâ loans. Credit markets generally began to sell off in an orderly way, while most other markets adjusted, with a lag, to tightening credit spreads. It was not until the northern winter before equity markets began to reverse their positive trend.
The manufacturers of CDOs to varying degrees financed their warehouses and production lines using repurchase agreements2 (repos). Again, these contracts took assets and liabilities off balance sheet. But, as with the standby lines in securitisation structures, to maintain contracts off balance sheet the borrower was obliged to top up collateral as its value deteriorated. Liquidity tightened further. Parcels of CDOs were being sold off as repos struggled to be rolled over or renewed. The increased supply of paper for sale saw lower and lower prices register, while waning confidence in the value of CDOs aggravated the liquidity effect. Liquidity became unobtainable for some borrowers.
The pressure to liquidate a difficult-to-value balance sheet became too much for Bear Stearns. In February 2008, it was sold to JP Morgan after the NY Fed established a company to buy $30 billion worth of assets at just 7% of their price immediately before trading in Bear Stearns stock was suspended.
CDOs continued to lose value. Statutory profits and reporting fell hostage to the accounting classification for warehoused stock. With each reporting period, waves of securities became categorised âlevel 3â, automatically requiring greater capital to be assigned to them. As it reeled from these effects, it became increasingly obvious that the banking system in the US was systemically undercapitalised. The Fed and US Treasury stepped in to restore order, but it was not until the decision to allow Lehman Brothers to file for bankruptcy that it became clear that the regulators had overestimated the effectiveness of their regulations.
In 2008 Comptroller Dugan reflected that while âinvestors should never rely exclusively on credit ratings in making investment decisions, the plain fact is that triple A credit ratings are a powerful green light for conservative investors all over the worldâ.3 And so it had been that in the preceding period re-packaging of these securities had not only spread throughout the US banks and money market funds, but to US agencies and beyond its shores to the balance sheets of financial institutions in other countries.
Write-downs and illiquidity combined to take down banks that had aggressively made loans to property developers. Icelandic banks failed, with the assets of the three banks taken over by its supervisor equal to more than 11 times the country's GDP â an early indication of the effective limits of financial globalisation. German Landesbanks in time announced their losses and needed to be re-capitalised. In the UK, banks were crippled, partly due to write-downs to goodwill and loan impairment on the balance sheets of recently purchased Dutch banking assets. And so it played out across most of the northern hemisphere and parts of Asia.
From initial insouciance about the prospect of Lehman Brothers filing for Chapter 11 following its death spiral of the preceding months, the act itself led to something of an avalanche throughout the US â investors in money market funds that held now worthless Lehman Brothers debt were unable to redeem shares for their $1 face value. A run on money market funds ensued, clogging the repo market that broker dealers depended on, and obliging the US Treasury to step in to guarantee money market funds. The global insurer AIG also needed volumes of support, as it had been the dominant provider of credit derivatives through its London branch.
Money market funds sought to liquidate assets to meet redemptions and fund the additional collateral now needed to secure financing. At the same time, the margins required in the repo market continued to increase, and it became impossible to use some assets as collateral for loans. Organisations that could, called funds from their foreign affiliates in order to meet the obligations of their US operations. US affiliates, mostly branches, of foreign banks were under even greater liquidity pressure than were local banks, partly because their access to deposits from US citizens and residents was limited. This forced them to raise funds from outside the US, further contributing to international contagion of the liquidity squeeze, with the effect felt most acutely in London.
As valuation of financial assets became more an art than a science, asset markets such as equities were sold down around the world, as agents sought to raise liquidity wherever they could. Major exchanges generally behaved well, but the continued downtrend was further fuelled by investors withdrawing commitments to managed money, leading to the closure of many funds.
Part of the problem in funding liquidity was that some assets pledged as security had been re-hypothecated by holders of the collateral. When collateral was recalled, often the specific paper was unavailable, obliging investors to sell other assets to meet their own margin calls and redemptions.
One consequence of the turmoil was that collateral requirements could not be calculated in the usual way. This is because margins and collateral requirements are calculated using statistical processes that draw on recent asset price return history. The turbulence now prevailing meant that these processes broke down, particularly for non-traded and over-the-counter instruments. Futures exchanges and prime brokers were thus obliged, in effect, to guess the amount of collateral required. This meant that âChinese wallsâ, intended to avoid conflicts of interests within organisations, were tested, as prime brokers set collateral levels at capitulation points for âclientsâ who were in some cases part of the same organisation.
Shortly after the failure of Lehman Brothers the system became unmanageable in private hands. Governments, advised by their regulators, stepped in.
All will remember the steps taken by governments to traverse the crisis with the Troubled Asset Relief Program, and ultimately quantitative easing by the Fed in the US, the recapitalisation of British banks with public money, the decision of the Irish government to guarantee all bank deposits (which later led to its call for help from the European Union) and the need for the Swiss authorities to save UBS.
After throwing a TARP-o-line over impaired assets and entities and providing extraordinary amounts of public support, regulators set about digesting the outcomes. The âlive laboratory testâ of failure presented them with many avenues to follow as they set about re-calibrating regulation and supervisory procedures.