When the history of the credit bubble comes to be written, the list of guilty actors will be lengthy indeed: irresponsible banks, greedy borrowers, foolish speculators, incompetent regulators, the central bankers who kept rates too low for too long. Yet one group of players has been especially culpable in creating the current mess.
The credit rating agencies have played a pivotal role in the global debt markets for over thirty years, with the stamp of approval from Moody’s, Standard and Poor’s or Fitch a prerequisite for the sale of a bond. The agencies’ scale of ratings – from AAA for an issuer of unimpeachable creditworthiness, to the Cs and Ds for issuers of highly speculative or defaulted securities (often called “junk”) – form the basis of the capital markets.
Credit ratings are written into the regulators’ rulebooks as the basis for bank capital requirements, and dictate how fund managers allocate their assets. For example, AAA ratings are a prerequisite for money market funds, the pooled vehicles that invest surplus cash and which have liquidity and safety of capital as their primary concerns.
In view of this important status, the rating agencies themselves have to be approved by national regulators – in the US, for example, an agency must be granted Nationally Recognized Statistical Rating Organization (or ‘NRSRO’) status by the Securities and Exchange Commission.
Historically the “gold-plated” AAA rating was earned only by certain developed country governments and a few dozen corporations with strong balance sheets and a minimal risk of default. The number of AAA-rated US corporates actually dropped from 60 or so in the 1960s to 9 in 2005, as there was a general decline in creditworthiness. At the other end of the scale, C and D ratings (usually called “junk”) applied to issuers either in default on bond payments or at imminent risk of default. In between, the grades of double and single A and the “non-investment grade” BBB and below served to highlight subtle differences in borrowers’ ability to pay.
For most of their history, the rating agencies performed the worthy if rather boring task of applying ratings and then moving issuers up and down the scale as economic circumstances merited.
How the credit rating agencies’ role has changed
However in the last decade a whole new area of business opened up, one far more lucrative than the bread and butter business of analysing company balance sheets or government accounts. The boom in structured finance – best described as the repackaging, restructuring and resale of existing debts – placed the ratings agencies in a qualitatively different role. Instead of rating already extant bond issues, they became intimately involved in the issuance process itself, advising the investment bankers and their clients on how to obtain the necessary ratings for their new loans.
The structured finance bond issuance production line offered fat fees to bond underwriters (the investment banks) and to the rating agencies. Moody’s alone earned nearly US$1 billion a year from rating structured finance issues in 2005 and 2006, dwarfing revenues from the more traditional activity of rating government, municipal and corporate bonds.
With such earnings on offer, a conflict of interest inherent in the agencies’ remuneration scheme started to lead them increasingly astray. The agencies had always been paid by the issuers themselves for rating new bonds. Historically this had not been seen as a major problem. True, some investors complained that the raters were too slow to downgrade the bonds of issuers that were getting into trouble, for fear of jeopardising commercial interests (revenue from future bond issuance programmes). In several cases (for example, the Russian debt default of 1998) the ratings downgrades came late into the process of financial deterioration, after investors had already suffered heavy losses.
In the area of structured finance, this built-in conflict of interest had far more grievous repercussions.
Structured finance is intimately linked to the financial technique of securitisation, the pooling and repackaging of existing cash flows into a range of senior and subordinate tranches. Here, Wall Street financial engineers achieved a remarkable result, achieved by a kind of alchemy. By collecting and pooling a range of, say, BBB-rated credits, and then arranging them into a series of tranches which are paid off sequentially, they could create a new set of financial instruments with different levels of risk.
The alchemy that led to more AAA ratings
Then, by modelling the likely default rates, the bankers could present the new bonds to credit rating agencies and get their approval. Here, seduced by the fees that securitisation offered and turning a blind eye to the fact that the models were in many cases untested or based on only a few years of benign economic data, the raters were only too happy to grant AAA-ratings to the top (senior) tranches.
The alchemy in the equation was the realisation that the mere securitisation process was enough to grant higher ratings overall – the (securitised) whole was, according to the raters, worth more, sometimes a lot more, than the starting pool of assets, if one judged by the ratings alone.
This might appear contrary to the laws of nature, similar to a perpetual motion machine. Indeed, many sceptics suggested that the securitised whole could not be worth more than the sum of the parts. If the risks appeared lower, they said, then it was because they were being hidden in the structure.
Such views remain heretical: securitisation has been a money machine for the banks for over twenty years, and has been the lynchpin of the enormous expansion of the global credit market. Yet in recent years the technique might be described as an all-devouring monster, as it was applied to debts of lower and lower quality.
US subprime market: a real money-spinner
The US sub-prime market became the most egregious example of the abuse of the credit rating process and, ultimately, of common sense. Mortgage loans made to lower-income and financially unsound Americans, who had in many cases lied about their incomes and borrowed many multiples of their (fictitious) salaries, were pooled and securitised, with their top-rated tranches granted the gold standard rating of AAA.
Not coincidentally, the commissions and fees involved in the sub-prime lending business were the highest of all, as the end borrower was being charged interest rates well above the the base or wholesale rate in recognition of his or her poor credit standing.
When bonds securitised in this way began to slide in price during 2007, as the US property market began to fall and it became clear that many of the underlying borrowers would default, the rating agencies initially did nothing. Eventually, when the AAA-rated tranches had lost up to 30% of their value, and the lower-rated pieces had become close to worthless, they moved to downgrade many of the issues.
However, rather than confront and explain the scale of their errors, the agencies made a sly and retrospective change in their policy. No-one had ever intended AAA for a securitised bond to mean the same as AAA for a government or corporate issue, they (quietly) argued.
If this was indeed the case, it was news to many investors. Many funds have lost a lot of money by holding what they thought were safe bonds. In the case of many hedge funds specialising in credit, such as the Bear Stearns funds that failed last summer, or the Australian Basis Capital, investors lost all their money, as leverage quickly reduced capital to zero. While it is difficult to feel sympathy for an investor in a hedge fund, other, more vulnerable investors have also been hit.
The Florida local government investment pool had heavy losses and could not meet redemption requests in November, freezing the cash of hundreds of school districts and towns. Across the Atlantic and far from the tropics, sub-prime losses from bonds issued by Citigroup hit eight northern Norwegian towns, threatening school budgets and elderly care homes.
Bond insurers begin to feel the strain
Meanwhile, as politicians worldwide argued that the sub-prime crisis was contained and markets would soon recover, another key group of participants in the global credit markets started to come under heavy strain. The bond insurers, Ambac and MBIA, used their AAA ratings to enhance the credit of hundreds of thousands of other issuers, and were heavily involved in the structured credit market. As sub-prime losses mounted, the likely payouts on defaulting bonds began to threaten the insurers’own financial health.
For MBIA and Ambac, the maintenance of the highest rating was critical, as any downgrade would automatically transmit itself to the many bonds they insured, threatening a cascade of falling prices. By the end of 2007 it became clear that the insurers’ likely losses would bankrupt them unless they had a multi-billion infusion of new capital. The companies’ share prices reflected this, losing most of their value during the year. The cost of insuring against the bond insurers’ default (easily gauged from the credit derivatives market) at the same rose to levels consistent with a rating in the Cs or Ds – that is, the market had performed its own assessment of these companies’ creditworthiness, and classified them as junk.
Nevertheless, to date Moody’s and Standard and Poor’s have stubbornly refused to downgrade the bond insurers, and their AAA rating remains. Although it is likely that they have come under heavy pressure from the banks and, behind the scenes, from regulators, to avoid a downgrade, the agencies’ unwillingness to recognise the insurers’ high risk of default flies in the face of reality, and is attracting increasing criticism.
It is also perhaps the culmination of the long process of loss of integrity that we have outlined. From acting in their first two decades as the investor’s friend, the credit rating agencies had become thoroughly corrupted by the peak of the bubble in 2007.
A lesson for investors: beware conflicts of interest
For them, the outlook is bleak. Their reputation indelibly tarnished, there must be a significant chance that they will in due course fall victim to lawsuits and bankruptcy themselves
However the credit bubble’s deflation will have much wider repercussions. The Basle-II accord, which sets bank capital adequacy requirements on the basis of credit ratings, and depends on discredited financial risk models such as value at risk (VaR), will probably have to be scrapped or rewritten. Many investment and insurance agreements that depend on credit ratings to allocate capital will also need to be amended. The discrediting of the existing ratings system has also seriously damaged secondary market liquidity: whole swathes of the bond market have simply stopped trading.
Recognising the gravity of the credit market seizure, the US central bank has already started an aggressive series of interest rate cuts. Whether this has more than a temporary effect in restoring confidence to an economic system overburdened by poor-quality debts remains to be seen. Yet it is unlikely that the cure for the economy can be more borrowing, which is the origin of the problem. The level of saving in all the Anglo-Saxon economies will have to increase, and this will come at the cost of economic growth, which over the last decade has depended on ever-greater levels of debt. Ironically, rate cuts may impede the return to normality by hurting savers and forcing them to save more for longer, thus dragging out the downturn.
For the investor, the lesson learned is an old one – to pay attention to conflicts of interest. Just as the internet bubble showed us that stock tips given by investment banks with a vested interest in selling new shares were worthless, the credit rating scandal has reminded us that we cannot trust a system where the raters are paid by the issuers.
Paul Amery is an independent financial analyst based in London, formerly a fund manager and bond trader