Rating the reforms

The financial crisis that started with the credit crunch in 2007 demonstrated ratings agencies’ business models needed serious reform at best. More scathingly, some have argued ratings agencies were one of the direct causes of the crisis. The agencies themselves have been busy trying to restore their reputations. Have they done enough?

The ratings agencies have not had an easy few years. The US Financial Crisis Inquiry Commission concluded that “the failures of credit ratings agencies were essential cogs in the wheel of financial destruction,” to the point that the “crisis could not have happened without the ratings agencies.”

Yet their businesses are still thriving. Critics say the problems affecting the ratings agencies are deep and systemic. Issuers can shop for the rating they want, so ratings agencies will always be under pressure to win business by offering good ratings. This is more than just a theoretical problem.

“Participants in the securitisation industry realised they needed to secure favourable credit ratings in order to sell structured products to investors,” said the FCIC. “Investment banks therefore paid handsome fees to the ratings agencies to obtain the desired ratings.“ The ratings agencies were under enormous pressure to win business, with issuers naturally desperate to influence them in any way possible. A product’s ratings have huge implications in terms of price and which investors can buy it, while at the parent level it determines borrowing costs. It would be naive to doubt issuers pressure agencies for a favourable outcome. But because the ratings process is so opaque, it is very difficult – if not impossible – to prove fraud, if ratings have been relaxed in order to win business.

Even if rules are not relaxed in a cynical bid to win business, there is concern the agencies have grown too close to the issuers they analyse. It is a form of quasi-regulatory capture, said Eric Kolchinsky, a consultant at the National Association of Insurance Commissioners, and a former managing director at Moody’s. [♦Kolchinsky has a pending lawsuit against Moody’s.]

When regulators spend too much time talking to a particular type of financial institution, their outlook can come to resemble that of the entities they are supposed to be policing. The same is to some extent true of ratings agencies, said Kolchinsky. Credit analysts spend so much time speaking to issuers that they can come to make similar assumptions about how they will perform in stressed markets, and what their ratings should be. “Its a bit like Stockholm syndrome,” said Kolchinsky. “If you spend that much time talking to issuers you start to see the world as they do.”

The ratings agencies have acknowledged the problems and moved to address them. S&P identified four main areas it needed to strengthen to restore its credibility in February 2008, in the immediate aftermath of the credit crunch, when swathes of Triple A rated securities were defaulting: governance; analytics; information; and investor education. Moody’s identified six areas: strengthening analytical integrity of ratings; enhancing consistency across ratings groups; improving transparency; increasing resources; managing conflicts of interest; and pursuing industry and market initiatives.

The Financial Crisis Inquiry Commission is blunter in its own list of failings identified at Moody’s, on which it based its case study, identifying “flawed computer models, the pressure from financial firms that paid for the ratings, the relentless drive for market share, the lack of resources to do the job despite record profits, and the absence of meaningful public oversight.”

Comparability

One of the key tenets to the ratings agency philosophy is the importance of comparability. “When ratings are comparable across sectors, investors can better use them to compare the credit risk of securities in different sectors, thereby helping them to assess whether there are potential discrepancies in the pricing of credit risk in different sectors,” wrote S&P in a report in November 2009.

“When ratings are comparable, an investor can more easily assess the yield offered by a bond relative to its risk as compared with other bonds from other sectors. We intend for each rating symbol – for example, AAA – to connote a comparable overall view of creditworthiness, wherever and whenever it appears,” said S&P in early 2011 – whether that related to a Bund, a Treasury or an RMBS. But the determination to provide comparable ratings across asset classes rings hollow when the agencies are so reluctant to take steps that would allow for greater comparability of ratings of the same types of assets but between ratings agencies.

While it does make sense for all types of credit to work to the same scorecard, so investors can make assessments between asset classes, in practice there are fundamental differences between sovereign credits, corporate credits and structured finance instruments, for example, in terms of volatility and the likelihood of extreme losses when things go wrong.

What needs to happen, said Kolchinsky, is for the agencies to be made to follow the model used by the accountancy firms. There, as with ratings agencies, the client pays for the service, and is free to take its business to any institution it wishes to. But although there is choice in the identity of the accountant, there is no choice in the methodology used to keep or audit the accounts.

All accountants use the same, fully transparent process, meaning there can be no shopping for favourable treatment, and any fraud is identifiable because the process used is transparent and understood by impartial third parties. This is currently not the case with the ratings agencies, which all have their own methodologies, which only they fully understand.

Their ratings do not even mean the same thing: while S&P and Fitch ratings evaluate probability of default, Moody’s evaluates the expected loss.However, a source at one of the ratings agencies dismissed this analogy as like comparing apples and oranges. One accountant or auditor expresses gives a definitive judgement on a company, while ratings agencies are just one opinion in the market, he said.

Neither would it be desirable, said Martin Winn, spokesperson at S&P. It is, he said, “the opposite of what market organisations and users of ratings have told the various recent enquiries into ratings in Europe and the US. What investors actually want to see is a diversity of clearly explained ratings opinions, so they can take their own view on which ratings are credible and useful and which are not.”

Any move towards one-size-fits-all common ratings, he insisted, “would in fact run counter to the efforts that regulators are taking to reduce any over-dependence on ratings, as it would convey a sense that uniform ratings carries an official stamp of approval.”

Standardisation, not just here but in the whole process of how ratings are assessed, the relative weightings of different factors and what exactly is taken into account, would increase transparency and remove some of the incentive, and most of the opportunity, for ratings agencies to bend the rules to win business.

Reform: quantity or quality?

All the ratings agencies have made a raft of changes to their businesses in the years since 2007. In 2009 S&P adapted its methodology for analysing structured finance instruments by introducing stress testing, subjecting securities to conditions it claims are comparable with the US Great Depression, and awarding AAA only when such conditions do not cause losses. Moody’s cites its additional disclosures on ratings announcements and its structured finance indicator as particularly important changes.

The ratings agencies all now use a suffix for the ratings on structured finance transactions, giving investors additional information, although their exact definition of what constitutes such a transaction is slightly different. The change is welcome, but is perhaps not the kind of radical step the market expected back in 2008 when the full extent of the crisis first revealed itself.

In December 2010 S&P made further changes to the structured finance ratings criteria that it said were in some respects “one of the most wide reaching criteria changes in recent years; it potentially affects nearly 25% of the outstanding, structured finance ratings globally.” The changes hinge on the principle of replacement of a counterparty if its creditworthiness deteriorates below specified thresholds, providing an explanation of how a rating is influenced by the rating of its counterparties and the strength of a replacement mechanism.

Fitch introduced recovery ratings and loss-severity ratings to measure the loss given default risk and recovery potential on performing and distressed structured finance securities. In January this year S&P acknowledged its ratings methodology for banks had failed to adequately reflect the unique economic and industry risks afflicting the sector, with changes expected there. In late 2010 it also proposed changes to sovereign ratings that would reflect the risk placed on some countries by their financial sectors, by refining the calibration of the five major ratings factors used.

All the agencies have both expanded their headcounts, with Moody’s in particular singling out research for added resources, while S&P requires staff pass extra exams in order to participate in ratings. S&P has also tightened its control framework and compliance reviews. Websites have been revamped to make information more accessible, and publications are more widely circulated. Moody’s emphasises measures such as the separation of the teams awarding original ratings and those responsible for ongoing surveillance.

Moody’s even argues it has made “more fundamental changes to our systems and operations” than those imposed on it by regulations, for example with the provision of independent oversight at the board level. It has also endeavoured to implement changes globally even where required changes only apply in certain countries – because, it argues, this is in line with its goal of delivering consistent ratings across jurisdictions.

However, some question whether these changes address the real issues of systemic risk, or do anything that will prevent another credit crunch situation, where highly rated securities turn out to be worthless.

“Nothing has really changed,” said one market observer. Much of the reform has been largely cosmetic or ineffective. There is a big element of moral hazard at play: investors that should be driving the process are not as motivated as they should be because governments, in the end, are the real losers when crises hit. And the taxpayers who are on the hook do not understand the complexities of the situation.

Take look-back analysis, for example, a solution to the perceived problem that rating agencies might want to favour clients they hoped to one day work for. “I think that the premise underlying the issue is just flat wrong. If a banker has a docile analyst he would want that analyst to remain where he is – at the rating agency,” the observer said. This would ensure the rating agency remained compliant, maintaining the status quo. In addition, “a docile individual does not have the right personality to be a banker. The people who I’ve seen hired out of rating agencies are those who are able to stand up to bankers.”

The agencies, of course, insist the changes are substantive. “The changes we have made independently – and the tough new regulatory regime that has been put in place for ratings worldwide – are fundamental and far reaching,” said Winn. “Our organisation, and the environment in which we are operating, are radically different than was the case a few years ago. Our ratings are more transparent and more consistent, and understanding of them in the market is much greater. We are now one of the most closely scrutinised and supervised corners of the financial system.”