Charles Cronin says that credit rating agencies must rebuild investors’ confidence in the way they rate structured products
Like the story of the emperor’s new clothes, the unravelling of the structured finance markets over the past year has exposed the failings of the world’s leading credit ratings agencies (CRAs). The difference is that while the emperor lost only his dignity, many institutional and retail investors, who based their investment decisions on ratings, lost their shirts.
Only last month, Standard & Poor’s came clean about the likelihood of investors recovering any of the money they have invested in structured bonds such as collateralised debt obligations (CDOs). S&P said that for any deals rated A or lower, recoveries were likely to be zero, while recoveries for AA-rated slices of such deals would be at best 5%. The most senior, or ‘super-senior’, AAA-rated tranches were likely to recover 60%, while junior AAA-rated tranches could expect to recover only 35%.
Of course, caveat emptor should be the guiding principle behind investment decisions, and investors can never duck the responsibility for conducting adequate due diligence on their investments. Indeed the CFA Institute code of ethics and standards of professional conduct stipulates that members must “exercise diligence, independence, and thoroughness in analysing investments, making investment recommendations, and taking investment actions” and “have a reasonable and adequate basis, supported by appropriate research and investigation, for any investment analysis, recommendation, or action”.
Unfortunately, many investment professionals, fiduciaries and managers had come to rely on the opinions of the CRAs, and, more importantly, the label ‘investment grade’ on those bonds with a rating triple-B and above. They were lulled into a false sense of security that these bonds carried a ‘seal of approval’ from the CRAs and as such were unlikely to default.The crisis has put the spotlight on the underlying methodologies and assumptions behind structured products. The mathematics quantifying the risk profile of these asset pools is beyond the comprehension of most investors. It is taken on trust that the maths experts know what they are doing and that their conclusions are robust. As can be plainly seen, there has been some recalibration, and this has accounted for much of the structured ratings downgrades. While recently travelling in Brussels I met a veteran of this business who said some of the “fudges” adopted by the CRAs “would have been laughed out of my Oxford mathematics forum”.
Where laymen could use their judgement in the credit rating process was in reviewing the underlying assumptions that went into the ratings. This, we understand, was proprietary to the CRAs. For example, in building a structured product the CRA may use assumptions of 15% house price inflation over the next two years and 7% for the following 10 years. This is an opinion, and someone considering investing in the product may hold a different opinion. Knowledge of the underlying assumptions is material to the investment decision-making process, but is denied to the investor. Hence, we support changes advocating full-unrestricted publication of CRA assumptions. After all, this is common practice in the equity market.
The CFA Institute Centre for Financial Market Integrity (The CFA Institute Centre), which seeks to maintain the highest ethical standards for the investment community, believes that the CRAs contributed to the collapse in the structured finance markets by relying too heavily on, and neglecting to disclose the limitations of, their statistical analyses. Furthermore, while the CRAs can always defend themselves by saying that their credit ratings have only ever reflected the probability of default, they did little to correct the growing perception among investors that ratings also gave an opinion on liquidity and volatility. CRAs could have done more to dispel this myth through greater disclosure of their assumptions specifically about correlations.
The practice of ‘notching’, or unsolicited rating a company or structured product by a CRA should raise the regulator’s eyebrow. A senior executive of a German company told me how he was telephoned by a CRA agency offering to rate his company’s ‘unrated’ debt. He was flattered and told them to go ahead. However, when the salesman pointed out this would cost a considerable amount of money, the executive declined.
The salesman then told the executive that the agency intended to assign a rating to the debt anyway, but would be unable to conduct full enquiries. A rating was duly assigned, which surprised the market and increased the company’s cost of capital. I cannot comment on whether the market had correctly priced this debt, before or after the unsolicited rating, but the sales technique is certainly worrying.
It is all too easy to make CRAs the whipping boys for the CDO collapse. After all, financial engineering is not an exact science. In the past, CRAs have provided an invaluable service to financial markets, in particular to the end-investor who often has little or no impartial information to hand. The CFA Institute Centre believes that the CRAs can rebuild investor confidence by making some adjustments to their business practices.
The CFA Institute Centre has made the following suggestions to both the International Organization of Securities Commissions (IOSCO) and the Committee of European Securities Regulators (CESR), reviewing the conduct of CRAs:
CRAs should use a rating nomenclature or categorisation that distinguishes structured products from both corporate and commercial paper ratings to help investors recognise the differences. For example instead of calling it AAA, call it treble alpha (ααα). Most debt funds are credit constrained under their investment policy statements, such as no lower than single A. By changing the nomenclature of structured products, they could only be included in the portfolio if an allocation was permitted by the trustee/fiduciary. CRAs should refine or otherwise eliminate the concept of ‘investment grade’. CRAs should encourage a global best practice of prohibiting ‘notching.’ CRAs should create an executive-level compliance officer position in their organisations to ensure implementation and enforcement of the codes set out by IOSCO and CESR, and require complete adoption of the codes to claim compliance.
CRAs should refrain from rating new structured products until the statistical data are sufficiently robust to produce a defensible rating.
The CFA Institute Centre also recommends that the CRAs use fair value reporting for financial instruments. Financial reports should show the fair value of all assets and liabilities that affect a company’s economic performance and condition to ensure that investors have a basis for evaluating asset values. CRAs should also conduct their own relevance and reliability tests on the robustness of their methodologies and assumptions, for example in the case of default, recovery rates and correlations.
The overall aim should be to put investors’ interests first. Regulators in all markets should ensure that investment firms and investment managers adopt, implement and enforce procedures to ensure the instruments they sell or use match their clients’ circumstances, goals, objectives and preferences.
As an advocate of self-regulation, we believe that the CRAs should organise themselves with a strong self-regulatory body, such as a Compulsory Association regulated by law. Such organisations are motivated to self-regulate in order to ward off government intervention. To this end, they endeavour to maintain a high degree of public confidence by setting standards to promote the public good and self-regulate the behaviour of their members.
CRAs provide an important service to the credit markets, and for that service to be valued in the future they must take the lead in their own reform.
Charles Cronin is head of the CFA Institute Centre EMEA