Response by the Swedish Financial Supervisory Authority and Sveriges Riksbank to the Consultation by the Commission services on Credit Rating Agencies (CRAs), 5 September 2008

Assessment of the Commission proposal – no clear case for regulation and supervision of Credit Rating Agencies

There is no doubt that the Credit Rating Agencies (CRAs) played a role in the recent crisis. As pointed out by i.a. the FSF- report of 7 April 2008 (1), credit ratings were shown to be based on inadequate data and insufficient analysis. In particular, this seems to be true in the area of structured debt, as investors have found out the hard way. As also pointed out by the FSF, CRAs have a long way to go to restore confidence in their activities. This clearly means that there is scope for improvements in their operations.

However, we are not convinced that legislation is the right way forward, especially not in the form proposed by the Commission.

First, in line with the “better regulation” agenda, regulation should only be introduced after a full impact assessment. The benefits should be clear and there should be no good alternative. Such an analysis has not been presented so far. This is problematic in itself, and may also be a reason why the Commission fails to present a satisfactory motivation for its proposal. It is stated in the introductory section that “the main objective of the Commission proposal is to ensure that ratings are reliable and accurate pieces of information”. In our view, ensuring the accuracy of ratings can not, and should not, be the role of the public sector. Such a role would go far beyond what we normally see as the role for the state in a market economy. It is also not clear to what extent the authorities have the expertise to make an assessment of the accuracy of ratings.

For regulation to be considered, the Commission thus needs to present a clearer motive. In our opinion, the regulator, in this as in other areas, should intervene only in so far as a market failure has been identified. However, in this area such market failures are not evident. While there may be short-term conflicts of interest stemming from the fact that the rating agencies are normally paid by those rated, the industry depends on the clients’ trust. Clients’ trust in the CRA ratings will be directly related to the agencies’ reputation, implying that concern for long run profits should discipline their behaviour.

Clearly, the market mechanism will only work properly if the investors make informed decisions and have an interest in monitoring the CRAs. It has become increasingly clear that a number of investors have “over-relied” on ratings and bought products whose risks they did not understand. The fact that the major CRAs had earned a high reputation may have been part of the explanation. However, their reputation has been cashed in for now and investors are likely to question ratings more in the future – a trend we find healthy. Consequently, the industry is presently working actively to improve its code of conduct and its business models. Openness and transparency on e.g. methods and remuneration models is clearly of key importance to restoring confidence. Investors can be expected to take an active role in monitoring this work. We thus believe that the conclusion that the current self-regulatory approach is inadequate is drawn too lightly.

Second, the conclusion that regulation is needed seems be at odds with the Commission’s own analysis of the problem. It is clear that investors did not fully appreciate that credit ratings focused on the credit risk and risk of default within one year, and largely ignored other risk such as market risk and liquidity risks. It is also possible that the use of credit ratings in some regulations, e.g. CRD, has given the ratings an undeserved status among investors. However, this problem can hardly be solved by regulating the CRAs and putting them under public supervision. Even if such supervision is not intended to evaluate the ratings themselves, there is a clear risk that investors will perceive the ratings as being approved by the government or at least that the supervision will lead to better ratings. In such a setting, it is difficult to see that investors in general would rely less on credit ratings and more on their own independent assessments. An official oversight mechanism for rating agencies could come back to haunt us if it turns out that banks have trusted a rating agency to which the authorities have given some form of approval. The obvious conclusion should instead be to reduce any reliance on credit ratings in regulations, i.e. to decouple the ratings from any public authority.

Third, the Commission proposal seems to imply the build-up of a complicated and costly supervisory structure. Given the concentrated and largely international dimension of the rating industry, the value added of such a European structure would seem to be limited.

The way forward – strengthened market based monitoring and better use of supervisory tools already at hand

Resources for legislation and supervision are limited. We should take a step back and focus on mending what needs to be fixed. Given that the rating industry is currently working hard to restore confidence, this is presently not a priority area for new legislation and supervision at the EU level. It will be useful to await the results of this work before contemplating measures at the EU level.

Should it be decided that there is a need to take measures at the EU level, any policy proposal must be clearly motivated and tailored to address the problem at hand. The present proposal for regulation of the CRAs does not fulfil this requirement. We can thus not support the introduction of public supervision of the CRAs. Neither of the two supervisory options proposed should be explored further unless the existence of a market failure is clearly identified.

Instead, it would be useful to give support to the ongoing initiatives to enhance openness and transparency. In line with the CESR conclusions, the IOSCO Code should be the standard on which CRA conduct of business should be assessed. A revision of the code in order to benefit from recent experiences and to ensure its continued relevance will be necessary. An example of a measure which would increase transparency would be to clearly signal the different characteristics of traditional ratings and ratings of structured products, for instance by introducing separate rating scales.

Also along the lines of the conclusions of CESR, we support the formation of an international body, formed of representatives of the industries concerned, to monitor the compliance of the IOSCO code of conduct.

The key issue for EU policy makers should be to ensure that the institutions which are of a broader importance to the financial system or whose activities are of important interest to consumers are supervised adequately. It will be necessary to review the role of ratings in the regulations regularly.

There also appears to be scope for using more actively the tools already available to supervisors in order to ensure that the institutions do not rely overly on external ratings and make an adequate risk assessment.

Note
1)  “Report of the Financial Stability Forum on Enhancing Market and Institutional Resilience”.
http://www.fsforum.org/publications/r_0804.pdf.