The Hindu Business Line : Investing faith in rating agencies

Isolated anecdotal instances are not a robust way of evaluating rating agency performance. They should be judged on the basis of a long-term record that shows that the higher ratings were consistently more stable and less likely to default than lower ones.

Roopa Kudva

The article “Credit rating agencies — answerable to none?”, which appeared in these columns on November 3, raised several important issues, not only for credit rating agencies but also the wider financial market. Additionally, there has been much debate around the business model of rating agencies, specifically on conflicts of interest in the issuer-pays model.

Who should pay?

The complaint against the issuer-pays model — where the entity issuing debt pays for the rating — is that it compromises the quality of analysis and ratings. Some suggest an investor-pays model instead, while others recommend third-party involvement such as a regulator.

Will the investor-pays model work? When a rating is assigned, the investor is generally not known.If investors were to pay for ratings, then only those paying will have access to the ratings. Lenders and the market cannot benefit from the ratings. Today, all the ratings are available to all — including retail investors — free of charge, and are widely disseminated by agency Web sites and the media because the issuers pay for them.

The issuer-pays model also gives rating agencies easy access to company managements, which provide insights into strategy that might otherwise not be widely known, and help the rating agencies evaluate them better. It is hard to imagine this level of information-sharing under an investor-pays model.

The issuer-pays model enables rating agencies to provide a quality and depth of analysis to the market that public-information-based opinions and model-driven approaches cannot.

Room for bias

The past 12 years have seen 500 long-term rating downgrades by CRISIL against 200 upgrades, which clearly rules out bias, if any, towards issuers.

Would the investor-pays model be free of conflicts? Hypothetically, if the issuer-pays model was to result in higher-than-warranted ratings, then the converse would be true for the investor-pays model. In the latter, rating agencies could give companies lower ratings than merited so that investors could get a higher yield than warranted. And pressure from investors to prevent rating downgrades could increase sharply, as downgrades result in mark-to-market losses on rated securities.

Are externally-specified rating methodologies the answer then? These would, at best, reduce the rating exercise to a mechanical checkbox approach. At worst, in the absence of upgraded methodologies that keep pace with market realities, rating agencies would miss out on key credit issues in a highly dynamic environment.

Anecdotal impressions are relevant, but may prove inaccurate in evaluating a rating agency’s performance.

Rating the raters

Isolated anecdotal instances are not a robust way of evaluating rating agency performance. How should the raters be evaluated? First, on the basis of a long-term record that shows that the higher ratings were consistently more stable and less likely to default than lower ones. Adherence to clear and widely-disseminated criteria is another parameter.

Governance practices are also critical in evaluating the independence of rating agencies. These include: multi-layer decision-making processes, independent criteria and quality assurance teams, non-involvement of analyst in fee decisions, strict trading policies for analysts, and de-linking analyst compensation from assigned ratings.

The Indian rating industry scores well on these parameters. It has a two-decade history of credible default statistics. Criteria are published and actively discussed with investors. CRISIL, for instance, holds investor discussion forums, where its methodologies and views are intensely debated by over 100 analysts each time. It has proactively provided a report card on its performance against many of these evaluation parameters.

Comparing rating agency performance across disparate global and Indian debt markets is specious. We believe it is inappropriate to draw conclusions on the Indian financial market from experiences in the US. While it is good to learn from global events, differences between markets should be acknowledged. The present problems in the US came about because sub-prime mortgages were bundled into securitised instruments; in India, sub-prime mortgages as a category does not exist, and the number of mortgage securitisation transactions has been minuscule in comparison.

In other ways, too, it is misleading to compare these markets: corporate debt in India is still primarily a loan market, while the US has for decades had a deep corporate bond market. CRISIL began using separate symbols to help investors identify complex structured instruments in 1992; global regulators are only now considering this measure. CRISIL added complexity levels as an additional disclosure on all its ratings nine months ago. This is a global first and a pro-bono service from CRISIL.

Indian credit rating agencies ARE answerable, as they are regulated by SEBI and subject to intense market scrutiny. SEBI, in fact, specifies a rigorous code of conduct and operating procedures for the agencies which include continuous monitoring of all ratings, prompt dissemination of ratings, detailed disclosure and strong compliance.

scrutiny and achievements

Further, rating agencies are constantly scrutinised by investors, the media and regulators.

Structurally, rating agencies are the most independent among all the providers of credit opinions — borrowers, lenders, investment bankers and brokers. Rating agencies alone have no interest in the success or otherwise of the rated transaction. And ratings have played a vital role in India’s credit markets: in the 1990s, rating agencies proactively flagged the impending risks in the NBFC sector well ahead of time. CRISIL was the first to highlight the contingent liabilities of State government guarantees as a risk. The potential problem in collective investment plantation schemes was nipped in the bud by rating agency alerts.

The market has keenly appreciated CRISIL’s actions highlighting the heightened risks of leverage in corporate balance sheets. Not surprisingly, Indian investors accepted ratings long before regulation recognised them. This bears testimony to the central role that rating agencies play in the Indian market.

(The author is Managing Director and CEO, CRISIL Ltd. blfeedback@thehindu.co.in)

Related Stories:
Why weren’t the rating agencies up to grade?