Respected Vice-Chancellor Dr. D.C. Reddy, Professor Kishan Rao, and, friends,

I am thankful to my large circle of friends who organised this gathering, and gave me an opportunity to be at home in Osmania University today. My association with the University is forty years old, a long time indeed, and in several capacities, as a research scholar, lecturer and Visiting Professor.

Dr. Nagaraj was my senior colleague and we had many things in common, including Bellary-connections. Dr. Nagaraj had zest for life, affection for all and was charming to everyone. He had a sharp wit and impressive articulation befitting the most popular teacher of economics. Dr. Nagaraj always had inquisitiveness that I admired. Appropriately, the subject of today’s address is a somewhat recent phenomenon in our country viz. credit rating. This is a subject relevant not only to students of economics, management and commerce but also to millions of savers and scores of borrowers. Of course, it has contemporary policy significance, since Reserve Bank of India (RBI) is emphasising development of debt markets, for which credible ratings are vital. Well-understood and efficient debt markets are critical at this stage of our economic progress. Large financing requirements of physical infrastructure need them. One of the lessons to be learnt from the recent East-Asian crisis is the importance of debt markets for financial stability.

I will give an overview of credit rating, and agencies involved in such rating, both domestic and international. The benefits expected and criticisms levelled will also be recalled. The main focus of the address would be the issues relating to sovereign rating and use of credit rating by regulators, especially in the banking sector. From the Reserve Bank of India’s direct concern, the subject has to be viewed in terms of sovereign rating of India, rating of States in India and the use of rating in regulating financial institutions such as banks and non-bank financial companies (NBFCs) as well as instruments such as Commercial Paper (CP); this in no way detracts from the importance of rating for corporates. It must be noted that the emphasis here is on the use of rating in the regulatory framework, especially of the RBI and not on regulating credit rating agencies, which is entirely in the jurisdiction of the Securities and Exchange Board of India (SEBI).

Credit, in simple terms signifies lending with corresponding borrowing and a process of repayment, usually at an agreed rate of interest. In the distant past, credit was essentially to finance seasonal agricultural operations, and hence highly localised. It became less localised when it financed trade. With the growth of industry and increased capital-intensity of production-process, banks became significant players in the business of credit, accepting deposits from savers and advancing loans to industry or corporates. However, in the evolution of financial markets, an important phase was the issue of bonds by corporates as a mechanism for raising money directly from savers. In such a process, there was a problem in ensuring that the savers have full information and a reasonable basis for assessing the worth of issuers of bonds or any such debt instrument. Credit rating as a business proposition evolved to address this problem.

It should be clear that availability of accurate and timely information is vital for the health and efficiency of financial markets. However, lack of symmetric information between lenders or savers and borrowers or issuers is generally observed since, for example, a manager of a project who issues bonds to borrow funds will have more information about a project than a lender (i.e.) investor in the bonds. For savers, gathering of such information is costly and time consuming. In a global environment, this asymmetry is even greater and the costs of collecting information is even higher. The practice of credit rating and the emergence of Credit Rating Agencies (CRAs) for the purpose are meant to help mitigate this problem of asymmetric information. A CRA may be defined as one which is engaged in the business of rating securities, and rating in turn means an opinion of the CRA regarding default probability of such securities, expressed in a standardised manner and made public. Default is generally taken to be a missed or delayed disbursement of interest/principal, bankruptcy, receivership or a distressed exchange. Rating is mostly solicited by issuers for specific debt issues, although unsolicited debt issues are not uncommon.

CRA is thus different from a mercantile credit agency like Dunn & Bradstreet, which usually supplies general information on corporates. It is also different from a credit bureau, which collates information on credit record of corporates or even individuals. Nor is it a credit assessing agency like the credit department of a commercial bank. The most significant aspect of credit rating is that it is an opinion made available for public, influencing decisions by participants in financial markets.

CRAs often provide information and insights to financial markets but even if they merely summarise or collate existing information, they offer service in summarising the existing disjointed information, and giving an assessment. To the extent CRAs play a vital role in functioning of financial markets, not only the savers or investors in bonds, but the regulators also have a stake.

The modern rating system dates back to 1909 when John Moody started rating US railroad bonds. Currently, four rating agencies dominate the international scene. They are Moody’s, Standard and Poor’s, Fitch IBCA and Duff & Phelps. While normally CRAs assign a rating on the request of an issuer, there are occasions when unsolicited ratings are assigned, and in many such cases, the fact that they are unsolicited is made explicit with an asterisk.

While the rating of corporate bonds started in early twentieth century, sovereign ratings represent a relatively new line of business for the agencies. The first industrial country to be rated was France, by S&P in 1959. Both Moody’s and S&P rated a non-industrial country, namely, Venezuela as recently as October 1977. Fitch IBCA entered the business of sovereign rating only in 1975.

In cases where sovereign does not seek a rating, but a corporate entity of such a country seeks a rating, CRAs do assign an implicit sovereign rating.

The Credit Rating Information Services of India Limited (CRISIL) initiated the concept of credit rating in India. CRISIL was established in 1987 and started operations in January 1998. Currently, four rating agencies are in operation in India, rating bonds, time deposits, CP and structured obligations. All the four Indian rating agencies have tie ups/alliances with international rating agencies - CRISIL with S&P, ICRA with Moody’s, CARE with Fitch IBCA and DCR (India) Pvt. Ltd. with Duff & Phelps.

A credit rating may be defined as an opinion of a CRA as to the issuer’s (i.e. borrower of money) capacity to meet its financial obligations to the depositor or bondholder (i.e. lender of money) on a particular issue or type of instrument (i.e. a domestic or foreign currency : short-term or medium or long-term, etc.) in a timely manner. In other words, the rating measures the relative risk of an issuer’s ability and willingness to repay both interest and principal over the period of the rated instrument. To put it differently, rating signifies the default probability of the instrument that is rated. The ratings are usually provided through a simple symbol system like AAA that we are as familiar with. It could be BBB or Baa3 or BBB- etc.

Broadly speaking, ratings are divided by the CRAs into three levels, viz. investment-grade, non-investment grade and default grade. The investment-grade is considered by CRAs to be a significantly safer grade than the rest.

The formal ratings are accompanied by detailed rationale, which some of the users and those rated study carefully to get nuances of assessment. The ratings thus indicated are subject to revisions. The CRAs keep reviewing the ratings, and upgrade or downgrade them as the circumstances change. There are reviews at fixed periods, say biannually as also whenever events occur or circumstances warrant a review of rating.

More important, the CRAs often put up a rating on what is called "watch", which alerts the concerned about an imminent review. They also indicate the purpose of watch, usually with positive outlook or neutral or negative outlook.

CRAs themselves explain what their ratings are meant to convey. They explain that ratings are meant to provide capital market participants with a single framework for comparing "credit quality". It is made clear that the credit quality is determined by its default probability, which in turn, depends on the ability as well as willingness to pay both interest and principal within the stipulated time schedule. Thus, "credit quality" is a multi-dimensional concept that comprises default probability, loss severity, financial strength, etc. While the composition is common for credit ratings, the weight of each component that results in an aggregate credit quality may differ even if the final rating turns out to be the same.

Different agencies give different weights to several factors that go into rating. Thus a CRA may give greater weight to political factors while giving a lesser weight to technical (economic or financial) factors. It must be noted, therefore, that similar ratings of different CRAs are not strictly comparable.

CRAs often seem to more in tandem but there are rating differences at any time. In any case, many participants do supplement information available from CRAs on rating with internal research or study before arriving at a decision to invest in a debt instrument.

Perhaps it would also be useful to explain what credit rating does not connote.

First, a rating is specific to the issue or debt or instrument that is rated. A rating is neither a general purpose evaluation nor overall assessment of credit risk associated in all debts contracted by an issuer.

Second, it is not a recommendation to buy, hold or sell. It is an opinion, perhaps well-informed opinion.

Third, they are not predictors of default but opinions about the relative probability of default and loss. Thus, the difference between the highest rated instrument and another rated a rung lower is that the probability of default of interest, and principal in the case of the former is lower than that of the latter.

Fourth, ratings are not guarantees against losses. Under no conditions do they or can they predict losses due to ‘shocks’ or highly unexpected situations.

Fifth, credit ratings relate only to credit and thus, for example has no relationship to risk preferences of investors or attractiveness of equity. Hence, the perceptions of different stakeholders viz., creditors, lenders, shareholders, etc. in responding to ratings could be different.

Benefits Expected

There are three main stakeholders in the credit rating by CRAs, viz., issuer, investor and regulator. Each stakeholder expects some benefit or usefulness.

The issuer expects enhanced access to borrowed funds, by diversifying funding base within a country and reducing geographical concentration. In fact, in some cases, ratings provide access to international pools of capital. Rating can, in several instances, open funding alternatives in the form of lengthening of maturities, diversity of institutions and diversification of currencies. The wider access to investor-base and investing instruments also increases financial flexibility and thus lowers funding cost over the long term. In fact, the benefits to issuer accrue as a result of potential benefits to investors and in many ways they share the benefit of wider, healthier and more efficient debt markets, attributable to the role of CRAs. Credit ratings are critical in all securitisation programmes. Credit rating in issues where repayments are backed by receivables help the issuer in raising funds at rates finer than what its own independent rating would indicate.

Investors utilise the rating to supplement their own credit evaluation process, especially when they do not have adequate resources or time or access to management to undertake independent risk analysis. Investors can view the rating as an opportunity to enter into new markets, domestic or foreign. In this regard, the information component of a credit rating is valuable to investors. Further, the information component helps in setting credit risk limits. For example, investors may decide not to invest in bonds below a certain investment rating. On the contrary, they may invest in an instrument just because it has a certain credit rating, which matches with its investment policy. Operationally, the markets often use credit rating to determine the credit risk premium. Some of the large CRAs publish default probability studies that are used by financial institutions to quantify credit risk exposures.

Regulators also benefit from the ratings of CRAs. The ratings and the accompanying detailed analysis by CRAs help in disseminating information and imparting transparency to all, especially to small investors, who otherwise may not have access to such information, and thus have made the regulators task less onerous. Regulators rely on credit ratings for a number of purposes. For example, some regulators prescribe that pension funds invest in debt instruments of investment rating only. Credit rating has also been prescribed by regulators as the basis for exemption from registration norms for issuers of asset-backed securities. Regulators of stock exchanges also use credit rating for prescribing margin money for brokers and dealers. Some regulators insist on a minimum rating for corporates to become eligible for issuance of CP. Specific credit rating limits have been made one of the eligibility criteria for issuance of bonds in some countries. At times, ratings from two CRAs have been prescribed for bond issues of a large size. In some countries, capital requirements for banks and security houses have been based on ratings. Minimum rating criteria has also been prescribed for acceptance of public deposits by companies by a few regulators. The regulatory capital charges in respect of insurance companies are also often related to credit rating.

A number of criticisms have been levelled against credit ratings or CRAs, and indeed of the whole credit rating system. It is necessary to recognise them and assess the validity of such criticisms.

First, it has been suggested that since issuers are charged for ratings by CRAs (i.e. the issues are paymasters), the independence of ratings becomes questionable. According to the argument, the CRAs may be tempted to assign higher ratings than warranted or hesitate to downgrade issuers from fear of spoiling business relationships. The argument adduced against this notion is that the reputational risk that CRAs face provides an overriding incentive to maintain high quality and accurate ratings.

Second, there have been suggestions that CRAs are not accountable for the ratings given by them. The rating agencies argue that every time a rating is assigned to the agency’s name, justification is given, and in any case, the agency’s integrity and credibility are on line.

Third, as observed in some cases, ratings may lead to herding behaviour thereby increasing the volatility of capital flows. This criticism gained ground during the Asian crisis when many commentators argued that the downgrading of the crisis-hit countries during a crisis might have worsened rather than help the situation. Rating agencies argue that ratings are not intended to predict the exact timing of default or when a crisis would occur and that change in rating would occur if the new information received so warranted. CRAs point out that most of the lending in East Asia was done by big banks with their own analytical capacity.

Fourth, it has been argued that credit ratings change infrequently since the rating agencies are unable to constantly monitor developments. Furthermore, owing to time and cost constraints, credit ratings are unable to capture all the characteristics of an issuer and issue. CRAs argue that they supplement their ratings with credit watches and outlook designed to indicate the agencies’ perspectives on factors that might prompt a rating review over a future period.

It is often argued that rating change affects prices and quantities since it forces certain portfolio managers to sell. There is also a view that prices may not be affected since the market would have already factored the developments leading to announcement. The results of studies are not uniform. One set of studies found that positive announcements in ratings were followed by movements in bond yield in the expected direction while negative rating changes did not have significant effect on yield movements. In contrast to these results, another set of studies found that yield movements occurred only when a downgrade in rating was announced. However, an analysis by IMF has shown that the largest announcement effects are noticed in respect of emerging market sovereign spreads. In other words, impact of rating changes is far higher in respect of emerging market economies and hence of special concern to India.

As mentioned, international rating agencies faced severe criticism in the wake of the Asian crisis since they facilitated large flows; they did not anticipate the events in Asia, and later they appear to have overreacted in a panic of downgrades attenuating the falling trend in currencies. Some of them have openly admitted their mistakes. They have announced that they have since changed their rating methodology to take into consideration the dynamics of capital flows. For instance, increased emphasis is now being placed on the proportion of short-term debt, private sector external debt, soundness of banks and corporates, etc.

Sovereign borrowers usually enjoy the highest credit standing for obligations in their own currency. If they retain the right to print money, the question of default is largely an academic one. The risk is that a country may service its debt through excessive money creation, effectively eroding the value of its obligations through inflation. When a sovereign borrows in foreign currency, the position is obviously different, since the sovereign borrower has to make available foreign currency to service the debt.

Sovereign risk generally refers to two types of risk : either a default by the sovereign Government on its foreign currency obligations or the risk that the country’s lack of foreign exchange will cause default of other entities in the country. The task of the credit rating agency is also to assess the sovereign’s ability and willingness to ensure availability of the foreign exchange necessary to meet obligations. Thus, while assessing a foreign currency bond of a corporate, not only the commercial risk of the corporate is relevant but also the currency-transformation risk attributable to macro policies of the sovereign. Hence, the sovereign rating is also important for other issuers in the country such as banks, financial institutions, public sector units, corporates, etc. The sovereign rating virtually forms a ceiling above which it is normally not possible for other borrowers to rise unless the issuer is issuing higher quality asset based securities, such as against export receivables.

As articulated by the Bank for International Settlements (BIS), sovereign rating is different from rating corporates in several ways. Sovereign ratings do not have long track record and the past record of sovereign defaults provides less reliable guide to the assessment of risk today, than in the case of a corporate. The reasons for a corporate default are easier to discern than that of sovereign, since in the case of former there are legal structures that can seize assets and change management. There is possibly a greater element of judgements or human will in sovereign defaults causing less predictability. Sovereign has the option, in extreme cases, of being able to access official sources for borrowings which is conditioned by non-economic factors. The responses of creditors to a potential sovereign default could also be different. While there are many common features between sovereign and corporate in debt markets, there are three special factors in assessing default probability of a sovereign, viz., behaviour of sovereign, of other official lenders and creditors themselves.

In view of the above nature of sovereign ratings, it is necessary to understand the factors that appear to determine the ratings. Empirical studies have tried to identify factors that explain over 90 per cent of the cross sectional variance of sovereign ratings. First, the most potent factor was GDP per capita in US dollar terms. High ratings were associated with high per capita income. Secondly, high level of economic development measured by the IMF’s classification as an industrial country accounted for a full rating increase. Thirdly, the history of default played a major role. For instance, a recent history of default set a rating back by about two notches. Fourthly, the rate of growth in Gross Domestic Product was a positive factor. Fifthly, high inflation was a negative factor. Sixthly, a country’s net foreign debt position was also negative. Finally, the political factors, especially political stability and willingness as well as ability to ensure appropriate macro economic policies do weigh heavily in rating.

The extent to which changes in sovereign ratings affect market prices and credit availability is difficult to assess. If the information available to rating agencies is also available to a significant number of participants on a real time basis and a larger number of investors or investment advisers make their own analysis, then changes in ratings may not exert important influence on market prices. The issue perhaps is two fold, viz., whether rating change is ‘news’ and more important whether such a change makes a difference in the applicable regulatory framework.

India has been assigned sovereign ratings by Moody’s and S & P for over a decade. In addition to these two agencies, India is also being rated by Duff and Phelps, Japan Credit Rating Agency, Japan Bond Research Institute and Fitch-IBCA. Moody’s started rating India in 1988 with an investment grade rating. S & P started rating India informally, also with an investment grade rating in August 1990. The external crisis of 1991 prompted, both these agencies to downgrade India’s long-term sovereign credit rating to non-investment grade. After the crisis, in 1994, Moody’s had moved India back to investment grade but by June 1998, it again brought it down to non-investment grade. S & P has kept India just below the investment grade since the Gulf Crisis. Recently, both these agencies affirmed their positive outlook on India. The international rating agencies are yet to increase India’s rating to investment grade.

Our external debt servicing ability is unquestionable in terms of track record, manageable magnitude of debt and capacity to service debt both in the short and the long run. At the time of crisis in 1991, India took upon itself the full burden rather than default on any obligation, and in fact, all obligations were met on the respective due dates. Unquestionably, Indian economy is far stronger now. In view of this unblemished record and current economic strength, it becomes difficult to explain the CRAs delay in upgrading India’s credit rating to investment grade. The ratings of India by these two agencies would appear to convey that India in 1998 is no better off than it was in 1991. This is not intended even by the rating agencies. The only explanation for this is the dynamics of rating wherein there are different inter-temporal standards of assessment.

Indian credit rating agencies rate domestic currency obligations and the States of the Union, somewhat indirectly. Implicit sovereign ratings of the States can be derived from their State level undertakings or special purpose vehicles set up with State Government guarantees. In fact, both the market as well as CRAs themselves treat such ratings as a proxy for rating of State Governments. In doing so, CRAs take into consideration the economic structure of States, finances of State Governments and the economic and political management including Human Development Index. However, when a State Government requests or solicits a rating for itself, the CRA takes up rating as has been done in one instance recently. While viewing a rating, it is necessary to appreciate the basis. For example, in the case of rating of a State, a CRA states that this rating view is relative to that of the other States of the Indian Union and with the Union Government of India viewed as "Highest Safety" for local currency obligations. Yet, another important consideration in rating of a State is its contingent liabilities, where unlike actual liabilities the obligations of State depend upon occurrence of a discrete event.

The guarantees, whether of the Union or the States, can and do involve a range of terminology with corresponding legal impact on financial liability. It is in this context that use of terminologies such as letters of comfort, and conditional or unconditional default guarantees become critical. Governments are well advised to note that once the legal liability on the guarantee gets established, it is akin to Governments’ own liabilities. However, in the case of letters of comfort, though such an automatic equation with a Government’s liability in respect of guarantees may or may not be technically appropriate, the wording of the letter could have some form of holding out expectation, and in any case, a Government may not ignore reputational risk involved in repudiation of moral obligations implicit in the issue of such a letter. In fact, international practice appears to treat the letter of comfort of a sovereign on par with a guarantee.

Regulators of both developed and emerging markets rely on credit ratings for a variety of purposes. USA introduced the concept of regulatory use of ratings in 1931. The Office of the Comptroller of Currency used ratings as a means to determine the basis of valuation of bonds. The use of ratings spread to other activities such as determination of capital prescription or margin money for brokers/dealers, disclosure requirements under Securities and Exchange Commission norms, exemption from registration and regulation for certain issuers of asset-backed securities, etc. The National Association of Insurance Commissioners (NAIC), which determines insurance company’s regulatory capital charges, also relies on ratings.

Japan promoted credit ratings in 1974 and regulators used the ratings of Japan Bond Research Institute (a rating agency) as one of the eligibility criteria for bond issues in the 1980s. The Ministry of Finance relies on ratings in a variety of ways, including regulation of money reserve funds.

In 1993, the European Community stipulated capital requirements for market risk for banks and security houses based on ratings. UK adopted rating based Capital Adequacy Directives in 1996. Favoured treatment is also accorded to firms engaged in securities business based on rating. France, Italy, Australia, Switzerland, Canada, Argentina, Chile, Mexico, Indonesia, Korea, Malaysia, Philippines, Taiwan Province of China and Thailand are other countries that have regulatory uses for ratings. In fact, the adoption of rating based regulations was the main force leading to the creation of rating agencies in emerging markets in Latin America and Asia.

An important issue is the criteria for recognising a CRA for use of its ratings in regulation. It is now commonly accepted that criteria are : assured continuous objectivity in methodology; independence from outside influences; credibility, though this should not be an entry barrier; access to all parties with legitimate interest; and adequacy of resources. Most regulators stipulate a list of recognised agencies whose ratings can be used to satisfy rating requirements. Broadly, there are three areas where extensive use is made of ratings in the regulatory process, viz., investment restrictions on regulated institutions; establishing capital requirements for financial and disclosure as well as issuance requirements. The issues faced by regulators in use of ratings include reconciling divergent ratings by different CRAs and deciding cut-off of level of ratings.

The RBI prescribes a number of regulatory uses of ratings. The RBI requires that a NBFC must have minimum investment grade credit rating if it intends to accept public deposits. Furthermore, unrated or underrated NBFCs in the category of equipment leasing and hire purchase finance companies are required to disclose the fact of their being unrated, to the public, if they intend raising deposits. Finally, as per money-market regulations of RBI, a corporate must get an issue of CP rated and can issue such paper subject to a minimum rating.

In the area of investments, SEBI stipulated that ratings are compulsory on all public issues of debentures with maturity exceeding 18 months. SEBI has also made ratings mandatory for acceptance of public deposits by Collective Investment Schemes. If the size of the issue is larger than Rs.100 crore, two ratings are required.

Pension funds can only invest in debt-securities that have two ratings, as per the stipulations of Government of India.

Of particular interest to policy-makers and of course to market participants is implications of use of ratings by regulators.

First, reputed rating agencies feel that when regulators rely on ratings, there is a potential for laxity. It tends to create demand for easy and accommodating raters. In other words, it stimulates `rating shopping’ i.e., issuers will be inclined to seek the cheapest and least demanding credit rating agency.

Second, and following from the earlier point, it creates a continuous revenue source for rating agencies.

Third, the ratings induced by regulatory requirement could induce ritualism and even enhance risk. Thus, issuers will get rated because they are required to do so and investors will tend to use it as a proxy for internal analysis. It is argued that in such situations, ratings may actually end up increasing the risk.

Fourth, regulatory use of ratings tends to provide official sanction for rating agencies’ assumptions and their ratings, thus increasing the risk of moral hazard. Sometimes, it is argued that while assessing creditworthiness of Government owned banks or public sector units, ratings may be higher than justified, though there is no evidence to support this argument.

Fifth, the more profuse use of ratings would increase the demand for ratings and, therefore, indirectly also increase the influence of CRA on financial markets. Thus, there is a potential challenge to the financial stability by CRA.

Sixth, it is held that when regulators tend to rely on ratings, often it seems to be under an assumption of uniform/standard ratings and rating criteria by different rating agencies, an assumption that is not necessarily valid at all times.

Seventh, there is an expectation, especially among some developing countries that ratings by themselves can be used as a stimulus to developing financial markets without realising that the requirement of mandatory ratings on domestic debt instruments etc. are no substitutes to a strong regulatory framework, transparency, accounting and disclosure standards. There is indeed the bigger issue of institutional factors relevant to debt and ratings at best is one aspect.

Finally, it is clear that there are many dilemmas for regulators in using the ratings. Prescription of eligibility criteria based on reputation of the rating agency, length of service, acceptability by market participants, and actual track record would only entrench the oligopolistic nature of the market. On the other hand, easy eligibility criteria would increase the laxity and moral hazard. There have been suggestions to develop official substitutes for regulatory induced rating, and this idea is not favoured by many since there is a fear of possible political influence on official bodies.

Another suggestion to increase accountability of CRAs whose ratings are used by regulators is prescription of codes of conduct for rating agencies to enhance market discipline. CRAs could be encouraged to disclose relevant aspects of their organisation, including among others, their ownership, staff, etc. They could be asked to disclose a record of their assessments and results. They may also be asked to publicise more explicitly their methodologies. The entire process of ratings could be codified by developing modalities for consultation, sharing of preliminary ratings, detailed explanations etc.

You would be aware that capital adequacy norms for banks in India are based on the Capital Accord of the Basle Committee. The Committee prescribed risk weighted approach for assets, with weights assigned to both on-and off-balance sheet exposure of a bank. According to the Committee, a minimum standard of 8 per cent capital adequacy ratio was to be put in place by 1992. The Accord had also prescribed that capital should be defined in two tiers, Tier I and Tier II. At least 50 per cent of capital should be in Tier I.

In the light of further developments, particularly in the banking sector, Basle Committee on Banking Supervision has issued a Consultative Paper on ‘A New Capital Adequacy Framework’ in June 1999. The New Accord of June 1999 proposes that capital adequacy rules be more closely aligned with risk profiles. The proposed capital framework rests on three pillars, viz., minimum capital requirements, supervisory review process and effective use of market discipline. The Basle Committee has acknowledged that the methods used to determine the capital charges for credit risk in the Current Accord is not overly sophisticated, especially in the context of financial innovations. Furthermore, the growing complexity of financial transactions has reduced the efficacy of the present system. The Committee has, therefore, proposed to replace the current standardised approach with preferential risk weighting.

To this end, the Committee proposes to use external credit assessments such as credit ratings for distinguishing credit risk. At the same time, the Committee has expressed some concerns about the incentives and the potential impact on account of the enhanced role for credit agencies’ ratings. The Committee has, therefore, cautioned supervisors against depending on rating agencies’ ratings in a mechanical fashion, and has recommended that banks should develop internal rating mechanisms simultaneously.

Despite the apparently guarded approach of the Committee to the use of rating by CRAs, there has been serious concern over this issue. For example, use of external ratings for capital adequacy may prompt a bank to exercise less care in granting credit. That would be contrary to the objective of setting up a supervisory framework in such a way that the banks themselves evolve interest in improving internally developed risk management techniques. In addition, it might be argued that rating agencies are being given undeserved influence in the context of regulating banks, because they determine by their rating, how high the capital requirements for banks should be. Moreover, it is also feared that strong competition among the agencies might lead to a dilution of the rating or that enterprises requiring a rating might engage in a kind of rating shopping.

It would be interesting to observe the comments of two major and internationally reputed CRAs on the New Accord. S & P is of the opinion that external ratings should supplement and not replace prudent lending standards by banks. S & P encourages the Basle Committee to "establish a methodology for reconciling the existing diverse set of policies, practices and rating scales currently in use among rating agencies or among the banks". Moody’s is also candid in limiting the role of external ratings in the New Accord. In their view, currently, credit ratings cover only a small portion of most banks’ portfolios and the external ratings approach is unresponsive to the unique needs of individual banks. Its implementation would introduce a number of adverse incentives into the credit risk services industry. They recommend external ratings as an interim measure until banks develop internal skills. They also advocate development of a criteria for rating agency recognition, which will balance the independence of a rating agency and setting of minimum standards.

In the aftermath of the Asian crisis and the scathing criticism on the failure of CRAs to predict the crisis and later on its role in precipitating it through downgrades, the role of credit rating agencies has been placed under microscopic scrutiny. The merits and demerits of regulating credit rating agencies and the issue of rating the rating agencies have been discussed in many international fora.

There is no international regulatory authority overseeing rating agencies. Whether they are regulated or not depends on specific country circumstances. In general, however, countries impose a modest regulation over CRAs. In USA, Securities and Exchange Commission gives recognition to CRAs as Nationally Recognised Statistical Rating Organisations (NRSO) for specific purposes. The main form of regulation is USA is in officially recognising a CRA. Thereafter, there is hardly any regulation. Similarly in UK, recognition as a rating agency is required from the Financial Services Authority (FSA). So is the case in Japan, Australia, France and Spain.

In India, in 1998, SEBI constituted a Committee to look into draft regulation for CRAs that were prepared internally by SEBI. The Committee held the view that in keeping with international practice, SEBI Act 1992 should be amended to bring CRAs outside the purview of SEBI for a variety of reasons. According to the Committee, a regulator will not be in a position to objectively judge the appropriateness of one rating over another. The competency and the credibility of a rating and CRA should be judged by the market, based on historical record, and not by a regulator. The Committee suggested that instead of regulation, SEBI could just recognise certain agencies for particular purposes only, such as allowing ratings by CRAs recognised by it for inclusion in the public/rights issue offer documents.

In consultation with Government, in July 1999, SEBI issued a notification bringing the CRAs under its regulatory ambit in exercise of powers conferred on it by Section 30 read with Section 11 of the SEBI Act 1992. The Act now requires all CRAs to be registered with SEBI. Since then, all the four CRAs in India have been registered with SEBI. SEBI Act now defines "credit rating agency", "rating", and "securities". Details of who could promote a CRA and their eligibility criteria are specified. The Act also mentions about agreement with clients, method of monitoring of ratings, procedures for review of ratings, disclosure of ratings and submission of details to SEBI and stock exchanges. Restrictions have now been placed on CRAs from rating securities issued by promoters or companies connected with promoters i.e. companies in which directors of CRAs are interested as directors.

It is clear that the credit ratings are playing increasingly important role in financial markets. The most significant change in the recent past relates to emphasis on their accountability and more important, the caution in regulators use of ratings. In India, rating is a more recent phenomenon, but the changing global perspectives on the subject do impact the financial system. In the light of the East Asian experience, it is clear that appropriate disclosure of information and accounting standards across the board are necessary to help viable rating systems. While freedom of expression and independence of CRAs would also help improve the systems, credit awareness by investors, especially on the operations of rating system needs to be encouraged. Several issues and dilemmas being faced by all stakeholders in the matter of credit rating should be self-evident from the presentation so far. However, flagging a few specific issues of policy significance to the RBI at the current juncture would be appropriate.

1. On Sovereign Rating of India by global CRAs, in the light of what has been explained earlier in this address, clearly there is a need to improve our articulation of both the country's social ethos and our economic strengths and thus enable the CRAs also to enhance their appreciation of our track record and strengths in the external sector. I had articulated the importance of assessment of behaviour of a sovereign in the context of assessment of sovereign rating and it is exactly here that the social ethos and national character demonstrated by our track record of early ‘nineties amidst political uncertainties needs to be reckoned.
2. Sometimes, it is argued that such an enhancement of our credit rating to investment grade would facilitate a sovereign borrowing by India. I believe that, for several reasons, it is neither necessary nor desirable at this juncture for Government of India to seek sovereign borrowing in foreign currency in international financial markets, but the issue could be reviewed sometime in future. Though sovereign borrowing is advocated for India on the ground that it will help develop a benchmark favourable to bond issues by our corporates in international markets, there are at present several disadvantages in sovereign borrowing that outweigh, in my view, the extent of possible benefit of benchmarking to our corporates.
3. I had explained earlier in the speech about the sovereign rating and guarantees in the context of States. The major issue in this regard relates to treatment by CRAs of ‘letters of comfort’ issued by Central and State Governments as closer to Government guarantees, while at the same time, the concerned Governments treat them as farther from Government guarantees. There is need for early resolution of this issue since the magnitudes of bond issues by central and state level enterprises are increasing with significant implications for both financial markets and fiscal transparency.
4. The State Governments have an increasing stake in obtaining appropriate rating either direct or implicit. The subscription to the borrowing programmes of different States with regard to both the amounts offered and the rate at which they are offered is being influenced by the bankers’ perceptions of the State Government and hence some relevance for rating. When the enterprises of a State seek borrowings through the State guaranteed bonds or Special Purpose Vehicles, the rating becomes critical. In fact, recently rating agencies are even articulating their expectations with regard to fiscal policies of specific States in this context. For example, a recent statement of a rating agency on the subject states that the situation calls for fiscal consolidation by improving revenue performance and restructuring expenditure in order to sustain the creditworthiness and performance of the States economy. The RBI, as debt manager of the States has to take into account the perception of investors in States’ securities and hence the stance of rating agencies. The biannual conference of State Finance Secretaries is being convened by the RBI on April 29, 2000 and the status of State finances is an important item on the agenda.
5. The quality of rating is likely to be enhanced with a firmer bankruptcy law and its effective implementation. Currently, RBI has taken steps to strengthen the Debt Recovery Tribunals and this should also contribute to speedy recovery by banks, thus improving the quality of credit rating.
6. Several initiatives with regard to the operational aspects of CRAs are being considered to meet emerging situations and these deserve to be encouraged. There is a view that CRAs could play a more active role in monitoring the structures of structured obligations that they have rated. It is held that CRAs may take into consideration the performance record of Trustees while rating issues. Further, rating agencies may give special consideration while rating, to specific covenants that the issuers may voluntarily undertake to impose discipline on their finances.
7. There are reports that some banks have on occasions shown a distinct preference for investment in rated paper, often through private placements, over normal credit operations. While diversified instruments for credit delivery are desirable for efficiency, excessive reliance on ratings without adequate internal assessment is inadvisable. Furthermore, there are reports that some banks tend to avoid the rigour of process of credit appraisal by preferring to invest in rated paper. Such an avoidance of normal advances impacts adversely on credit availability to small and medium enterprises. Both as a prudential requirement and as sound business strategy, banks will be well advised to treat the ratings with a mix of respect and caution that they deserve.
8. The role of ratings in CP is also under review, by an internal group set up to improve the regulatory framework for CP in money market. On the basis of the Report of the Internal Group, the RBI expects to release shortly draft proposals for review of regulatory framework for Commercial Paper, which would include a review of prescription regarding ratings.
9. The policy makers, and this includes the regulators, are facing some dilemmas in regard to ratings. They are compelled by the circumstances to place increasingly greater reliance on CRAs judgements, and at the same time, they are inclined to increasingly question the CRAs’ judgements. Operationally, the regulators face the problem of specifying or recognising a particular CRA’s rating since ratings do differ among CRAs. Prescribing two ratings rather than one for regulatory purposes may not fully address this issue. One approach is that of the United Nations National Association of Insurance Commissioners which itself resolves differences of opinion among CRAs. The Hong Kong Monetary Authority resolved the issue by comparing ratings of sample of issuers and subtracting the appropriate number of notches from ratings provided by agencies that on average, grade higher. There have been proposals for rating the CRAs by regulators but have not been found to be acceptable so far. It is necessary for regulators to explore ways of resolving the differences among CRAs while using them for regulatory purposes. Furthermore, when only higher rating is insisted upon by the regulator, there is a take it or leave it implication, resulting in pressures to obtain only higher rating. In such circumstances, the financial markets would effectively shut out several investors and issuers willing to participate in ratings with higher default probability. There is thus, another dilemma for regulators here. The RBI will examine these issues while considering the use of ratings in the regulatory framework and would welcome suggestions.
10. More generally, RBI has been in the forefront in advocating caution in use of ratings in the New Accord on Banking Supervision. There has been a widespread appreciation and indeed acceptance of our cautionary stand on the subject – especially the consequences of a short-end bias in rating. Such a bias has a potential to encourage short-term liabilities and in the process imparts some instability to financial systems inherent in excessive short-term liabilities. In terms of operational detail also, some observers feel that several notches of ratings are clubbed together in the New Accord.
11. Caution is also warranted when comparing ratings by international CRAs, which are what may be called global ratings and national or domestic ratings. The global ratings take into account capacity in terms of foreign currency obligations while national ratings would accord top notch rating for sovereign almost automatically. Moreover, default probability embedded in a particular rating – say in India by an Indian CRA will reflect the domestic standards, while for the same rating, a global CRA would capture a different level of default probability.
12. In general, ratings are very useful for investors, issuers and regulators but they need to be used carefully. I would conclude, by saying that credit ratings are like lampposts, which are meant to provide illumination for all, though a drunk could use them for support !

Thank you,