Credit Rating Agencies: Meeting the Needs of the Market? Researchers: Angus Duff Sandra Einig

Transcription

1 Credit Rating Agencies: Meeting the Needs of the Market? Researchers: Angus Duff Sandra Einig

2

3 CREDIT RATING AGENCIES: MEETING THE NEEDS OF THE MARKET? by Angus Duff University of Paisley Sandra Einig Oxford Brookes University Published by The Institute of Chartered Accountants of Scotland CA House, 21 Haymarket Yards Edinburgh EH12 5BH

4 First Published 2007 The Institute of Chartered Accountants of Scotland 2007 ISBN EAN This book is published for the Research Committee of The Institute of Chartered Accountants of Scotland. The views expressed in this report are those of the authors and do not necessarily represent the views of the Council of the Institute or the Research Committee. No responsibility for loss occasioned to any person acting or refraining from action as a result of any material in this publication can be accepted by the author or publisher. All rights reserved. No part of this publication may be reproduced, stored in a retrieval system, or transmitted, in any form or by any means, electronic, mechanical, photocopy, recording or otherwise, without prior permission of the publisher. Printed and bound in Great Britain by T. J. International Ltd.

5 Contents Foreword... i Acknowledgements... iii Executive Summary... v Abbreviations... xi 1. Introduction... 1 Background... 1 Purpose of the study and the research approach Credit Rating Agencies A brief history of the ratings industry Ratings processes... 1 The demand for ratings services Financial performance of credit rating agencies Demands for self regulation of credit rating agencies Barriers to entry Summary Literature Review Market participants attitudes to credit rating agencies Audit theory and audit quality Summary Interviews with Key Market Participants and the Development of the Ratings Quality Survey... Interviews... Questionnaire development... Technical features... Service features... Summary

6 Contents 5. Conduct of Survey Financial managers Other interested parties Investors Summary Results of the Questionnaire Survey: All groups Characteristics Individual items Summary Comparison Between Groups Characteristics Items Summary Conclusions Main findings and comments Policy implications Further research References... Appendix... About the Authors

7 Foreword The recent turmoil in the international financial markets following the subprime mortgage crisis in the US has raised questions about the role of credit rating agencies in this crisis and as to whether credit rating agencies are meeting the needs of the market and if there is a need for increased regulation. This timely publication looks at the views of market participants before the crisis and provides an insight into the rapid development and practices of the agencies. The authors recognise the lack of research in this topical area and identify five areas for future research: to consider the relationship between the credit rating agency and the issuer; to identify whether an expectations gap exists; to gather the views of market participants in other countries, particularly in the US; to better understand the processes and methodology used by the agencies; and to see how market participants views have changed as a result of the turmoil encountered in the credit markets of This report commences with background information on the industry, including information on: the structure of the market; the ratings process; the perceived global domination of the market by two main players; the criticisms the agencies have faced since the late 1990s; and international calls for greater supervision and accountability of the agencies. Against this background commentators have questioned whether credit rating agencies are providing an adequate service and quality ratings. This research attempts to resolve some of these issues by gathering views of participants and interested parties to identify the components of ratings quality and then comparing the views of three separate groups of market participants in the UK: corporate treasurers; investors; and other interested parties. The other interested parties group was made

8 ii Foreword up of finance academics, finance professionals, consultants, journalists and bankers. The report then analyses the findings from the interviews and questionnaires. The interviews were used to help identify 1 characteristics of rating agencies ranging from technical features to service features which were used to design the survey questionnaire, which featured 71 questionnaire items. In the survey reputation, trust and values were ranked as the most important characteristic, with transparency and timeliness also ranked highly. The highest ranking items were focused on integrity, ethical standards and credibility as well as more competency based items such as the accuracy of ratings and qualification of staff. User groups did not place any importance on other services offered by the agencies. Users were confident in the independence of credit rating agencies they dealt with. Finally, the report concludes on the findings of the study and policy implications for the industry. The policy implications are that: there is no evidence to suggest a need for formal regulation of the agencies, as existing codes of conduct are considered to be adequate; the codes, and compliance with them, should be monitored closely; the agencies themselves need to maintain and improve their commitment to quality and improve their communication; agencies ancillary services should be formally separated from their ratings work; and any unsolicited ratings should be clearly identified as such. This project was funded by the Scottish Accountancy Trust for Education and Research (SATER). The Research Committee of The Institute of Chartered Accountants of Scotland (ICAS) has also been happy to support this project. The Committee recognises that the views expressed do not necessarily represent those of ICAS itself, but hopes that the project will contribute to current debate on credit rating agencies and their role in the financial markets. David Spence Convener of Research Committee October 2007

9 Acknowledgements The monograph benefits from comments of two anonymous reviewers, together with John Grout of the Association of Corporate Treasurers and Christine Helliar of The Institute of Chartered Accountants of Scotland. We would like to thank the 16 individuals who agreed to be interviewed and shared their experience and perspectives of the ratings industry. We are indebted to those individuals who agreed to pilot our questionnaire and who made helpful suggestions and comments. Most significantly, we are grateful to those respondents who found time to complete our questionnaire. Finally, the Research Committee and the researchers are grateful for the financial support of the Trustees of the Scottish Accountancy Trust for Education and Research without which the research would not have been possible.

10

11 Executive summary Background In the past decade, interest in the activities of credit rating agencies (CRAs) has grown significantly. This reflects: (i) the global growth of ratings services, driven by new debt issues and structured finance products such as credit derivatives; (ii) criticisms of the CRA industry s seeming inability to forecast the Asian crisis of 1997, and the wellpublicised corporate scandals of the early part of this century, including the recent subprime mortgage crisis creating turmoil in financial markets; and (iii) the perceived duopolistic nature of competition in the CRA industry, with domination by two CRAs: Moody s, and Standard & Poor s (S&P). In response to these issues, commentators have suggested greater self regulation of the industry, with the United States debating the need for formal regulation. However, only a small amount of research exists which considers what constitutes quality in the provision of ratings services. This is in contrast to studies of the audit and assurance industry, where an accumulation of literature has developed the concept of audit quality. Purpose of study and research approach The purpose of this investigation is to apply the audit quality literature to conceptualise those desirable qualities of CRAs. These theoretical qualities were refined by undertaking 16 interviews with issuers of debt who use ratings services, debt investors, and a range of other interested parties. The interviews and audit quality and related literatures were used to develop a 71 item questionnaire relating

12 vi executive summary to potential CRA qualities. The 71 items relate to 1 underlying characteristics of ratings quality, labelled: co-operation; expertise; independence; investor orientation; internal processes; issuer orientation; methodology; reputation; responsiveness; timeliness; transparency; trust; values; and service portfolio. This investigation was conducted in the period 2005 to The questionnaire was sent to 2,00 interested parties representing three primary groups using CRA services: financial managers in UK corporations that are issuers of debt; debt investors (bond holders); and other interested parties (accounting and finance academics, accountants in corporate finance functions, corporate and investment bankers, financial journalists, and treasury consultants). Key findings Relative importance of ratings quality characteristics There was a broad consensus across the three groups regarding the usefulness of the 1 ratings quality characteristics. These characteristics were generally rated in the following order: reputation, trust, and values, followed by the relatively clustered characteristics of transparency, timeliness, expertise, investor orientation, methodology, co-operation, independence, issuer orientation, internal process, and responsiveness, with service portfolio ranking last and well-behind the other characteristics. Relative importance of 71 individual questionnaire items The items rated most highly by all stakeholder groups tended to be related to the integrity of the CRA, their ethical standards and credibility. Rated almost as highly were items which directly related

13 executive summary vii to the competence of CRA outputs such as accuracy of ratings and associated reports, and the education and qualifications of staff. The lowest ranking items related to the ability of the CRA to provide specialised ancillary services. Absolute importance of 1 ratings quality characteristics In general, the financial manager group rated the 1 characteristics as more important than the other interested parties, with debt investors located between these two groups. This result perhaps reflects that the primary relationship is between the financial manager of the issuer and the CRA, with debt investors and other interested parties somewhat detached from this process. Differences between the three groups When making inter-group comparisons between the three groups, the most polarised differences exist between issuers and debt investors. Other interested parties views fall between the views of issuers financial managers and debt investors. As expected, financial managers rate issuer orientation higher than debt investors, and debt investors rate investor orientation higher than issuers. Financial managers also rate the characteristics of trust and methodology higher than debt investors. These characteristics reflect the one-to-one relationship that financial managers usually have, in their role as issuers, with the CRA. Debt investors and financial managers have differing views of what timely means in terms of rating updates. Debt investors would like to see a rating updated as soon as possible. Issuers prefer stability in their rating, as volatility in ratings may breach financial covenants in bank borrowing agreements. When financial managers are compared with other interested parties, differences relate to four characteristics: issuer orientation;

14 viii executive summary responsiveness; trust; and timeliness. Similar to the financial managerinvestor comparison, issuer orientation, responsiveness and trust relate to the close relationship between issuers of rated debt and the CRA. Also, timeliness means different things to financial managers and other interested parties, with financial managers preferring stability in ratings and other interested parties preferring ratings to reflect current events. Finally the differences between debt investors and other interested parties are in items that are investor-specific or issuer-specific. That is, those items that focus on the service and attention that the particular group receives from CRAs. Financial manager issues Sixty one per cent of responding financial managers issue debt in public markets. Two-thirds of financial managers maintain a relationship with a CRA, either for debt issuance or investment purposes. Moody s and/or S&P are used by 95 per cent of financial managers maintaining a relationship with a CRA. However, the position is far from a duopoly, with 52 per cent of respondents indicating that they use Fitch or another CRA, either with Moody s and/or S&P, or as their sole CRA. Multiple CRA relationships are commonplace: 8 per cent of respondents use two CRAs and 33 per cent use three CRAs. Debt investor issues Only one-third of responding debt investors indicate that it is policy to require a rating before making a debt-related internal investment. Although a majority indicate that a rating is important, 5 per cent indicate it was just one input of many.

15 executive summary ix Other interested parties issues Two-thirds of other interested parties indicate that they make regular use of ratings produced by CRAs. This validates expectations that a wide range of stakeholders make use of ratings information. The most common uses are: research, teaching and training (18 per cent); advising clients (17 per cent); assessing the creditworthiness of clients/suppliers (17 per cent); and for investment decisions (1 per cent). Discussion and policy implications Background research conducted for this project identifies the strong growth in the ratings industry enjoyed by just a few CRAs. This growth, combined with recent corporate scandals of a global nature not foreseen by CRAs and the recent subprime market crisis, has stimulated demands for greater scrutiny and regulation of CRAs work. This project explores relevant background issues to the ratings industry, and elicits what constitutes quality in CRAs work. The questionnaire survey identifies 1 quality characteristics desirable in CRAs. A great deal of consensus exists among the three groups suggesting that each of these stakeholders are looking for similar qualities in CRAs. As the most important characteristics relate to the reputation, credibility and integrity of the CRA, it is essential that CRAs do not allow their standards to slip in the light of their expansion. Behind those characteristics relating to the standing of the CRA in the market are those aspects that reflect: technical qualities of the CRAs service provision; the timeliness of ratings upgrades/downgrades; the transparency of CRA decision-making; and their expertise and methodologies. CRAs interviewed went to great lengths to explain that their methodologies were disseminated via their websites and available to all parties. However, issues relating to communication were apparent in many interviews. CRAs could therefore benefit from making additional

16 x executive summary efforts to explain their methodologies to stakeholder groups. These could include: presentations to issuers about the conditions needed for ratings upgrades, and what might precipitate a downgrade; presentations to investor groups about their distinctive methodologies; and open doors sessions to other members of the financial community to create a greater understanding of their work. Surprisingly, those items relating to the characteristic of independence did not score highly in the surveys. However, it would be dangerous to interpret this as meaning that independence is not an important characteristic of a CRA. Interviewees all believed that CRAs were independent, and that it was one aspect of their work that they were entirely confident about. The most appropriate interpretation of these findings is that, independence is a sine qua non of CRA work. This is one finding that CRAs and all market participants must take great comfort from. The characteristic of CRAs ranked least important was their service portfolio. The provision of added-value ancillary services by CRAs was not seen as particularly valuable by any of the user groups. The fallout from the subprime crisis has reiterated calls to separate ratings and consulting arms. CRAs either need to better communicate the value of these services, or stick to their core business of providing ratings information. A finding of this research is that a large number of stakeholder groups make use of ratings information for a variety of purposes. However, it is apparent that many individuals do not have a significant understanding of what the ratings industry does or of the ratings process. Issues of communication have already been addressed. However, one avenue for CRAs to explore would be to educate potential members of the financial community, and develop stronger links with universities and professional bodies in the accounting and financial services sector. As well as improving an understanding of the industry, courses or programmes focusing on the analysis of debt securities and structured finance could increase the supply of junior analysts.

17 Abbreviations ABBREVIATIONS USED IN REPORT ACT AFP AFTE AMF CEBS CESR CRA CRARA DBRS EC EU FEE Fitch IGTA IOSCO Moody s NRSRO NYSE OIP SEC S&P UK US Association of Corporate Treasurers Association of Finance Professionals (US) Association Francaise des Tresories D Enterprises (France) Authorité Des Marchés Financiers (French Securities Regulator) Committee of European Banking Supervisors Committee of European Securities Regulators Credit Rating Agency Credit Rating Agency Reform Act Dominion Bond Rating Services European Commission European Union Federation of European Accounting Experts Fitch Inc., a division of Fimalac International Group of Treasury Associations International Organization of Securities Commissions Moody s Investors Services Inc. Nationally Recognized Statistical Rating Organization New York Stock Exchange Other interested party Securities and Exchange Commission Standard and Poor s, a division of The McGraw-Hill Companies Inc. United Kingdom United States of America

18

19 1introduction Credit rating agencies play a vital role in global securities and banking markets. It is essential, therefore, that they consistently provide ratings which are independent, objective, and of the highest possible quality. (European Commission, 2006 p.c59/2) This chapter outlines the background to the study, the conceptualisation of ratings quality, and the research approach together with the structure of the report. Background The role of credit rating agencies Credit rating agencies (CRAs), usually at the request of the issuers of a debt security, provide a rating which provides an indication of the probability of investors receiving their money back in accordance with the terms under which they invested. Ratings can be applied to a wide range of debt securities including long-term debt, medium-term notes, commercial paper, bank loans and preference shares. The CRA assesses the likelihood of borrowers (debt issuers) defaulting on their repayments. A rating reflects the credit worthiness of the issuer, together with any credit enhancement feature attached to the security, such as a guarantor, a letter of credit provider, or a bond insurer. The assessment is delivered in a letter-based rating, and provides issuers, investors, and other stakeholder groups with an objective criterion to assess the credit risks associated with a particular debt. Therefore, ratings influence the interest rate issuers will pay, and their ultimate cost of capital.

20 Credit rating agencies: meeting the needs of the market? The ratings industry is dominated by two agencies: Moody s Investors Service Inc. (Moody s), and Standard & Poor s (S&P), a division of The McGraw-Hill Companies Inc. Fitch Inc. (Fitch), a division of Fimalac, S.A., is a third CRA providing global coverage, but as yet does not have the same market share as Moody s and S&P. Each agency uses their own scale and nomenclature see Table 1.1. Panel A of this table summarises the international long-term ratings scales of the three major CRAs. CRAs also assign short-term credit ratings to debt obligations that have original maturities of one year or less, for example, commercial paper. Panel B summarises the international short-term scales of the three major CRAs. Investment grade ratings signal a low probability of default and high likelihood of repayment. At the other end of the credit spectrum, speculative (junk) ratings indicate a high probability of default or that default has already occurred. Table 1.1 Long-term and short-term credit ratings of the three largest CRAs (adapted from Voizey, 2006) Panel A: International long-term credit ratings Fitch Moody s S&P Investment grade Highest quality/ best quality/ extremely strong AAA Aaa AAA Very high quality/high quality/very strong AA Aa AA High quality/upper medium grade/ strong A A A Good quality/medium grade/adequate BBB Baa BBB Non-investment grade Speculative/lower medium grade/speculative less vulnerable BB Ba BB Highly speculative/low grade/more vulnerable B B B Poor quality/currently vulnerable CCC Caa CCC High default risk/highly speculative/currently highly vulnerable CC Ca CC High default risk/extremely poor/imminent default C C C In default D C D Note: Fitch and S&P append rating with + or - to denote relative status within major rating categories. Moody s append ratings with 1, 2, or 3 to denote relative status.

21 introduction Table 1.1 Long-term and short-term credit ratings of the three largest CRAs (adapted from Voizey, 2006) (Continued) Panel B: International short-term credit ratings Fitch Moody s S&P Highest/superior/strong F1+, F1 P1 A1 Good/strong/satisfactory F2 P2 A2 Fair/acceptable/adequate F3 P3 A3 Speculative/not prime/speculative B Not prime B High default risk/ - /vulnerable C - C Default/ - /default D - D The structure of the CRA market A recent Basel Committee survey of CRAs indicates there are more than 130 agencies worldwide (Basel Committee, 2000). These CRAs vary considerably in terms of their size, the industries they cover, their regional focus and methodologies used. The three largest CRAs: Moody s, S&P, and Fitch operate on an international basis and boast global coverage for their ratings. Each of these CRAs offers ratings for sovereign and corporate borrowers, and specific issues of fixed-income securities. The three largest CRAs derive the majority of their revenue in the form of issuer fees for ratings. These ratings are then made publicly available, with summary information on the rationale for their decision available free-of-charge via their websites. Additional income is earned by fees charged to subscribers who are allowed detailed access to ratings information, and by the provision of ancillary business services. Smaller CRAs operate by offering a service to specific industries (e.g. insurance) or operating in developing countries by analysing local gradations of issuer credit risk that would otherwise be overwhelmed by country risk, that all issuers in that market face (IOSCO, 2003). In contrast to the larger agencies, smaller CRAs derive most of their income from subscriber fees.

22 Credit rating agencies: meeting the needs of the market? CRAs have enjoyed considerable growth in the past two decades. For example, in the 21st century revenues for Moody s, S&P, and Fitch have grown by 17.6 per cent per annum see chapter two and figure 2.1. This growth reflects two related processes (Sinclair, 2005). First, disintermediation, where bonds and notes sold on capital markets have replaced bank loans as a source of debt finance. Second, securitisation where mortgages, credit card receivables and bank loans are transformed into securities that can be traded in capital markets. These processes provide companies, institutions, and non-profit organisations with the ability to access debt markets, rather than borrowing from a traditional relationship bank. Therefore, issuance of debt has grown dramatically over the past two decades. The growth in the holding of bonds relative to loans by institutional investors in the UK, is shown in Figure 1.1. Figure 1.1 Trends in financial assets held by institutional investors in the UK in US Dollars (Millions) (OECD, 200) Other Shares Loans Bonds

23 introduction Recent criticisms of credit rating agencies However, despite the utility of information produced by CRAs, their activities have drawn criticism from a range of interested parties including regulators, politicians, issuers, investors, and professional bodies such as corporate treasurers. A significant source of criticism of CRAs has been their inability to predict large-scale failures such as the 1997 Asian financial crisis or well-publicised corporate failures of Parmalat, Enron, or WorldCom. Each of the three leading CRAs Moody s, Standard & Poor s, and Fitch, provided Enron with investment grade ratings until four days prior to its collapse. Such events have led commentators to question whether CRAs are providing adequate service to market participants. Despite access to insider information, there is a feeling that CRAs did not question Enron suffciently: The Committee staff has concluded that the credit rating agencies approach to Enron fell short of what the public had a right to expect It is diffcult not to wonder whether the agencies practical immunity to lawsuits and non-existent regulatory oversight is a major problem. (US Congress Committee, 2002 p.116) How reasonable such criticisms are, given certain issuers motivation to mislead CRAs, is open to question. Other criticisms of CRAs made by various commentators at the start of the 21st century include: the lack of competition within the CRA market, which is dominated by two CRAs, reflects high barriers to entry within the CRA market;

24 Credit rating agencies: meeting the needs of the market? the excessive profitability of CRAs, reflects the lack of competition within the CRA market and lack of leverage issuers possess to negotiate tariffs; and the lack of transparency in the ratings process and the use of unsolicited ratings. The next section identifies the purpose of this report and the research approach adopted. Purpose of the study and the research approach Research issues The strong growth in the market for ratings, with questions of the ability of CRAs to predict corporate failures, together with the concerns of regulators, politicians, issuers, investors, professional associations and regulators sets the scene for a research project of interest to a wide audience. A relative paucity of research considers how market participants view the role of CRAs and the concept of ratings quality. Perspectives from other areas of related academic work are therefore examined to provide a framework for the investigation. In particular, the project focuses on the audit quality and service quality literatures which have some potential to inform contemporary thinking on CRAs. CRAs are similar in a number of ways to auditors as suppliers of audit services. Similar to auditors, CRAs operate in a unique economic market in which the CRA is a central participant. A significant source of demand for ratings comes from third parties (especially fixed income investors) who do not directly pay for the rating service. The debt issuer (like the audit client) is often a forced participant in the market, required by investors to engage the services of a CRA to obtain a rating of the credit quality of its securities.

25 introduction The focus of the research project is ratings quality and the principal aims of the project are to: elicit and compare interested parties views regarding the determinants of the quality of ratings provided by credit rating agencies; develop a holistic conceptualisation of ratings quality, borrowing elements from audit theory; and compare interested parties attitudes regarding a broad set of items relating to ratings quality, derived from the semi-structured interviews and the audit quality and service quality literatures. A major purpose of this investigation is that it allows the attitudes of the main interest groups in credit ratings to be systematically compared, allowing the identification of major points of agreement and disagreement to be identified. Evidence of this nature is potentially valuable to regulators, such as the International Organization of Securities Commissions (IOSCO), Committee of European Securities Regulators (CESR), Committee of European Banking Supervisors (CEBS), and the Securities and Exchange Commission (SEC), whose aim it is to encourage best practice and resolve potential conflicts of interest between groups. Questions asked The investigation is informed by two sources of data. First, from the 16 semi-structured interviews undertaken with issuers, debt investors and other interested parties, to understand the determinants of ratings quality and to ascertain the relationship CRAs have with financial markets and associated market participants. Second, from a questionnaire survey developed to elicit the importance of specific characteristics of CRAs from a wide range of market participants. These characteristics were derived from a detailed review of the academic and professional literature

26 Credit rating agencies: meeting the needs of the market? relating to audit quality and service quality published up to the start of 2006, and informed by the findings of the 16 semi-structured interviews. In total, a comprehensive list of 71 items was created. For convenience these were grouped into 1 categories. Table 1.2 lists the main categories, together with a code and the number of items. Table 1.2 Categories of questions used in the questionnaire Category Code Description Reputation Values Expertise Methodology Timeliness Issuer orientation Investor Orientation Responsiveness Service portfolio Rep. Value Exp. Meth. Time. Issue. Invest. Resp. SP The reputation and standing of the CRA within the financial community How the CRA goes about doing business The sunk costs CRAs have incurred in producing ratings, their coverage of markets The confidence market participants have in the techniques CRAs use in their assessment of issuers The ability of ratings to accurately reflect events happening within the issuer s organisation the general economy. The ability of CRAs to offer a high level of service to the issuers The ability of CRAs to offer a high level of service to the investors Relationship issues between CRAs and individual issuers and investors The ability of CRAs to provide services other than ratings No. items Independence Indep. The detachment of CRAs from issuers 7 Internal process Proc. Processes CRAs employ to ensure ratings are of the highest quality 6 Transparency Trans. The willingness and ability of CRAs to communicate the ratings process 6 Trust Trust The level of trust between CRAs and issuers and investors 7 Co-operation Coop. The degree of communication between issuers and CRAs 6 Total

27 introduction The questionnaire consisted of five sections. For sections one to three, respondents were asked to base their responses on their general views of credit rating agencies, rather than their experiences with a single CRA. Specifically they were asked to think about the kind of rating agency that would produce ratings of the highest quality. A five-point scale, ranging from 5 (absolutely essential) to 1 (not at all essential) was used. Section one consisted of factors relating to the ideal CRA. The second section consisted of factors relating to CRA s lead analyst and other staff. Section three consisted of other relevant factors. Section four was an open-ended section that asked for free comments regarding the quality of credit ratings and general views on CRAs, and the ratings process. The final section asked questions related to the background and experience of the people completing the questionnaire, and their experience of using ratings information. Respondent groups and response rate Financial managers were UK corporate treasurers and finance directors drawn from the Association of Corporate Treasurer s (ACT) membership directory. Investors were UK institutional investors drawn from the Euromoney Institutional Investors Database. Other interested parties were a relatively heterogeneous group of finance academics, consultants, financial professionals connected with finance professional bodies, financial journalists, and corporate and investment bankers all from the UK. Of the 2,00 questionnaires distributed (with two follow up mailings), 08 responses were received. This represents a response rate of 18.2 per cent - see Table 1.. A more detailed description of the sampling procedures and sample composition is described in chapter five.

28 10 Credit rating agencies: meeting the needs of the market? Table 1.3 Respondent groups and response rates Group No in sample No. excluded from sample* No. of responses Response rate Corporate treasurers % Investors % Other interested participants % Total 2, % Note: * these represent questionnaires returned as the recipient no longer worked with the organisation. Structure of the report Chapter two provides some background to the environment in which CRAs operate. Specifically the chapter describes: the ratings process; identifies potential sources of demand for rating information; explains how the CRA market has become dominated by just a few players; reviews the recent financial performance of CRAs; and considers international demands for greater supervision and accountability of CRAs. Chapter three contains a literature review. It begins by a consideration of the recent professional literature that documents proposals for changes in the way CRAs conduct business. The next section then describes the audit theory literature which is used to gain an insight into the interaction between the CRAs and their stakeholders. Specifically this section considers: ways in which CRAs and auditors as suppliers of audit services are similar or dissimilar; an overview of audit theory and definition of audit quality; and the economics of audit quality. The chapter considers how each of these literatures extend to the ratings industry. Finally, empirical research considering different

29 introduction 11 stakeholder groups of audit quality is reviewed as this has the potential to inform the present study. Chapters four, five, six, and seven present the results of the present research. Chapter four presents and discusses the findings from the 16 interviews. The list of characteristics used in the questionnaire is presented along with an identification of their source. Chapter five explains how the questionnaire survey was conducted and reports the results of specific background questions asked of the three groups. Chapter six reports the findings of the questionnaire survey for all groups. Chapter seven compares findings between the groups for characteristics and individual items. Chapter eight summarises and concludes. Implications of the research for policy makers are discussed along with suggestions for future research.

30

31 2credit rating agencies This chapter provides an introduction to the work of CRAs. Specifically, five issues are addressed: how a rating of a security is conducted; the demand for ratings services; the growth in revenue enjoyed by major CRAs in the 21st century, and their related financial performance; demands for greater regulation, self regulation and oversight of the activities of CRAs from both US and international viewpoints; and finally, potential barriers to entry in the CRA market. A brief history of the ratings industry Historically, CRAs developed from mercantile credit agencies that supported business and trade creditors. The first of these agencies was The Mercantile Agency, established in 181, followed by R.G. Dun and Company in 1859, which became Dun and Bradstreet, Inc., the owner of Moody s until 2001 (for a full history, see Sinclair, 2005). The first individual to focus on debt issuers was Henry Poor in his History of Railroads and Canals of the United States. Poor s Publishing Company later merged with the Standard Statistics Bureau (founded 1906) in 191. The only other suppliers of debt ratings information were Fitch Publishing Company, established in 192 and Duff and Phelps Credit Rating Co., which specialised in public utilities. Duff and Phelps merged with Fitch in Fitch had previously acquired some smaller CRAs that attempted to challenge Moody s and S&P, such as IBCA (in 1992) and Thomson BankWatch (in 2000). To summarise, historically the ratings industry has been dominated by a few suppliers. Historical analyses suggest this has been due more to a common perception that the ratings industry:

32 1 Credit rating agencies: meeting the needs of the market? was simply too small to justify the start-up costs incident to covering the large universe of publicly held bonds. (Coffee, 2006, p.29) Furthermore, when a threat appears to the oligopoly the industry leaders enjoy, the smaller competitor is subsumed. Ratings processes The processes used by individual CRAs vary considerably. It is important to distinguish between two types of rating. Solicited ratings occur where the issuer has invited the CRA to rate it, usually for a fee, or has at least participated in the rating. Unsolicited ratings are undertaken by the CRA without reference to the company s management. Such unsolicited ratings are usually carried out by CRAs to improve market coverage of their ratings, or as a marketing device to encourage the company to commission a rating. At the start of the rating, a lead analyst is appointed by the CRA. The lead analyst requests information from the issuer and collects other available sources of information to undertake an assessment of the issuer s industry and economic environment. The lead analyst will meet with the company s management and visit its offces and production facilities. A rating is usually based on an assessment of quantitative and qualitative indicators, which are reported to a ratings committee. Individual CRAs have their own methodologies, which they disclose to a greater or lesser degree to the public via their websites. The ratings committee usually consists of the lead analyst, senior managers at the CRA and junior analysts who work on the rating. The final decision is made by a ratings committee, not by a sole individual. A report is then prepared with a draft recommendation. Issuers are allowed to comment on the report to ensure the information collated by the lead analyst is correct. If the CRA has relied on incorrect information or is able to present new evidence, the CRA may reconsider its decision.

33 Credit rating agencies 1 The demand for ratings services Issuers seeking to market a debt issue commission the services of a CRA to grade their securities. This is undertaken primarily for five reasons, described below: To define investment eligibility, as many investment organisations will have policies which specifically prohibit investment in noninvestment grade bonds. Therefore, the value of ratings has become institutionalised into organisational practice. Without an appropriate rating, investor appetite for an issue will be thin. Consequently, some commentators (Langohr, 2006) suggest CRAs provide a gate-keeping function to enter global capital markets, where the rating provides the ticket for entry. To reduce the information asymmetry effects (Myers and Majluf, 198) between the issuer and investors. In undertaking an independent review of the probability of default of a debt issue, the CRA is able to provide objective and independent information to the market about the issuer s future prospects. Furthermore, the CRA has access to the company s management and is able to access information about the organisation s future strategy and prospects not available to the market. Therefore, the information gap between issuers and investors is reduced. The effect of reducing this information asymmetry is to reduce the risk premium associated with a debt issue, lowering the margin payable on the issue and reducing the organisation s overall cost of capital. CRAs are able to provide an advisory role to management, offering feedback about the long-term effects of their approach to risk. CRAs can advise management of the effect of changes in strategy to their perceived risk profile, and the effect such a change would have on their credit rating, and consequently cost of capital.

34 1 Credit rating agencies: meeting the needs of the market? Ratings are used by regulators, for example, the Committee of European Banking Regulators in the European Union (EU) and the Securities and Exchange Commission in the United States (US) to assess capital adequacy and permitted investments for regulatory purposes. As well as providing information upon which investors can base investment decisions, ratings are used to evaluate trading partners, financial counterparties, for example, in financial instruments such as swaps, and business partners, for example, where an organisation enters into a joint venture agreement or strategic alliance. For internal treasury purposes, such as assessing counterparty risk. Typically a corporate will have an internal policy which will set absolute and relative limits to the amount that can be invested with any one counterparty. This limit may vary according to the nature of counterparty, with higher limits applying to counterparties with higher credit ratings. Financial performance of credit rating agencies Having considered the market structure of the ratings business, it is useful to understand what effect this may have on the financial performance of the CRAs themselves. The leading CRAs have demonstrated significant growth in revenue and on the basis of publicly available information appear to be highly profitable. The revenues of S&P, Moody s and Fitch grew at an average of 17.6 per cent per annum from 2001 to 2005, reaching $,826 million see Figure 2.1.

35 Credit rating agencies 1 Figure 2.1 Revenue of three major CRAs, Revenue $millions Fitch Moody's S&P Moody s represents the only CRA quoted on a stock exchange (New York Stock Exchange, NYSE). S&P and Fitch are owned by McGraw- Hill and Fimalac respectively, and therefore the performance of these ratings businesses is harder to observe. It is therefore only possible to examine Moody s in any detail. Table 2.1 identifies some of the key financial performance statistics of Moody s relative to those companies also quoted on NYSE. This demonstrates that Moody s performance is superior to the average of its peers quoted on NYSE. Table 2.1 Financial performance of Moody s versus NYSE, source Thomson Disclosure Moody s NYSE mean P.E. ratio (2005) Net income margin % (2005) It is diffcult to assess how comparable Moody s financial performance is to Fitch and S&P. Moody s exhibit the strongest average growth in revenue over the five-year period (21.5 per cent p.a.) compared to S&P (16. per cent p.a., source: McGraw-Hill Inc annual report)

36 1 Credit rating agencies: meeting the needs of the market? and Fitch s (13.8 per cent p.a., source: Fimalac 2006 annual report). In conclusion, the demand for ratings has generated significant sales growth for the three leading agencies, and in particular the structured products market, for example, credit derivatives. In the first quarter of 2007, structured products accounted for 3 per cent of Moody s Investor Services total revenue, an increase of per cent on the first quarter of 2006 (source: Moody s 2007 first quarter results). However, the market concentration enjoyed by the three largest CRAs operating on a global scale, together with their strong financial performance led to criticisms that the ratings market is an oligopoly and anti-competitive. It is a matter of speculation how sustainable such growth is in the longterm. The production of ratings is a labour-intensive business requiring the employment of new analysts. Therefore, the ratio of experienced ratings staff to new hires will decrease. It is plausible that, without the facilities in place to rapidly develop new analysts, the quality of ratings information will fall, which in turn will reduce an important barrier to entry, namely reputation, to new participants. Demands for self regulation of credit rating agencies Concerns about the work and market structure of CRAs have led to a variety of responses from regulators, governments and professional bodies on an international basis. This section is structured as follows: first, the concept of the SEC s Nationally Recognized Statistical Rating Organisations (NRSRO) is discussed. Second, recent debates in the US are summarised. Third, international perspectives are discussed including the International Organization of Securities Commissions (IOSCO) Code of Conduct Fundamentals for credit rating agencies, and an exposure draft developed by corporate treasurers in the UK, US and France. The third subsection, describes the European response to the work of CRAs including that of regulators, the European Commission and the European Parliament.

37 Credit rating agencies 1 Introduction and recognition of the work of CRAs A degree of regulation has existed in the ratings industry since 1975 with the establishment of Nationally Recognised Statistical Rating Agencies (NRSRO) by the SEC. At the time, the SEC had no intention of becoming the de facto regulator of CRAs (Chambers, 2006). NRSRO status was originally created by the SEC to identify those agencies that it could rely on to distinguish among grades of creditworthiness in various regulations under the federal securities laws, that is, for its own internal purposes. Specifically the SEC was motivated to formulate rules on net capital requirements for broker-dealers. To allow dealers investing in highly rated securities to hold less capital the SEC needed to designate those CRAs that were competent to judge the quality of securities in which the brokers invested. Since then the role of the NRSROs has expanded; the SEC has accelerated issuance approval for issuers that are regular users of capital and have a high enough rating with an NRSRO. At present there are five NRSROs: Moody s, S&P, Fitch, Dominion Bond Rating Service Limited (DBRS), and A.M. Best. Although the SEC created the NRSRO label specifically for use in its own jurisdiction, NRSROs ratings are now used as benchmarks in federal and state legislation, financial and other regulators, foreign regulatory schemes, and private financial contracts. As the financial community began to place greater reliance on CRAs ratings, the NRSRO concept found greater currency. Likewise, although NRSRO designation is confined to the US, the global nature of ratings has resulted in the designation taking on the role of an accreditation standard worldwide. The SEC undertook a significant investigation into the role of CRAs, as described in the following section. As part of this exercise, the SEC conducted a consultation into the future of NRSRO designation. Arguments against NRSRO status are:

38 0 Credit rating agencies: meeting the needs of the market? that the NRSRO designation acts as a barrier to entry into the credit rating business (see discussion of Credit Rating Agency Reform Act in the following section); the SEC s lack of explicit regulatory authority over NRSROs; and the SEC s view that it would be inappropriate to impose a more comprehensive regulatory framework on CRAs (SEC, 2005). Perspectives from the US In January 2003, the SEC produced a preliminary report concerning the relationship between CRAs and the US capital markets (SEC, 2003a). This preliminary report was in response to the provision of Sarbannes- Oxley Act 2002 and the US Congress s concerns about the role that CRAs may have played in the corporate failures at Enron and other corporations. Specifically, the preliminary report examined: the role of CRAs and their relation to the securities markets; impediments faced by CRAs in performing that role; measures to improve information flow to the market from CRAs; barriers to entry into the rating business; and conflicts of interest faced by CRAs (SEC, 2003a). To follow up the preliminary report the SEC issued a concept release considering the role and function of CRAs in June 2003 (SEC, 2003b). Specifically, the SEC (2003b) investigated issues concerning:

39 Credit rating agencies 1 The utility of the NRSRO designation: including alternatives to NRSRO, the appropriateness of the existing recognition criteria, the recognition process, and the evaluation of ratings quality. Examination and oversight of CRAs. Is more direct, ongoing oversight of CRAs required? Are there minimum standards or best practices to which CRAs should adhere? Should CRAs incorporate general standards of diligence in their analysis and the training and qualifications of CRA analysts? Should CRAs be required to (rather than recommended to) register as investment advisers? Conflicts of interest. Should CRAs implement procedures to manage potential conflicts of interest when CRAs develop ancillary fee-based business; do firewalls exist to separate this business from the ratings business? Are issuers unduly pressured to purchase ancillary advisory services by a rating analyst at a NRSRO? Alleged anti-competitive, abusive and unfair practices. For example, smaller CRAs have alleged that S&P and Moody s have attempted to squeeze them out of certain structured finance markets by a practice of notching, that is lowering ratings, or refusing to rate certain asset pools unless a substantial proportion of the assets are also rated by them. Similar concerns exist with unsolicited ratings where a CRA attempts to induce an issuer to pay for a rating it has not requested, for example, by sending a bill for an unsolicited rating. Information flow: should the ratings process be more transparent? How could the quality of information to users of credit ratings be improved? Should all ratings information be available to all market participants, regardless of subscription?