The effect of increased competition on the quality of credit ratings

FACULTEIT ECONOMIE EN BEDRIJFSWETENSCHAPPEN FACULTEIT RECHTSGELEERDHEID

The effect of increased competition on the quality of credit ratings De toegevoegde waarde van credit rating agencies bij de berekening van minimum kapitaalvereisten

Anne Goor S0203622

Masterproef aangeboden tot het behalen van de graad MASTER IN DE ECONOMIE, HET RECHT EN DE BEDRIJFSKUNDE Optie financieel recht en economie

Promotor: Prof. Dr. Paolo Casini Co-promotor: Prof.Dr. Veerle Colaert Academiejaar 2012-2013

Summary

This master thesis is a critical analysis of the functioning of credit rating agencies and the quality of their credit ratings. We start with a short introduction about the background and the evolution in the regulation of those agencies. Further the paper gives more insights in the advantages and disadvantages which are related to credit raters. The emergence of an oligopolistic rating market and the reasons for this existence are discussed afterwards and highlight the importance of our research question, which is whether the quality of credit ratings would increase in a situation with more competition. According to the reputational capital view, credit raters publish truthful ratings in order to receive a good reputation and hence to attract more clients. We argue that this view fails to explain market behavior in situations of increased competition because in that case the value of a good reputation declines. By means of an economical model we do research after the effect of the entrance of a third rating agency, named Fitch. We end our thesis with an introduction to some alternative models which could serve as a solution to the conflicts of interest related to credit rating agencies.

Acknowledgements

This way we would like to thank a few people who helped us with the accomplishment of our master thesis.

First of all we would like to thank our promotors professor Paolo Casini and professor Veerle Colaert who stimulated us in the subject and helped us find the right direction for our research question. By encouraging us to work out a plan and to do some beginning research in the second semester of the previous year, the workload in the second year became less heavy and more achievable.

We would also like to thank Werner Thiels and Lies Gevaert working at Record bank and Belfius respectively for the time they spend introducing the subject to us.

Professor De Croux deserves our extra attention for guiding us through Eviews and helping us setting up a regression model using panel data.

A lot of credit goes out to Gianni Pauwels, who spend an incredible amount of time helping us retrieve the data. This enabled us to do the regression tests with credit ratings, which are hard to retrieve.

Last but not least, we would like to thank everybody in our close environment. Our parents and sister, who motivated us in the difficult moments. And of course our friends who were there with some pep talk after a disappointing day

ACKNOWLEDGEMENTS| III

Contents

Acknowledgements ......................................................................................................... iii Contents ............................................................................................................................ iv List of abbreviations ........................................................................................................ iv Introduction ....................................................................................................................... 6 1

Literature review ..................................................................................................... 9 1.1

1.2

1.3

2

EMPERICAL STUDY.............................................................................................. 25 2.1 2.2 2.3

2.4 3

Model setup .................................................................................................. 25 Data .............................................................................................................. 28 Results .......................................................................................................... 30 2.3.1 Test 1: Regression with firm credit ratings ......................................... 30 2.3.2 Test 2: Correlation between ratings and bond yields ......................... 35 2.3.3 Test 3: Predicting default.................................................................... 37 2.3.4 Interpretation of the results................................................................. 41 Alternative interpretation ............................................................................... 42

ALTERNATIVE MODELS ...................................................................................... 44 3.1 3.2 3.3

4

History of credit rating agencies ..................................................................... 9 1.1.1 concept ................................................................................................. 9 1.1.2 Evolution of the regulation on CRA's.................................................. 10 The functioning of credit rating agencies ...................................................... 12 1.2.1 Information asymmetry problem ......................................................... 14 1.2.2 Coordination function ......................................................................... 15 1.2.3 Advantages & Disadvantages of the use of CRA's ............................ 16 Oligopolistic market ...................................................................................... 18 1.3.1 Entrance barrier .................................................................................. 19 1.3.2 Entrance to ratings markets ............................................................... 20

Investor pays model ..................................................................................... 44 Government pays model .............................................................................. 46 Alternatives ................................................................................................... 46

Conclusion ............................................................................................................ 49

Bibliography .................................................................................................................... 51 Attachments .................................................................................................................... 58

CONTENTS| IV

List of abbreviations

SEC

= Securities and Exchange Commission

NRSRO

= Nationally Recognized Statistical Ratings Organizations

CRA

= Credit Rating Agency

IOSCO

= International Organization of Securities Commissions

CRD

= Capital Requirements Directive

BIS

= Bank for International Settlements

S&P

= Standard & Poor’s

CESR

= Committee of European Securities Regulators

ESMA

= Europese Autoriteit voor Effecten en Markten

BW

= Burgerlijk Wetboek

NAICS

= North American Industry Classification System

MSCI

= Morgan Stanley Capital International

FISD

= Fixed Income Securities Database

LIST OF ABBREVIATIONS| IV

Introduction

In our master thesis we focused on the role external credit rating agencies play in the global financial system and the effect on those credit ratings when there is an increase of competition. A credit rating is defined as “an opinion regarding the creditworthiness of an entity, a debt or financial obligation, debt security, preferred share or other financial instrument, or of an issuer of such a debt or financial obligation, debt security, preferred share or other financial instrument, issued using an established and defined ranking system of rating categories” (Regulation 1060/2009)1. A credit raters’ main function is to produce valuable information about the creditworthiness of a corporation or a financial instrument and make this available to the public. Those credit rating agencies play a crucial role in the decision-making process of investors, most of them blindly make their decision based on a published credit rating. On the other hand are credit ratings an important factor in the regulation of financial institutions by determining the minimum capital requirement. When banks supply financial services and provide clients with the opportunity to invest their money with them, a maturity transformation will occur in which the bank will have to turn short term investments into long term loans. There is a potential problem when such credit takers are insufficient liquid to fulfill the retrievals of those loans, which will cause financial institutions to be unable to satisfy the retrievals of their creditors. This may lead to a domino effect which will make the financial system collapse. Therefore, banks are required to hold a minimal amount of capital which functions as a buffer for unexpected retrievals. To accomplish this, credit ratings are given per financial transaction and are a reflection of the creditability of the investments. The process of providing ratings comes with some issues, partly due to the oligopolistic market structure of the rating industry. The three main credit rating agencies, which are recognized by the SEC and make together 90% of the whole rating market, are Standard & Poor’s, Moody’s and Fitch. This oligopoly position 1

Article 3(1) (a) Regulation (EC) 1060/2009 of the European Parliament and of the Council of credit rating agencies

INTRODUCTION | 6

can be the cause of negative effects. When a rating agency produces ratings, it are the issuers who pay for those services and who prefer favorable ratings. The investors, on the other hand, are the ones who rely on those ratings for free and who prefer accurate ratings. Because an issuer depends on the services of these agencies for their ratings, those agencies might take advantage of this position. The huge increase in power and collective dependency of the rating agencies have contributed a lot to the weakening of the financial system and potentially to the financial crisis. The main reasons for this must be found at the level of the regulation which is enforced on the credit raters. In 1988 the First Basel accord came into action. These are international standards which main goal is to assure the stability of the financial system. The Baselcommittee mainly contains standards and guidelines which must be followed by member states but there still exists a lack of formal entitlements. This has generated the question whether the system has a potential for improvement and how this can be done. We investigate whether it is the big dependency which must be avoided by creating more competition, or whether it is better to look for alternative solutions.

The main goal of our research exists in determining whether increased competition can offer a solution for the perverse effects which come with the use of credit rating agencies or whether it is better to use alternative models like ‘the investor pays model’ , ‘government pays model’ or other models. In a system with more competition, the opportunity is offered for the market to be self regulating in a way perverse effects can be avoided. This brings us to our research question, which is: ‘What is the effect of increased competition on the quality of credit ratings?’. In part one of our thesis, we critically review the debate on credit rating agencies, with special attention for the particular form of the rating market which leads to an oligopoly position for the three major credit rating agencies. Part two contains an empirical study in which we do research after the effect of increased competition on the quality of credit ratings by setting up three different tests and by analyzing the outcomes. Thereafter, part three focuses on alternative models

INTRODUCTION | 7

and their potential to increase the quality of credit ratings. We end our master thesis with an overall conclusion.

INTRODUCTION | 8

1

Literature review

1.1

History of credit rating agencies

1.1.1

concept

The market for credit ratings was born in 1909 with the attendance of the first credit rating agency, named Moody’s (Richard, 2002). It was in the year 1900 that John Moody established his former company ‘John Moody & Company’ and published manuals which enabled to analyze the credibility of financial instruments (Partnoy, 1999). It was until 1909 that John Moody decided to enlarge its trading activities and to make itself able to judge the credibility of an instrument. The market for credit ratings has increased a lot during the following years and has opened itself for the entrance of competing credit rating agencies as Standard & Poor’s and Fitch. The European directive describes CRA’s as follows: “Credit rating agencies publish advice about the creditability of a certain issuer or the quality of a certain financial instrument at a certain date. These advices on their own do not form recommendations within the meaning of this directive. Rating agencies must take into account intern behavioral guidelines and procedures to make sure that the published ratings are a correct reflection and that they publish potential interests or conflicts of interests which are related to the financial instruments or the issuers who are the subject of the ratings" (Regulation 2003/125/EG). Hence, credit rating agencies make it possible to get rid of the information asymmetry, which exists between the investor and the issuer of an investment and which is discussed further in the paper.

Because of the importance of credit ratings in the financial markets, the regulation of these ratings is crucial. In the past, confidence was given to the fact that credit rating agencies put much weight to the value of a good reputation and that building such a reputation would cause the perverse incentives to be avoided. This theory is also called the reputational capital view (Richard, 2002). The credit

LITERATURE REVIEW| 9

rater knows that he depends on the orders of its clients for his future income and hence that he must create confidence to maintain a reputation of reliability. This made the reputational capital theory an excellent system for self regulation in the rating market and seemed to work until the arrival of the financial crisis (Hunt, 2009). Credit rating agencies and the sometimes biased credit ratings played a major role in the financial crisis and created the need for more regulation concerning credit rating agencies.

1.1.2

Evolution of the regulation on CRA's

S&P, Moody’s and Fitch published in the beginning of the nineteenth century bond ratings. These were the first publicly disclosed ratings on the market and were soon sold to investors (Partnoy, 1999). At that time it was still an investorpays model in which it was the investor who paid the credit rater for the rating. The financial world began to trust more and more on those credit raters (White, 2009). In 1936 the bank regulators drew up a decree whereby it was prohibited for banks to invest in ‘speculative investment securities’. These investments are rated BB+ or lower (according to S&P and Fitch) or Ba1 or lower (according to Moody’s). This way the investment behavior and hence decisions of banks were dependent on the opinion of the three big rating agencies, providing S&P, Moody’s and Fitch an oligopoly power.

The United States Securities and Exchange Commission introduced in 1975 the new American concept ‘Nationally Recognized Statistical Rating Organizations (NRSRO)’. 2 The three biggest raters were immediately classified under this statute. The purpose of the SEC was not to let banks keep the same amount of reserve, but to make the amount dependent on the risk of the investment. This risk

2

U.S. Securities and Exchange Commission, "Report on the Role and Function of Credit Rating Agencies in the Operation of the Securities Markets, As Required by Section 702(b) of the Sarbanes-Oxley Act of 2002", January 2003, 5-15. See also sec3(62) of the "CREDIT RATING AGENCY REFORM ACT OF 2006" for a definition of NRSRO


was determined by the level of the credit rating, which in turn was determined by a credit rating agency. Only ratings coming from NRSRO’s could be used to calculate the capital requirements, whereby the oligopoly position of S&P, Moody’s and Fitch was only intensified. By implementing the NRSRO-statute the entrance to the rating market is being hindered even more. Furthermore, it is very difficult for an entrant to build up so much information, reputation and connections as the three big incumbents have already done in a period of 100 years. Currently there are 10 rating agencies that enjoy the NRSRO statute (Hunt, 2009). However, the competition in the market is quickly tempered if we know that S&P, Fitch and Moody’s together have a market share of more than 90%.3

In 2004, the IOSCO (International Organization of Securities Commissions) already implemented some kind of guidelines the CRAs must stick to when giving a rating, named the IOSCO code. Unlike the Rating Agency Reform act, the ‘code of conduct fundamentals for credit rating agencies’ is also applicable outside America. As described by article 4.1, “comply or explain” is the guiding principle which states that rating bureaus have to publish their code of conduct and indicate how the IOSCO-principles are implemented in their own code (Sommer, 1996). When they do not implement the IOSCO fundamentals, they have to explain how and why their code deviates from the IOSCO-principles.4 The code can be divided in four important parts. First the rating agencies must guarantee the quality and integrity of the rating process. Next, CRAs must take measures to assure the independency and to avoid conflicts of interest. The third part requires transparency and clarity of the ratings and in the final part rating bureaus are required to work out procedures which guarantee the confidentiality in handling dossiers.

3

U.S. Securities and Exchange Commission, "Speech by SEC Commissioner: Remarks at the Commission Open Meeting, by Commissioner Kathleen L. Casey", 3 December 2008, http://www.sec.gov/news/speech/2008/spch120308klc.htm. 4 The Technical Committee of the International Organization of Securities Commissions, Code of Conduct Fundamentals for Credit Rating Agencies, December 2004, 11p.


Because of the big responsibility of credit rating agencies when giving credit ratings, it is important that strict supervision also exists at European level. The European commission carefully analyses the way the rating agencies function. On 16 September 2009, a first European directive came into action, which aimed at assuring credit ratings used in the community to be independent, objective and of adequate quality (Regulation 1060/2009). An amendment to the CRA Directive entered into force, on 1 June 2011, giving exclusive supervisory power to the European Securities and Markets Authority (Regulation 513/2011). Because a lot of flaws and disadvantages which are caused by rating agencies are not dealt with in the existing European regulation concerning credit rating agencies, the European Parliament is working on a proposal to enhance the regulatory framework.5 This has generated the desire of an increase in entrance possibilities, which will result in issuers no longer being dependent on only three credit rating agencies, named Moody’s, Standard & Poor’s and Fitch. Such an increase in competition could have a positive effect on the existence of potential conflicting interests. The economical effects which are generated by an increase in competition is analyzed in our economical model, further in the paper.

1.2

The functioning of credit rating agencies

Credit rating agencies play a crucial role in the functioning of the financial markets as they determine the credibility of an investment and the risk for default. On the other hand, credit raters offer monitoring services by which they are able to force issuers to take corrective measures to avoid a downgrade (De Haan & Amtenbrink, 2011). This is called the ‘watch’ procedure and will be explained in more detail. The three main credit rating agencies which are usually very comparable in structure and method are all making use of the issuer pays model (Deryn, 2009). Credit ratings can be defined on a scale of numbers and letters and differ according to the credit rater who published them. Within those rating 5

Proposal for a regulation of the European parliament and of the council amending regulation (EC) N° 1060/2009 on credit rating agencies


categories, modifiers are used to make a further diversification. As you can see from the table, both Fitch and S&P make use of pluses and minuses, where Moody’s makes use of numbers.

Investment grade

default

Moody’s

S&P, Fitch

Numerical value 6 assigned *

AAA Aa A Baa Ba B Caa Ca C D

AAA AA A BBB BB B CCC CC C D

28 24,25,26 21,22,23 18,19,20 15,16,17 12,13,14 9,10,11 7 4 2

A credit rater is an intermediary between the issuing entity and the investor by making valuable information available to the public and reducing the information asymmetry this way (Tang, 2008). On the other hand, a credit rating agency is able to change an existing credit rating when an event occurs which has a significant effect on a given rating. Most of the time, the intention to change a rating will be notified by a credit rater via ‘outlooks’ and ‘watchlists’ (De Haan & Amtenbrink, 2011). A rating outlook is an indication of the opinion of a credit rater about the direction a rating is likely to move over a medium period, more specific a one- or two- year period (Hamilton, 2004). A watchlist, on the other hand focuses more on the short time. The introduction of an outlook or a watchlist makes it possible for the credit rater to coordinate the behavior of both the investor and the issuer and to engage in an implicit contract with the rated firm (Leenaars, 2003). Below, both the information-transformation function and the coordination function are discussed in more detail.

6

*Multiple numerical values for a single rating level represents ratings with a + qualifier, no qualifier and a – qualifier respectively for ratings from S&P and Fitch. For ratings from Moody’s, the numerical values represent ratings with numerical modifiers 1, 2 and 3 respectively.


1.2.1

Information asymmetry problem

The economic function of a credit rater is defined as getting rid of existing information asymmetries by analyzing inside information and making this public in the outside-world, also known as the information-transformation function (Boot, 2008). According to this view, the use of external credit raters does have added value in the calculation process of credit risk and hence is able to reduce the existing information asymmetry. This view can also be understood in the following citation of Moody’s: “Generally, institutional borrowers know more about their companies than do their lenders. Moody’s helps to reduce this asymmetry of information. Accordingly, credit ratings, in aggregate, lower the costs of borrowing and lending and increase overall market efficiency for both issuers and investors” (Boot, 2006, P.3).

The asymmetry problem can best be understood as a situation in which the issuer has information about the value of the firm which the investor doesn’t have and which the issuer can not communicate in a credible way to the market (Bruyninckx, 2008). In those scenarios, a credit rating agency can function as an intermediary between the issuer and the market by analyzing the issuer and by reporting the gained information in an objective way to the public. Information- asymmetry does not only exist at the level of issuer – investor, but also between the rated entity and the credit rater. This problem can be seen as an agent-principal relation between on the one side the agent or CRA and on the other side the issuer or the principal (Boot, 2006).The credit rater calculates a rating in request of his principal, the issuer. Doing so, the agent has to take into account the interests of his principal and respect those, but very often his assignment will be inspired by its own conflicting interests (Tang 2008).Between both parties, information asymmetry could occur; this means that both of them own information which the other party doesn’t have (Donkers, 2010).This reciprocal information asymmetry will cause both §1 ex ante as §2 ex post effects, which will be explained further in this paper.


§1 Ex ante effects: Before contracting with the credit rater, the issuer will have at his disposal some information which he thinks is valuable and which he will only release to the credit rater when he is convinced that he can gain an economical benefit this way (Jorion, 2005).The credit rater in his turn will have information about the methods and procedures to come to an optimal rating, something he will use tactically during the negotiations with the issuer to optimize its income. Here the risk exists that the credit rater will only aim at generating loyalty with the potential client, the issuer, and less at creating a qualitative good rating (bruyninckx, 2008).These ex ante effects are a form of adverse selection, which attracts bad risks because the one who keeps the most information hidden, will often have the greatest interest in negotiating a rating agreement.

§2 Ex post effects: Because of the information benefit, the risk exists that the credit rater will change its attitude towards the issuer once the agreement is formed and will not behave in a way that was agreed or which may be expected from him (White, 2010; Donkers, 2010). Because the credit rater has the possibility to keep his rating methods, the information and the processes to determine the rating secret to the public, he has a certain amount of freedom to make his decisions. He can give a personal influence to the rating, which makes the rating more positive than in reality, to please the issuer but which also leads to the rating losing its objectivity. Because the credit rater is dependent on the assignments of the issuer, this danger will be very present. This is an illustration of the concept ‘moral hazard’, which enters because the issuer can not observe every activity of the credit rater and so will be dependent on the reliability of this last one (Leenaars, 2003).

1.2.2

Coordination function

Investors depend for their decision whether they should or should not invest mainly on a given rating. On the other hand is the credit rater able to put pressure on a firm by threatening to change a given credit rating. This system of interactions can be described as the coordination function of credit ratings (Boot,


2006). A credit rating agency is capable of blocking an entire investment decision of a potential investor by publishing a non-investment grade rating. Those blocking effects can be found in simple investment restrictions or in specific regulation, both of them influencing the behavior of market participants and their investment decisions (Fried & Howitt, 1980).The major problem is situated in the fact that firms have information which the investors do not have, which leads to a lack of faith in the market. On the other side will the existence of an implicit contract between the credit rater and the rated firm cause a potential danger for pressure on the rated firm (Banier & Hirsch, 2008). In case an event occurs which could have a significant effect on the existing ratings, the credit rater can rely on the implicit contract by demanding from the rated firm to take certain correcting measures in order to retain the current credit rating. This procedure is also known as the ‘credit watch’ procedure (Frost, 2006). This system of interactions will be maintained because the rated firms know the importance the investors attach to a published rating and hence they have to try to keep the credit raters satisfied in order to receive a wanted rating. The consequences of a change in an existing rating can be of significant importance for the financing of a rated firm, which causes the firms to respect the credit rater and the implicit contract.

1.2.3

Advantages & Disadvantages of the use of CRA's

There are several advantages which can be ascribed to credit rating agencies (Partnoy, 1999). First of all, it is important for investors to gain information about the risk they are taking when making a certain investment, in order to receive an adequate compensation for this amount of risk. Compensation is given in the form of interest, whereby interest rates are related to the level of risk of an investment instrument (Brigo & Mercurio, 2006).This means that good firms, which have low levels of risk, are able to receive higher credit ratings and hence might borrow money at more favorable interest rates. Those firms, which are already performing well, are able to expand their business even faster and hence will stimulate the overall economy. By determining the appropriate level of compensation, credit


rating agencies enable investors to make a well balanced trade-off between risk and interest rate by providing a risk-return ratio. Since not all investors are opposed to risky investments, they are now able to be rewarded in a correct manner for taking on such high levels of risk. Finally it is important to notice that credit raters can function as a warning for underperforming firms which have low levels of credithwothiness because they hold on too much debt or because they are not able to pay back in time (White, 2010).

On the other hand, a lot of disadvantages and flaws are linked to the use of credit rating agencies. The methods those rating agencies use to calculate those ratings became more and more subject to discussion and critics, because they did not always generate optimal results (Cantor & Packer, 1994). By this, a credit rating agency will be confronted with all kinds of unwanted side effects of the rating process which cause the ratings to lose their objectivity (Boot, 2006). A credit rating agency who functions based on the issuer pays model, generates ratings when they are asked for by the issuer, most of the time this is the firm in which people can invest. It is the issuer who pays for the rating and who creates a source of income for the credit rater (Kudva, 2010). This causes the credit rater to lose its independency and will lead to ratings which are more positive for the issuer, but which is not always a true reflection of the reality.

Another potential side effect can be found in the fact that credit raters have a limited liability. They can not be held responsible for false ratings they have published. This generates a certain amount of quasi-immunity (Bruyninckx, 2008). Knowing that only a limited number of recognized rating agencies are present in the market, named Moody’s, Standard & Poor’s and Fitch, the market mechanism will not reach a complete functioning. The quality of the ratings will suffer from this because the credit rater knows that the issuer only has limited possibilities to switch to another credit rater.

The goal is to find models which aim at increased transparency between the investor and the investment by informing the investor in a better way. Another


working point is to avoid entrance obstacles which limit competition in a way that has a negative effect on the quality of the ratings. The potential of increased competition to improve the quality of the rating is investigated in this paper.

1.3

Oligopolistic market

The credit rating market has been characterized for years as a duopolistic market, dominated by Standard & Poor’s and Moody’s. In 1990, a third player, named Fitch, made its entrance by increasing its market share through various acquisitions. At present, the rating market consists of three major players, which are together in the possession of a market share of 90% (Smant & Van Der Ent, 1994).This high degree of concentration creates a noncompetitive market in which only a limited number of players have market power. This leaves the issuers with a limited possibility to switch to a competitor (Becker & Milbourne, 2011). The proposed measure resulting from the European proposal n° 2011/0361, indicating that the issuer must consult two different credit rating agencies in order to become an objective rating, only worsens the effect of limited competition (proposal 2011/0361). Up until recently, the possibility to entrance in the rating market in America was disturbed by a regulatory barrier to entry, created by the SEC (The Securities and Exchange Commission). In 1975 the process for recognizing credit rating agencies, ‘Nationally Recognized Statistical Rating Organizations’ (NRSRO), came into action. 7 This NRSRO statute formed a significant entry barrier because only NRSRO’s could issue ratings which are officially recognized by the SEC. The European regulation undertakes all kinds of measures to increase competition and the proposal for a new European regulation intends to go even further to achieve this goal (De Haan & Amtenbrink, 2009). Yet the increase in the number of players and hence in competition does not guarantee qualitative ratings. The 7

U.S. Securities and Exchange Commission, "Report on the Role and Function of Credit Rating Agencies in the Operation of the Securities Markets, As Required by Section 702(b) of the Sarbanes-Oxley Act of 2002", January 2003, 5-15. See also sec3(62) of the "CREDIT RATING AGENCY REFORM ACT OF 2006" for a definition of NRSRO


assumption of increased quality only holds when there is a factor which causes high quality raters to benefit and low quality raters to be punished (Hunt, 2009).

1.3.1

Entrance barrier

An entrance barrier is an obstacle which avoids that a credit rater will have automatically entrance to the rating market. The credit rating market is characterized by all kinds of entry barriers. Further in this paper, a distinction will be made between natural (§1) and institutional (§2) entrance barriers. §1 Natural entrance barriers: A credit rater is dependent on the issuer for the rating assignment and to generate an income this way. The prospects that he will attract new clients in the future, makes that the credit rater will have attention for the way he is seen at the public (White, 2010). He will profit from building a reliable and credible reputation, to assure the amount of clients for the future. This reliability demands ratings to be formed independent from the issuer and to contain a minimum amount of objective information (Hunt, 2009). Besides these natural barriers, the entrance is also disturbed by law, by all kinds of rules and institutional restrictions which make that not every credit rater is a nationally recognized entity. A NRSRO statute gives a credit rater the exclusive status of recognized credit rating agency (Partnoy, 2006).

§2 Institutional entrance barriers: The credit rating industry has been dominated during a long time by two rating agencies, forming a duopolistic market (Camanho & Deb, 2011). Those two agencies were Moody’s and Standard & Poors. This was mainly due to the entrance

of

the

NRSRO

“Nationally

Recognized

Statistical

Rating

Organizations”. Since 1975 rating agencies could only be in the possession of such a certificate when they complied with all the different aspects, and which is granted by the “Securities and Exchange Commission” (SEC) (Partnoy, 2006).


Because investors prefer credit rating agencies who own such a certificate, this will have a limiting effect on the entrance of new credit rating agencies, without a certificate (White, 2008). Paul S. Atkins, 2008, argumented following quote: “The unintended consequence of the SEC’s approach to credit rating agencies was to limit competition and information flowing to investors. The legislative history reflects a genuine concern that the SEC facilitated the creation of – and perpetuated – an oligopoly in the credit rating business.”8

These entrance barriers gave credit rating agencies a prominent place in the financial market. The high amount of concentration which gives an oligopoly power to the three main players, draws our attention to the stability of the credit rating market and the negative consequences that follow. Therefore we can ask the question whether a system of self regulation, which is made possible by more competition on the market, is able to offer a solution.

1.3.2

Entrance to ratings markets

The entrance of the third rating agency, named Fitch, allows us to analyze the effect of increased competition in the rating market. Where issuers could previously only ask ratings and services from Standard & Poor’s or Moody’s, now the possibility is created to switch to a third agency, Fitch. Fitch’s roots go back to 1913, when it was founded by John Knowles Fitch under the name Fitch Publishing Company (Wolfson & Crawford, 2010). But at that time Fitch was nothing more than a small player which had no or only a small effect on the existing oligopoly. It was in 1989 that Fitch became a noticeable player in the credit rating market, because of the recapitalization of Fitch (Packer & Cantor, 1995). This recapitalization was followed by a merger with IBCA limited in 1997. From that time on, Fitch continued to grow and became a serious competitor of the two existing agencies (Becker & Milbourne, 2010).

8

U.S Securities and Exchange Commission, "Speech by SEC commissioner: Remarks to the institute of international bankers by SEC commissioner Paul S. Atkins", 3 maart 2008, www.sec.gov/news/speech/2008/spch030308psa.htm


This way the amount of competition increased, which led to all kinds of effects. Further in this paper we try to analyze these effects and determine how this has influenced the quality of the ratings. More specific we put up a regression model to measure the effect of increased competition presented by the entrance of Fitch on the quality of the ratings published by the incumbent players, Standard & Poor’s and Moody’s. The quality of a rating is an important factor in our research, that’s why it is crucial that this term is clearly defined. Quality is defined as the difference between the published rating and reality. A qualitative rating has to be informative, precise and most important; it must be a clear representation of the truth. Ratings with low quality, most of the time high ratings, will be in favor of the issuer but will be detrimental for the investor, because the rating is influenced by preferences and will be subjective because it reflects other things than the probability of repayment (Huang, Chen, Hsu & Wu, 2004). The income of a rating agency depends on the fees being paid for the ratings. It is the issuer, most of the time the company that is rated, that will have to pay for those fees and the investor can consult these ratings for free (Deryn, 2009). It is advantageous for the issuer to receive a rating which is as good as possible, while the investor has more interest in a rating being a correct reflection of the reality. It is up to the credit rater to balance those two interests and to create a trade-off between current incomes and reputation.

1.3.2.1

Reputational, market-sharing and disciplining effect

In a credit rating market where the possibilities to enter are limited, the reputation of the credit rating agencies will be of major importance (Becker & Milbourne, 2011). Therefore the rating agency will try to publish ratings which are very reliable and which reflect the truth, because this will determine whether the issuer will or will not ask again for his services in the future (Partnoy, 1999). It is important that the investor has confidence in the rating, because if not, the issuer would no longer believe that it is profitable for him to ask for a rating. This could


be defined as follows: “Every time a rating is assigned, the agency’s name, integrity and credibility are on the line and subject to inspection by the whole investment community” (Wilson, 1994). When the possibility to enter is offered and finally competition is allowed, the importance of reputation will decrease. This because increasing competition will cause expected future rents for the incumbents to fall (Becker & Milbourne, 2010). This way, competition will lead to ratings that try to favor the issuer by giving ratings which do no longer reflect reality. A potential effect of increased competition will exist in the fact that the ratings of the existing rating agencies will improve (Allen, Carletti & Marquez, 2009). So a rating which was a B-rating before, might after the entrance of a third player, be transformed into an A-rating. The main idea, which was also the result of the research by Bar-Isaac and Shapiro (2010), is that there are greater incentives to provide accurate ratings when reputation losses are higher, because this way there’s a lot to gain from informative ratings (Bar-Isaac & Shapiro, 2010). The reputational theory states that competition can undermine the future rents for the incumbents, but we must argue whether this is also the case when there is a situation of massive increase of the credit rating industry, which generates an increase in total rents (Mathis, Mcandrews & Rochet, 2009). This is why we must conclude that the reputational theory only makes sense when more competition corresponds to lower future rents, holding market size constant. The seminal study of Mathis et al. (2009), shows that reputation by itself is an insufficient mechanism to discipline the reporting of objective ratings. In a situation of increased competition, credit rating agencies tend to inflate ratings and this can only be avoided when a major fraction of the credit rater’s income stems from other sources than rating financial instruments (Schultz-Larsen & Hasling Kyed, 2012). The decreasing importance of reputation can be found in a number of statements which fall under the theory of ‘market-sharing effect’ (Pragyan, 2012). The investigation of Klein and Leffer’s (1983) shows that building reputation will be of big importance when this generates the possibility to create more income in the future. Because there are more players in the market in a situation of increased


competition, the incomes per player will be limited and so the effect of reputation will decrease, and because of this effect, the importance of the reputational view will decrease as well (Becker & Milbourne, 2009). Another consequence is that rating agencies will, especially when situated in a competitive market with an elastic demand side which is very price sensitive, opt to decrease their price to attract more issuers (Hill, 2009). Before the entrance of the third rating agency, Fitch, credit raters had a kind of oligopoly position, which offered them the possibility to count high prices. Credit raters tried to maintain this oligopoly position after the entrance by charging extremely low prices and maintaining this way a large part of the market share. This is called the disciplining effect (Mathis, Mcandrews & Rochet, 2009). As a conclusion we can say that the reputational effect might be weaker in situations of high competition, which means that the incentives to maintain high quality ratings declines (Becker & Milbourne, 2011). In our economical model we provide evidence for this conclusion.

1.3.2.2

Rating shopping

Another effect we must take into account in a situation of increased competition is rating shopping. This phenomenon describes the situation in which the issuers look around for the best rating from the different rating agencies and choose only to publish the best one (Skreta & Veldkamp, 2009). When the barriers to entry are dismissed and competition is allowed, the issuers can choose between more parties to rate their products (Sangiorgy, Sokobin & Spatt, 2009). Theoretical models by Skreta and Veldkamp (2009) show the effect of rating shopping on the quality of the ratings. Results show that more complex financial products are a trigger for more rating shopping, which has as consequence that the ratings are more biased. The effect of rating shopping will be bigger when one rating agency uses criteria which are more lax than the criteria the competitors use (Adelson, 2006). Although increased competition worsens the effect of rating shopping, this isn’t as much the case for corporate securities (Becker & Milbourne, 2010). Because in


test 2 we use ratings of corporate bonds, the problem of rating shopping will be mainly avoided for that test. This can be explained because corporate issuers can not choose whether or not to publish a given rating because Moody’s and Standard&Poor’s have a policy which states that all taxable corporate bonds publicly issued in the US must be rated and published. So it doesn’t matter when an issuer doesn’t want to pay for a rating, the rating will be published anyway. This phenomenon is also called ‘unsolicited ratings’ and describes the situation in which the evaluation of the creditworthiness of an instrument or an issuer is granted without any involvement of the issuer (Cohen, 2011). In our research, after the entry of Fitch, some issuers might ask for the ratings of this credit rating agency, but they do not have the possibility to eliminate the ratings she already received from the incumbents, Moody’s and Standard & Poor’s, this way the variable ‘number of ratings issued by incumbents’ won’t be affected. However, our sample for the empirical study consists of both bond and firm ratings, for this reason an alternative interpretation of the results, which includes the effect of rating shopping, will be added to the empirical study.


2

EMPERICAL STUDY

2.1

Model setup

We wish to examine the potential effect of increased competition on the quality of the credit ratings by doing three different tests which are mainly based on previous research of Becker and Milbourne (2011). Increased competition is determined by the entrance of Fitch, which was in 1913. Our sample ranges from 1996 to 2006, this because Fitch only became a remarkable player since the late 90’s caused by , as already mentioned in this paper, a major recapitalization. This entrance resulted in a significant increase in the market share of Fitch. This will be calculated by the fraction of the total number of bond ratings issued by Fitch in a particular year over the total number of bond ratings issued by Standard & Poor’s and Moody’s in the same year in a specific industry. To classify the industries we used the North American Industry Classification System (NAICS), see table 1 and the Morgan Stanley Capital International World (MSCI World), see table 2. In the figures below you can see the graphical evolution of the market shares of Fitch over our sample period in both industry classification systems, starting in 1996 until 2006.

Fitch's market shares NAICS 0,4 0,3 0,2 0,1

Market shares

0

EMPERICAL STUDY | 25

Fitch's marketshare MSCI World 0,3 0,25 0,2 0,15 0,1 0,05 0

Marketshare

The graphs represent the mean of the market shares over the different industries per year and are a reflection of the level of competition in the rating market. In 2000 and 2005 we notice a peak, due to the acquisition of BankWatch and the acquisition of Algorithmics respectively. Table 1 shows the market presence of Fitch over the different industries according to the NAICS, which indicates that by the end of our sample period, Fitch was especially present in Real Estate, Rental And Leasing with an average market share of 0,42 (NAICS 53) followed by Retail Trade with an average market share of 0,39 (NAICS 45). Also Public Administration, Management of Companies and Enterprises and Utilities are characterized by a prominent presence of Fitch. In table 2, you can see the market presence of Fitch over the different industries according to the MSCI World index. This table shows that by the end of our sample period, Fitch was especially present in Finance with an average market share of 0,38, followed by the Communication industry with an average market share of 0,29. We analyze how the quality of ratings corresponds to the increased competition, caused by the entrance of Fitch. Quality is defined as the difference between the rating, which was determined by the rating agency, and the true credibility of the financial instrument or entity. It is important to notice that the rating quality is not always objective. This means that the failure of a firm which received a high rating, is not automatically due to a false rating/low quality, but is influenced by a lot of other external factors.


To measure our research question, we set up two regression models using panel data and one correlation model. In our first test, we analyze the regression between the firm ratings and the evolution of the market shares, in which credit rating is the dependent variable. One possible outcome is that when there is more competition, the level of the ratings will be higher (e.g. AA), meaning that the quality decreased. This can be explained because of the decrease in reputational effect, which causes the rating agencies to favor the issuer by giving him a high rating. In that case the high rating is not due to an increase in creditworthiness of the entity but solely to enhanced competition. Another possible outcome is that competition does not decrease the quality of the ratings. In that event we can not assume that there is a decrease in the reputational effect and that even with increased competition, credit ratings are reliable. We expect to come to an outcome which confirms our idea that a general increase in the level of the ratings is a direct effect of competition, meaning that competition leads to credit raters favoring the issuer instead of giving credible ratings. The problem with these increases in ratings is that decrees can be circumvented by this. In 1933 a decree came into action which contained a prohibition for banks to invest in speculative bonds, which do not reach to ‘investment grade’ level. These investments are rated BB+ or lower (according to S&P and Fitch) or Ba1 or lower (according to Moody’s) . When the level of a rating increases only because of competition, the creditworthiness of the instrument will stay the same, and this way banks are allowed to invest in an instrument in which they would not be allowed to invest without competition.

In our second test we analyze whether a rating is qualitative. Quality is measured by analyzing the correlation between the bond yield and the bond rating. The intuition behind this measure is that a bond rating, when qualitative, must reflect nothing but the prediction of a future default. A bond rating that has a high correlation with the bond yield is a rating which transforms information in a correct manner and can be seen as a qualitative rating.


We expect the correlation to be lower when there is more competition in the rating market, this because of the decrease in quality as explained above.

Finally we test the predictive power of the default of ratings when the level of competition changes. This will be done by comparing situations of default with the rating that was given one year before this default event. Since ratings are supposed to predict the likelihood of default, we can measure whether a rating becomes less credible when there is more competition and will lose its predictability. In our research, situations of default will be defined as a rating with level ‘D’.

2.2

Data

For the firm ratings in our sample we made use of the Compustat North America Database. There we found the monthly evolution of the ratings for the different firms from 1996 to 2006, rated by Standard & Poor’s. To make the test more relevant, we duplicated the test with another industry classification system, namely MSCI World, which gives us the evolution of the ratings from Standard & Poor’s, Moody’s and Fitch per year. For the analysis we transformed the rating scale into a numerical variable, in the table below you can find an overview of the transformation process.


Investment grade

default

Moody’s

S&P, Fitch

Numerical value 9 assigned *

AAA Aa A Baa Ba B Caa Ca C D

AAA AA A BBB BB B CCC CC C D

28 24,25,26 21,22,23 18,19,20 15,16,17 12,13,14 9,10,11 7 4 2

For the duplication of test 1 we took the average of the three ratings. For the market shares, we used the total number of bond ratings issued by S&P, Moody’s and Fitch, which we drew from the Mergent Fixed Income Securities Database (FISD). As previously mentioned, we divided the market share of Fitch by the market share of the three credit raters over the different industries. The industries refer to the 2-digit NAICS and for these codes we relied on the Oriana Database. Via compustat we could only generate ratings from firms of three industries, namely Utilities, Manufacturing-Wood, Paper, Printing and ManufacturingMetals, Machinery, Computers. This will not bias our results because we still have a substantial number of observations. The remaining sample consists of 13350 observations. We retrieved bond yields from Bloomberg to do the correlation analysis between yield spreads and ratings. Yield spreads are defined as the difference between yield to maturity of a bond and the yield to closet maturity treasury bond. Because the total number of companies according to the NAICS was 24416, we only used firms from the industries Accomodation, Real Estate and Construction. Because every firm issues a number of different bonds, we only took one random bond transaction per firm, assuring that the different bonds we use have the same maturity.

9

*Multiple numerical values for a single rating level represents ratings with a + qualifier, no qualifier and a – qualifier respectively for ratings from S&P and Fitch. For ratings from Moody’s, the numerical values represent ratings with numerical modifiers 1, 2 and 3 respectively.


For our third test we used firm ratings and default events (a ‘D’ rating) and drew those from the Compustat Database. As mentioned before, a future default event is defined as having a ‘D’ rating within the next year.

2.3

Results

In this chapter we discuss the results of our test which aim at discovering the effect of increased competition in the rating market on the quality of the ratings.

2.3.1

Test 1: Regression with firm credit ratings

In the first test, we regress firm credit ratings on Fitch’s market share. We start with a regression where we use the level of the ratings as a dependent and market share as independent variable.

Regression model: RATt= α + β1MARt + ε Dependent variable: RAT = rating level issued by the incumbents in year t Independent variable: MARt = market share in period t (with t ranking from 1996 to 2006)

In the table below you can see the results for the dataset with industries from NAICS. variable C Market share

Coefficient 20,79 -2,27

t-statistic 203,93 -7,32

probability 0,000 0,000

R Squared 0,004

When we do not include control variables, we see that the coefficient on Fitch’s market share is negative, with a probability factor of 0,000, which indicates that market share is a significant variable in the model. The negative coefficient


suggests that more competition pushes ratings towards the lower end of the rating levels (for example toward C). This pattern is also presented in the graph which plots the frequency of each level of ratings for high and low levels of competition. We define low levels of competition as the period before 2000 and high levels of competition as the period from 2000 to 2006. 35 30 25 20 before 2000

15

From 2000-2006

10 5

AAA AA+ AA AAA+ A ABBB+ BBB BBBBB+ BB BBB+ B BCCC+ CCC CCCCC C D

0

From this graph we can see that higher level ratings are less common under high competition and lower level ratings are more common under high competition. As you can see for example the frequency of ratings of level A is 13,7% under low competition and only 9,7% under high competition. When we look at the R^2 of our model, we notice that this is very low (0,004). This low R^2 indicates that the proportion of variation in the dependent variable which is explained by the independent variable is redundant.

The estimated coefficient in this model could be unreliable because we did not add controls. We now turn to a model in which we control for year- and firm fixed effects. Variable C Market share

Coefficient 18,97 3,51

t-statistic 85,94 5,03

probability 0,000 0,000

R Squared 0,64

This pushes up the R^2 to 0,641 and makes the coefficient on market share positive (3,51) and significant.


When we control for the fixed effects of the lagged ratings, we see that the R^2 even goes up to 0,97, meaning that the proportion of variation which is explained by the independent variable has increased to 97 percent. The coefficient of market share is positive (0,05) and significant. regression model: RATt = α + β1MARt + RATt-1 + ε Dependent variable: RAT = rating level issued by the incumbents in year t Independent variable: MARt = market share in period t (with t ranking from 1996 to 2006) RATt-1 = lagged effects of ratings

Variable

Coefficient

t-statistic

Probability

R squared

C

0,28

7,46

0,000

0,97

Marketshare

0,05

3,88

0,000

Rating(-1)

0,98

620,45

0,000

This positive correlation between Fitch’s market share and the level of credit ratings issued by Standard & Poor’s suggests that with more competition, the levels of credit ratings increases, as we assumed before.

To make sure that these results were not due to our sample choice, we duplicated the test with an industry classification system from another database, namely MSCI World. These results show that the coefficient on market share is slightly positive (0,38) and significant when we do not control for fixed effects. Variable C Market share



Probability 0,000 0,0001

R Squared 0,0004

This weak effect can be illustrated by the graph which shows small differences between the levels of ratings with high and low competition.


25 20 15 Before 2000 10

from 2000-2006

5

AAA AA AA+ AAA A+ ABBB BBB+ BBBBB BB+ BBB B+ BCCC CCC+ CCCCC C D

0

As you can see for example all the A rating levels are comparable under high and low competition. When we have a look at R^2, which is again very low (0,0004), we can conclude that the predictive power of the model without controlling for fixed effects is again redundant.

When we do control for firm- and year fixed effects, to make the model more reliable, we see that this pushes up the R^2 to 0,73 and hence increases the predictive power of the model. The coefficient of Fitch’s market share remains positive (0,34) and significant. Variable C Market share


t-statistic 1312,27 -5,49


R Squared 0,73

When we control for fixed effects of lagged ratings, we see that this pushes up R squared to 0,98 and makes the coefficient on the market share positive and significant (0,23). From which we can again conclude that in situations of increased competition, the levels of credit ratings are significant higher.


Variable

Coefficient

t-statistic

Probability

R squared

C

0,18

12,06

0,000

0,98

Market share

0,23

3,51

0,000

Rating (-1)

0,99

1405,69

0,000

These results diminish the value of the use of credit ratings in the financial system because the increase in the level of these ratings are not a representation of the change in creditworthiness of the firms and do only reflect the increase in competition.

Drawbacks We must draw attention to some issues related to our sample. First, there might be some selection in our sample of market shares of Fitch. This lack of randomness is due to the fact that Fitch will be more likely to enter into industries in which firms receive low ratings and hence are more likely to ask for a third opinion. The higher market shares in those industries can be explained by the assumption that it was easier for Fitch to grow its presence in booming sectors, where credit demand was high. Second, our findings could be biased because of the effect of firm heterogeneity. This could be avoided by including accounting-based firm controls like firm size, profitability, indebtedness, etc. However, this was almost impossible for our research because our data originates from different databases, which requires transforming any extra control variable manually to link them to the correct time period. The test could be enhanced by analyzing whether the effect of competition is felt stronger for firms with higher levels of indebtedness and hence are more concerned about their ratings. This could be done by including several levels of leverage in the model and measure how they interact with Fitch’s market share. For the same reason as mentioned before, this was impossible with the data we had available.


2.3.2

Test 2: Correlation between ratings and bond yields

In the second test we examine the level of information contained in a credit rating under different levels of competition. This is done by measuring the correlation between ratings and bond yield spreads. In this approach we assume that ratings are able to predict the value of bonds, whereby we use the evolution of bond yields to assess the informativeness of ratings. The bond yields are calculated using bonds from MSCI World industry firms whereby we only included bonds with maturity of ten years. We chose to only use bonds with the same maturity to increase the comparability. The correlation between bond yield spreads and ratings will be lower when the ratings are less qualitative and less informative. Because in that case, ratings will predict other things than the chance for repayment. Yield spread is defined as the difference between the yield for the government bond with closest maturity and the yield to maturity of the bonds. For the government bond data with maturity of ten years, we made use of the Federal Reserve’s H15 report. The result of the correlation test between ratings and yield spreads is shown in the table below. rating

yield

rating

1.0000

-0.0897

yield

-0.0897

1.0000

From the table we can see that ratings and yield spreads are negatively correlated, indicating that when rating levels increase, yield spreads decrease.

We now do a regression to measure the interaction between the credit ratings and Fitch’s market share to include the effect of competition. Regression model: RATt= α + β1MARt + ε Dependent variable: RAT = rating level issued by the incumbents in year t


Independent variable: MARt = market share in period t (with t ranking from 1996 to 2006)

As you can see from the table below, the coefficient on credit ratings is positive (0,38) and significant, indicating that credit ratings increase with higher levels of competition. As a consequence we can derive that when there is more competition, the level of credit ratings is higher and the bond yields are lower. Variable C Market share




Next we measure the effect on the correlation between credit ratings and bond yields when the level of competition increases. We do this by splitting up our sample. In one part we measure the correlation in a period of low competition (1996 to 2000) and in the other part we measure the correlation in a period of high competition (2001-2011). The results are shown below and indicate that credit ratings and bond yield spreads are less correlated in a period of high competition.

1996-2000

rating

yield

rating

1.000

-0.0877

yield

-0.0877

1.000

2001-2011

rating

yield

rating

1.000

-0.0662

yield

-0.0662

1.000

The overall results of the above tests show that correlation decreases as competition increases, meaning that a high market share of Fitch implies bond yields to be less related to credit ratings. In terms of information transmission we can conclude that credit ratings are less informative about bond yields in periods of high competition. This is in line with our above results that lower quality ratings are a consequence of higher competition.


Drawbacks By analyzing the results, we must take into account that when bonds come closer to maturity, the price of the instruments will increase because the remaining maturity drops and hence the risk decreases. This causes the yield to decrease as well and is in line with our results that an increase in rating levels (decrease in risk) is negatively correlated with the yield level. Another point of interest is the fact that we only used bonds with maturity of ten years and only one random bond per firm. Our results are probably not biased by the previously mentioned effect of rating shopping because there is limited evidence for rating shopping among corporate bond issuers, see Cantor and Packer (1997) and Jewell and Livingston (1999).

2.3.3

Test 3: Predicting default

Our final test concerns a measure of informativeness of credit ratings, which is examined by their ability to predict future defaults. The ability to predict default is one of the major functions of credit ratings, which makes the relevance of this test very high. On the other hand we must notice that corporate defaults are rare which decreases the explanatory power of this test. An ideal rating should be able to predict default by taking only accounting data about the rated firm into account, and go beyond what easily measurable data can predict.

We regress future default in one year on market share, a dummy for investment grade ratings and an interaction of market share and the dummy for investment grade ratings. Regression model: DEFt= α + β1INVt + β2MARt + β3INVt*MARt + ε Dependent variable: DEFt = Default in year t Independent variable: INVt = Investment grade rating in year t


MARt = market share in year t INVt*MARt = interaction effect The results of the test, which shows the prediction that a firm will receive a ‘D’ level rating within 1 year are shown below. Variable C Investment Market share Investment*market share

Coefficient 0,22 -0,22 -0,04 0,03

t-statistic 20,58 -18,67 -1,06 0,87

Probability 0,000 0,000 0,288 0,385

R Squared 0,18

The dummy for investment grade ratings is significant and has a negative coefficient (-0,22) meaning that firms which received an investment grade rating, are less likely to default within one year. The interaction between the market share and the dummy variable is positive (0,03), which means that when there is more competition, the difference between a default rate predicted by an investment grade level rating and a rating with a level beneath investment grade (speculative grade)10, decreases. When we do the test again but replace the dummy variable with the rating variable, we notice that the results are comparable. Regression model: DEFt= α + β1RATt + β2MARt + β3RATt*MARt + ε Dependent variable: DEFt = Default in year t Independent variable: RATt = rating level in year t MARt = market share in year t RATt*MARt = interaction effect Variable C rating 10

Coefficient t-statistic 0,56 21,42 -0,03 -20,25


R Squared 0,08

Speculative grade is a rating with a level beneath BBB- for S&P and Fitch and beneath Baa3 for Moody’s


Market share rating*market share

-0,95 0,05

-11,68 11,43

0,000 0,000

The coefficients on rating and market share are both negative (-0,03) and (-0,95) respectively and significant. The coefficient on the interaction between rating and market share is again positive (0,05) and significant. By looking at the R^2 , which is only about 0,09, we notice that the predicting power of the model decreased.

By analyzing the results of both models, we can conclude that higher competition does coincide with lower quality ratings. This because the interaction in both models was positive, meaning that the power to predict defaults of speculative rating levels decreases when competition increases and hence the quality of the ratings decreases.

The results of the above mentioned models might be biased because of the absence of controls. In the next model we will examine the predictive power of ratings after controlling for year and firm fixed effects. Results show that the overall fit of the model is better (R^2 of model 1 increases from 0,18 to 0,22 and the R^2 of model 2 goes up as well from 0,08 to 0,14) and that the introduction of fixed effects does not change the overall conclusion about the predictiveness of ratings.

Variable C Investment Market share Investment*market share

Coefficient 0,22 -0,01 -0,22 0,03

t-statistic 16,75 -0,22 -17,52 0,91

Probability 0,000 0,827 0,000 0,360

R Squared 0,22

Variable C rating Market share rating*market share

Coefficient 0,55 -0,90 -0,02 0,03

t-statistic 19,99 -10,42 -17,09 8,15

Probability 0,000 0,000 0,000 0,000

R Squared 0,14


We must draw attention to the fact that industries vary in the possibility to predict defaults. This could bias our results because it might lead to smaller differences in default predictions between investment and non-investment grade firms in those industries where it is harder to predict defaults. We can control for this by including industry fixed effects and an interaction of ratings with those industry dummies.

Results show that by controlling for industry fixed effects, the predictive power of the model slightly increases (R^2 goes up from 0,08 to 0,09). Regression model: DEFt= α + β1RATt + β2MARt + β3RATt*MARt + β3IND22+ β3IND32 + β3IND31 + β3IND33+ ε Dependent variable: DEFt = Default in year t Independent variable: RATt = rating level in year t MARt = market share in year t RATt*MARt = interaction effect IND22 = if the firm is present in industry 22 = 1, if not = 0 (22: utilities) IND32 = if the firm is present in industry 32 = 1, if not = 0 (32: chemicals) IND31 = if the firm is present in industry 31 = 1, if not = 0 (31: manufacturing) IND33 = if the firm is present in industry 33 = 1, if not = 0 (33: furniture)

Variable rating Market share Rating*market share I_22 I_32 I_31 I_33

Coefficient -0,02 -0,77 0,04

t-statistic -18,76 -9,36 9,80

Probability 0,000 0,000 0,000

0,50 0,55 0,48 0,53

19,00 20,83 15,49 20,40

0,000 0,000 0,000 0,000

R Squared 0,09


The coefficients on the different industry dummies show that the default prediction does not differ significantly across the different industries (all coefficients are about 0,5), which can also be seen from the interactions between rating- and industry variables.

Drawbacks A major disadvantage related to this test is that corporate defaults are rare, which has a limiting effect on the results of the test. Our sample consists of 139 firms ranking from 1996 to 2006 and 105 events of default at the one year horizon after a rating was given.

2.3.4

Interpretation of the results

The overall conclusion we can draw from the three above mentioned tests is that the entry of the third rating agency, Fitch, had a significant negative impact on the quality of the ratings which is in line with the existing literature. This decrease in quality can be defined both as an increase in rating levels and a decrease in default predictability as a consequence of increased competition. These conclusions are consistent with the previously mentioned reputational capital view. According to the reputational capital view, efforts of credit raters to maintain a credible reputation would decrease when there was a drop in the potential gains of a good reputation. The decrease in those reputation building efforts would cause the rating quality to diverge as well. As also mentioned before, entrance of Fitch is more likely in situations where low ratings are published because of the effect of a third opinion. This excludes the potential danger that our findings are biased because of a reverse causality effect in which entrance is more likely in situations where high ratings are given, and hence as a consequence when market share increases, rating levels increase as well. Our findings are in line with those from Camanho et al. (2011), which showed as well that increased competition weakens the disciplining effect of reputation as opposed to a situation with fewer players. When there are more players on the rating market, credit rating agencies are not


concerned enough about their reputation to prevent them from lying about the level of the ratings. As previously mentioned, competition generates two effects, named the disciplining and the market-sharing effect. The results of our tests show that the disciplining effect will be larger compared with the market-sharing effect as a consequence of increased competition which generates concerns for their reputation. The overall conclusions of our three tests may formulate a caveat to the European proposal of 15 October 2011 for a new CRA regulation in which one of the proposed measures is that of increased competition.

(Proposal

2011/0361) One must however keep in mind that besides these negative consequences, competition obviously also generates benefits, by creating additional information because of the extra ratings published, which has a positive effect on the financial market.

In our economic model we made use of the issuer-pays model, in which there exists a lack of independency between the credit rater and the rated entity and hence the incentives of both converge. This problem can never be solved by increased competition, as this will only cause the disciplining effect to increase in order to favor the issued entity. A possible solution can be to reintroduce the investor-pays model. Another possible solution is the government-pays model. Two alternative solutions - the platform-pays model and the payment-upon results model - and their positive and negative factors will be discussed in the next part of our thesis.

2.4

Alternative interpretation

The general conclusion of the three tests was that an increased level of credit ratings due to an increased level of competition was explained by a decline in reputation building efforts. Another possible explanation for our observations, being the increase in rating levels, is that with the entrance of Fitch, the low-quality firms had an incentive to switch to the entrant firm. Fitch, doing efforts to increase its market share, tried to attract firms by giving them a favorable rating and the low-quality firms saw an


opportunity for an upgrade of their ratings. As a consequence, most of the highquality and hence high-rated firms, are dedicated to the incumbent firms, which caused the general rating level of these firms to increase. To affirm this assumption, we drew a graph which reflects the evolution in the solvability of the clients of S&P, the incumbent firm. Solvability is measured as total assets over total liabilities. Our dataset consists of the NAICS firms we used in the above tests.

Evolution solvability clients of S&P 4 3,8 3,6 3,4 3,2 3

Total assets/Total liabilities

As you can see from the line graph, there is a general increase in the solvability of the clients of S&P, meaning that either the equity has increased or the liabilities have decreased. These are the previously defined high-quality firms, which receive higher ratings and cause the general rating level of S&P to increase. This view can serve as an alternative interpretation for our results of increased rating levels.


3

3.1

ALTERNATIVE MODELS

Investor pays model

Before 1970, an investor pays model was applicable in the rating industry. This means that it is the investor who pays for obtaining a rating. This enterprise model could be an alternative for the ruling issuer pays model, but neither is this model free from critic and problems (Seaborn, 2010).

A comparison between both models definitely reveals that two different interests are being served and that the rating results could differ according to whose preference is being granted. In the issuer pays model, the issuers pay for the ratings, so they will benefit from their offer being rated as high as possible so they have a good chance to attract investors. This way issuers could ‘shop’ for the highest rating. Contrary, in the investor pays model, it are the investors who pay and they want a rating as accurate as possible. Here the opposite may occur, and ratings may be lower-than-warranted so investor can achieve a higher yield (Kudva, 2010). Although,it is too easy to impute the rating bias completely to the choice of business model. Also other factors, like experience, market power etc. play an important role.

In the issuer pays model, the rating agencies often receive private information about the enterprise. The rating process implies that the rating bureaus conduct an in-depth research in the issuing company and obtain a lot of information regarding that company that is not publicly known. In contrast, the investor pays model remains focused on publicly available information because there is little or no interaction with the company that is issuing an obligation (Caprio, 2012).

One of the most important differences is that the ratings in the investor pays model are no longer distributed among the public, but only available for the investor who paid for them. So a lot of small potential investors will be bypassed because they can not afford a rating. But a problem of free-riding could emerge

ALTERNATIVE MODELS | 44

(Fennell & Medvedev, 2011). Thanks to modern technology, information can be disseminated way faster and so it does not take long before ratings paid for by one investor, leak out to the wider public. This way there will be investors who never paid for a rating themselves, but do profit from the availability. The willingness to pay of investors decreases, leading to declining revenues for the rating agencies, which does not benefit the quality of information. However, this problem is being weakened by the fact that there is always a delay on the leaking of information, and some investors though are willing to pay to avoid the delay (Deb & Murphy, 2009).

As already mentioned, before 1970 rating bureaus made use of the investor pays model, whereafter they transferred to the issuer pays model. Reasons for this transfer could be found in the fact that the emergence of the photo copy machine enlarged the free-riding problem, whereby rating agencies feared losing part of their revenues (White, 2010). Also more complex financial products asked for a more in-depth analysis, which causes the need for CRAs to collect more financial resources by charging the issuer (Jiang, 2012). Another reason can be found in the bankruptcy of Penn-Central Railroad on 1970. This triggered a wake-up call for bond issuers to ensure that their bonds were low risk and so they were prepared to pay the rating agencies for this (Hill, 2004). It needs to be noted that the size of the fee is not regulated.11 It depends on the size and nature of the instrumented being rated. Another remarkable point is that a small part of the revenue of the three big rating agencies still stems from subscriber-based services (Fennell & Medvedev, 2011). Also, while most of the NRSRO’s are employing an issuer-pays model, still a few other use a subscriberbased business model. In figure 1, you can find the ten NRSRO’s together with the business models they use. 12

11

U.S. Securities and Exchange Commission, "Report on the Role and Function of Credit Rating Agencies in the Operation of the Securities Markets, As Required by Section 702(b) of the Sarbanes-Oxley Act of 2002", January 2003, 41. 12 U.S Securities and Exchange Commission, "Summary Report of Commission Staff’s Examinations of Each Nationally ecognized Statistical R Rating Organization, As Required by Section 15E(p)(3)(C) of the Securities Exchange Act of 1934", November 2012, 22p.


3.2

Government pays model

Another possible model is the government pays model. The EU Commission called this a public utility model.13 This model is sometimes put forward because the services of rating agencies perform a public function. The government creates the rating agency and closely monitors it. It does not have an exclusive rating responsibility, but the ratings can be used to be compared with those of the private rating agencies. Finance could come from various resources: taxes, fees paid by the issuers etc. The advantage, just like the issuer-pays model, is that the ratings would be publicly available, free of charge. But also this model has its own disadvantages. A possible conflict of interest may arise. It is not inconceivable that the government may exercise some pressure on the ratings when a company needs to be rated in which the government has a particular interest (Sweeney, 2010). It is also possible that people perceive the ratings of a public entity very reliable and of high quality. An over-reliance on those ratings can exist, even more than with the private ratings. Another possible danger is that the private rating agencies are crowed out and barriers to enter may be reinforced. This is not to the benefit of market competition. Today, no rating agency is based on the government-pays model. However, with the necessary resources, time, and alignment of all the governments, a high quality business model may be introduced (Alcubilla & Del Pozo, 2012).

3.3

Alternatives

Apart from the three models described above, other suggestions are possible too. One suggestion of the Financial Services Authority is to create a platform model.14 The platform would work as an intermediary between the issuer and the 13 European Commission, "Public Consultation on Credit Rating Agencies", November 2010, 28

. , , Occasional paper, November 2011, 2224. This platform model is also suggested in the following paper: J.MATHIS, J.MCANDREWS and J.C.ROCHET, “Rating the raters: Are reputation concerns powerful enough to discipline rating agencies?”, Journal of monetary economics 2009, 657- 674.

14 FENNELL D. and MEDVEDEV A., An economic analysis of credit rating agency business models and ratings accuracy, Financial Services Authority


rating agency. The issuer can not choose which CRA he wants to use, but it is the platform that selects a rating agency. The issuers pay the fees to the platform, and the platform pays them to the selected CRA. This way rating shopping is eliminated. However, a lot of issues may arise. For example, which CRA is chosen for which transaction? Will there only be one platform or multiple platforms? Does the fee has to be based on the quality of the rating or a flat fee? It is clear that there needs to be a lot of thinking and developing before this platform can be introduced.

Recently such a platform has been created. End 2011 "Wikirating" went online. Wikirating is an online credit rating platform whereby everyone can contribute. Any individual can issue ratings, based on a method described by Wikirating. An experts board provides their expertise to assess the methods. All input is published and can be modified by others. A forum is available to discuss the ratings. The aim is to provide a more reliable risk assessment (Greenwood, 2012). Momentarily there are 742 registered users and Wikirating aims at reaching a whole community of active users and even at being an alternative for the private agencies. As the table, illustrated in figure 2, indicates 15 , the ratings of S&P, Moody's, Fitch and Dagong tends to be higher, compared to the ratings of Wikirating. This could be an indication of the already mentioned criticism that ratings of the three big CRA's were downgraded too slowly in the years before the crisis. This because the CRA's were too lax and they wanted to please the issuers.

Another solution put forward by the European Commission is the payment-uponresults model. 16 Here, the fees of the rating agencies are based on future performance. If a bond they rated performs well, the agencies should receive a higher fee. This could be realized by putting a substantial proportion of the

15 Printscreen of a few countries, from www.wikirating.org/wiki/List_of_countries_by_credit_rating_-_comparisons. In total 198 countries are listed. Polling and SWI are 2 methods used by Wikirating for calculating the rating. The period of the polling runs from November 18th, 2011 - December 30th 2011, period for the ratings of S&P, Moody's, Fitch and Dagong runs from March to November 2011. 16 European Commission, "Public Consultation on Credit Rating Agencies", November 2010, 27.


payment into an escrow account. When the bond effectively performs well, the money can be returned to the CRA's. Rating agencies can also borrow from that fund so they can finance their activities. It is also required to publish the schemes on which the deferred payment is based. This way transparency is created and investors can determine themselves which scheme is preferred. But the downside of this model is the question who assesses the results and in what way (Alcubilla & Del Pozo, 2012). Rating agencies might have an incentive to manipulate their methodology to reach the desired result.

Section 939F of the Dodd Frank Act already required the Securities and Exchange Commission to evaluate alternative models. 17 It is clear that there are a lot of opportunities available to circumvent the conflicts of interest that are inherent in the issuer-pays model. But as described above, creating another compensation model would come with its own benefits and adversities and might introduce other conflicts of interest. It would also take a lot of effort, time and financial resources to switch to a complete different model.

17

U.S. Securities and Exchange

Commission, "Report to Congress on Assigned Credit Ratings, As Required by Section

939F of the Dodd-Frank Wall Street Reform and Consumer Protection Act", December 2012

, 83p.


4

Conclusion

We measured the effect of increased competition on the quality of credit ratings by analyzing the effect of the entrance of a third rating agency, Fitch. Results show that this coincides with lower overall quality, which is defined as an increase in the level of credit ratings and a decrease in informativeness of the ratings. However, we must note that this increase in level of credit ratings could not be exclusively assigned to the effect of increased competition. One must take into account that a credit rater is able to review and upgrade a credit rating when a firm is actually doing better. An alternative explanation for the increase in the level of credit ratings could be the fact that with the entrance of Fitch, the lowquality firms had an incentive to switch to the entrant to receive a higher rating and hence most of the high-quality firms are dedicated to the incumbents, which causes the level of ratings of those incumbents to increase. Our thesis is based on the research by Becker and Milbourne (2010), where they focused on North American firms, we enlarged the sample to worldwide industries. Our results were in line with those from Becker and Milbourne, which is that increased competition could not serve as a solution to increase the quality of credit ratings. We did a literature study to find alternative solutions to solve the previously mentioned conflicts of interest and came to the conclusion that there are alternatives which each have the power to solve the independency problem, but other side effects might appear. Stronger regulation concerning credit rating agencies can contribute to diminishing the conflicts of interest related to credit raters, especially when this regulation is harmonized on European level. Currently a proposal for a new European directive, proposal N° 2011/ 0361, is introduced but is not implemented yet. The evolution in the regulation concerning credit rating agencies is examined in more detail in the legal thesis of Anne Goor. An interesting continuation to our research could be to investigate the behavior of the clients and to consider whether they could be categorized based on their switching behavior. A possible outcome is that the low-quality firms have an incentive to switch to the entrant to receive a higher rating, leaving the incumbents with the high-quality firms, and hence the higher ratings.

CONCLUSION| 49

Bibliography

LEGISLATION

Article 3(1) (a) Regulation (EC) 1060/2009 of the European Parliament and of the Council of credit rating agencies

Regulation 2003/125/EG of the European parliament and of the board, as regards the fair presentation of investment recommendations and the disclosure of conflicts of interest

Regulation (EC) N° 1060/2009 of the European parliament and of the council of 16 September 2009 concerning credit rating agencies

Regulation (EU) N° 513/2011 an amendment to the CRA Regulation of 1 June 2011

Proposal for a regulation of the European parliament and of the council amending regulation (EC) N° 1060/2009 on credit rating agencies

The Technical Committee of the International Organization of Securities Commissions, Code of Conduct Fundamentals for Credit Rating Agencies, December 2004, 11p. Text books and collections ALCUBILLA, R. & DEL POZO, J.

(2012). Credit Rating Agencies on the Watch List, Analsis of European Regulation.

Oxford: University Press.

BRIGO, D. & MERCURIO, F. (2006). Interest rate models- theorie and practice: with smile inflation and credit. Berlijn: Springer.

BIBLIOGRAPHY | 51

CAPRIO, G. (2012). Handbook of Key Global Financial Markets, Institutions, and Infrastructure. New York: Academic Press. RICHARD, S. (2002). Rating, rating agencies and the global financial system. New York: Publisher US.

Contributions in journals ADELSON, M. (2006). Rating shopping, now the consequences. ASR Daily 2006, 12-30.

ALLEN, F., CARLETTI, E. & MARQUEZ, R. (2009).Credit market competition and Capital regulation. Oxford journals Economics and Social sciences, 983-1018.

BANNIER, C. & HIRSCH, C. (2010).The economic function of credit rating agencies: what does the watchlist tell us?. Journal of Banking & Finance, 3037-3049.

BECKER, B. & MILBOURNE, T. (2011).How did increased competition affect credit ratings?. Journal of Financial Economics,493-514.

BECKER ,B. & MILBOURNE, T. (2011).Reputation and competition: evidence from the credit rating industry. Journal of Financial Economics, 493-514.

BOOT, A. (2008).Credit ratingbureaus en de stabiliteit van de financiële sector. Maandblad voor accountancy en Bedrijfseconomie, 227-234.

BOOT, A. (2006).De toegevoegde waarde van credit ratings. the review of financiel studies, 11-18.

BRUYNINCKX, T. (2008).Rating agencies: inhoud, regelementering en aansprakelijkheid. Jura Falconis, 75-98.

BIBLIOGRAPHY | 52

CANTOR, R. & PACKER, F. (1994).The credit rating industry.Federal Reserve Bank of New York Journal Quarterly review, 1-26. DE HAAN, J. & AMTENBRINK, F. (2009).Regulating Credit Rating Agencies in the European Union: A Critical First Assessment of Regulation 1060/2009 on Credit Rating Agencies. Common market law review, 15–49.

DERYN, D. (2009).Credit Rating Agencies and the Credit Crisis: How the Issuer Pays Conflict Contributed and What Regulators Might Do about It. Columbia Business Law Review, 607- 633.

DONKERS, K. (2010).Credit rating agencies: informative-asymmetrie en civielrechtelijke aansprakelijkheid.Maandblad voor accountancy en bedrijfseconomie, 506-515.

FRIED, J. & HOWITT, P. (1980).Credit rationing and implicit contract theory.Journal of money, credit and banking, 463-471.

GERRITS, A. (2011).Amerikanen uit hun droom gewekt.De Telegraaf, 8-12. GREENWOOD, J. (2012).Move over S&P, Moody’s :make way for Wikirating. Financial Post

Geraadpleegd via:

, 12-38.

http://business.financialpost.com/2012/01/26/move-over-sp-moodys-make-way-for-

wikirating/.

HUNT, J.P. (2009).Credit Rating Agencies and the 'Worldwide Credit Crisis': The Limits of Reputation, the Insufficiency of Reform, and a Proposal for Improvement. Columbia Business Law Review, 109-209. JIANG, J. et al. (2012).Does it matter who pays for bond ratings? Historical evidence. Journal of Financial Economics, 607-621.Geraadpleegd via http://dx.doi.org/10.1016/j.jfineco.2012.04.001 JORION, P. (2005).Informational effects of regulation FD: Evidence from rating agencies.Journal of financial economics, 309-330.

BIBLIOGRAPHY | 53

KUDVA, R (2010).Issuer-pays model ensures ratings are available to the entire market. The economic times,6-8.

LEENAARS, H. (2003).Risicomanagement van banken. Maandblad voor accountancy en bedrijfseconomie, 346-347.

MATHIS, J., MCANDREWS, J. & ROCHET, J.C. (2009). Rating the raters: Are reputation concerns powerful enough to discipline rating agencies?.Journal of monetary economics, 657- 674.

PARTNOY, F. (1999).The Siskel and Ebert of Financial Markets? Two Thumbs Down for the Credit Rating Agencies.Washington University Law Quarterly 77, 619-712. PACKER, F. & CANTOR, R. (1995).The credit rating industry. The journal of fixed income, 10-34.

SOMMER, A.A. (1996).IOSCO, Its mission and achievement. Northwestern University School of Law Northwestern Journal of International Law & Business, 15-27.

SKRETA,V. & VELDKAMP, L. (2009).Ratings shopping and asset complexity: A theory of assets inflation. Journal of monetary economics, 678- 695. TANG, T. (2008).Information asymmetry and firms’ credit market access: Evidence from Moody’s credit rating format refinement. Journal of financial economics, 325-351.

WHITE, L.J. (2010).Markets: The credit rating agencies. Journal of economic perspectives, 211-226.

BIBLIOGRAPHY | 54

WHITE, L.J. (2009).The Credit-rating Agencies and the Subprime Debacle. Critical Review 21 no. 2-3, 389–399. WHITE, L.J. (2010). The Credit Rating Agencies: How Did We Get Here? Where Should We Go? . Journal of Economic Perspectives, 21-33. WOLFSON, J. & CRAWFORD, J. (2010).Lessons From The Current Financial Crisis: Should Credit Rating Agencies Be Re-Structured?. Journal of Business & Economics Research, 82-115. BAR-ISAAC, H. & SHAPIRO, J. (2010). Ratings quality over the business cycle (working paper NYU and University of Oxford), 30p. BECKER, B. & MILBOURNE, T.(2010).How did increased competition affect credit ratings? (Working paper Harvard Business School), 49p. CAMANHO, N., DEB, P. & LIU, Z. (2012).Credit rating and competition (Financial Markets Group London School of Economics and Political Science), 44p. COHEN, A. (2011).Rating shopping in the CMBS market (Federal Reserve board paper), 40p. DEB, G. & MURPHY, G. (2009).Credit Rating Agencies: An Alternative Model (London School of Economics and Financial Markets Group and Bank of England), 12p. DE HAAN, J.& AMTENBRINK, F. (2011).Credit rating agencies (De Nederlandsche Bank Working paper), 42p. European Commission. (2010).Public Consultation on Credit Rating Agencies (Working paper).

FENNELL, D & MEDVEDEV, A. (2011). An economic analysis of credit rating agency business models and ratings accuracy (Financial Services Authority

Occasional paper), 24p.

FROST, C.A. (2006). Credit rating agencies in capital markets: A review of research evidence on selected criticisms on the agencies (working paper University of North Texas) 45p. HAMILTON, D. (2004). Rating transitions and defaults conditional on watchlist, outlook and rating history (Working paper Moody’s analytics New York) 24p.

BIBLIOGRAPHY | 55

HILL, C. (2004).Regulating the rating agencies (American law and economics association annual meetings), 54p. HUANG, Z.,CHEN, H., HSU, C., CHEN, W. &WU, S. (2004).Credit rating analysis with support vector machines and neural networks: a market comparative study (working paper from the university of Arizona), 558p. MATHIS, J. (2009).Rating the raters: Are reputation concerns powerful enough to discipline rating agencies? (Working paper

Toulouse school of economics), 12p.

PARTNOY, F. (2006).How and why credit rating agencies are not like other gatekeepers (working paper University of San Diego School of Law), 45p. PRAGYAN, D. (2012). Essays on the impact of competition on financial intermediaries (working paper of the finance department of the London school of economics for the degree of doctor philosophy), 135p. SANGIORGY, F., SOKOBIN, J.& SPATT, C. (2009).Credit rating shopping, selection and the equilibrium structure of ratings (working paper Stockholm school of economics), 42p. SCHULTZ-LARSEN & T., HASLING KYED, J. (2012).The market for credit ratings: Competition and misalignment of interests (working paper department of economics, University of Copenhagen), 36p. SEABORN, P. (2010).Business models and incentives in rating markets: how 'who pays' matters (Working paper Rotman School of Management) ,18p. SWEENEY, E. (2010).An Examination of How Investor Needs are Served by Various Ratings Business Models: Ensuring Analytical Independence and Freedom from Conflicts of Interest at Credit Rating Firms (Standard's & Poors

paper,

New York)

White

,8p.

U.S. Securities and Exchange Commission (2003).Report on the Role and Function of Credit Rating Agencies in the Operation of the Securities Markets, As Required by Section 702(b) of the Sarbanes-Oxley Act of 2002 (working paper ) 45p.

,

U.S. Securities and Exchange Commission (2012). Report to Congress on Assigned Credit Ratings, As Required by Section 939F of the Dodd-Frank Wall Street Reform and Consumer Protection Act , 83p. U.S. Securities and Exchange Commission, Speech by SEC Commissioner: Remarks at the Commission Open Meeting, by Commissioner Kathleen L. Casey, 3 December 2008, http://www.sec.gov/news/speech/2008/spch120308klc.htm.

BIBLIOGRAPHY | 56

U.S Securities and Exchange Commission, Speech by SEC commissioner: Remarks to the institute of international bankers by SEC commissioner Paul S. Atkins, 3 maart 2008, www.sec.gov/news/speech/2008/spch030308psa.htm U.S Securities and Exchange Commission, Summary Report of Commission Staff’s Examinations of Each Nationally

Recognized Statistical Rating Organization, As Required by Section 15E(p)(3)(C) of the Securities Exchange Act of 1934, November 2012, 22p. Sites

http://www.moodys.com [27/02/2013] http://multiple-markets.com/3ratingschart.htm [12/11/2012] http://www.law.cornell.edu/cfr/text/17/240.15c3-1 [08/03/2013] http://www.sec.gov/news/speech/2008/spch030308psa.htm [08/03/2013] http://www.fitchratings.com [12/11/2012] http://www.standardandpoors.com [12/11/2012] http://www.wikirating.org [08/03/2013] www.wikirating.org/wiki/Wikirating:Stats [09/03/2013]

BIBLIOGRAPHY | 57

Attachments

Table 1: Fitch market share per NAICS industry in periods 1996-2000 and 2001-2006

NAICS Industry name 11 21 22 23 31 32 33 42 44 45 48 49 51 52 53 54 55 56 61 62 71 72 81 92

Agriculture, Forestry, Fishing and Huntin Mining, Quarrying, and Oil and Gas Extraction Utilities Construction Manufacturing: Food, Textile, Apparel Manufacturing: Wood, Paper, Printing, Petroleum, Chemicals, plastics Manufacturing: Metals, Machinery, Computers, Electrical, Furniture Wholesale Trade Retail Trade: Motor Vehicles, Furniture, Electronics, Food, Gas Retail Trade: Sporting goods, Books, Florists, Office Supplies, Mail-Order, Vending Transportation and Warehousing: Air Transport, Water Transport, Trucks, Pipelines Transportation and Warehousing: Messengers, Storage Information Finance and Insurance Real Estate and Rental and Leasing Professional, Scientific, and Technical Services Management of Companies and Enterprises Administrative and Support and Waste Management and Remediation Services Educational Services Health Care and Social Assistance Arts, Entertainment, and Recreation Accommodation and Food Services Other Services (except Public Administration) Public Administration

Fitch market share 1996-2000 2001-2006 0,0890909 0,228662 0,2155278 0,264468 0,2983342 0,3434 0,09058582 0,284794 0,16278246 0,304006 0,14286914

0,25898

0,19036782 0,13961388

0,275442 0,243745

0,08873534

0,275108

0,1773136

0,397773

0,110862 0,07326618 0,17411466 0,2926001 0,16730892 0,17948532 0,16053182

0,168662 0,19949 0,283829 0,3574 0,420202 0,26916 0,341544

0,1253776 0,12857142 0,18440676 0,14359684 0,13456064 0,1242837 0,15113262

0,294356 0,276974 0,269093 0,211729 0,303645 0,223923 0,360416

ATTACHEMENTS | 58

Table 2: Fitch market share per MSCI World industry in periods 1996-2000 and 2001- 2006

Industry

Fitch marketshare 1996-2000

2001-2006

Basic Materials

0,09066

0,135983

Communication

0,14306

0,290567

0,0812

0,114417

Consumer, non-cyclical

0,07452

0,124133

Diversified

0,07502

0,155033

Energy

0,0695

0,1549

Financial

0,2067

0,378517

Industrial

0,8132

0,255233

Technologie

0,06754

0,107033

Utilities

0,21396

0,270117

Government

0,23426

0,281733

Consumer, cyclical

Figure 1

ATTACHEMENTS | 59

Figure 2

from www.wikirating.org/wiki/List_of_countries_by_credit_rating_-_comparisons.

ATTACHEMENTS | 60

Samenvatting

Deze paper omvat een kritische analyse omtrent de rol die credit rating agentschappen spelen in de financiële wereld. Hierbij zal dieper ingegaan worden op de bijdragen en het belang van credit ratings bij het tot stand komen van het minimum kapitaal van financiële kredietinstellingen. Vervolgens zullen de meest voorkomende

problemen

die

geassocieerd

worden

met

credit

rating

agentschappen besproken worden. Als gevolg hiervan zal onderzocht worden of de huidige regulering inzake credit rating agentschappen een oplossing kan bieden om tegemoet te komen aan deze conflicten. Tot slot zal ook het Europese voorstel tot een vernieuwde regulering aan bod komen en de mate waarin deze bijdraagt aan het tot stand komen van een efficiente rating markt. De paper zal besloten worden met een algemeen besluit.

Inhoudsopgave

Inhoudsopgave ................................................................................................................. iii Lijst van afkortingen ........................................................................................................ iv Inleiding.............................................................................................................................. 5 Tekortkomingen credit rating agentschappen ............................................................... 8 1.1 1.2 1.3

1.4 2

Minimum kapitaal vereisten voor financiële instellingen ................................. 13 2.1

3

Berekening minimumkapitaal ....................................................................... 13 2.1.1 Interne rating systeem ........................................................................ 14 2.1.2 Externe rating systeem....................................................................... 15

Evolutie in de regulering ...................................................................................... 16 3.1 3.2

3.3 3.4 4

Gebrekkige ratingmethoden van ratinginstituten ............................................ 8 Exclusiviteit van de ratingagentschappen ...................................................... 9 Gebrek aan onafhankelijkheid van de CRA’s ............................................... 10 1.3.1 Credit rating agentschap oefent druk uit op de onderneming ............ 10 1.3.2 De onderneming oefent druk uit op het credit rating agentschap ...... 11 Geen civielrechtelijke aansprakelijkheid ....................................................... 11

IOSCO-gedragscode .................................................................................... 16 Nood aan strengere regulering? ................................................................... 18 3.2.1 Capital Requirements Directive .......................................................... 18 3.2.2 Market Abuse Directive ...................................................................... 19 Europese regulering inzake credit rating agentschappen ............................ 19 Voorstel tot vernieuwde Europese verordening ........................................... 24

Besluit .................................................................................................................... 28

Bibliografie....................................................................................................................... 29

INHOUDSOPGAVE | III

Lijst van afkortingen

SEC

= Securities and Exchange Commission

NRSRO

= Nationally Recognized Statistical Ratings Organizations

CRA

= Credit Rating Agency

IOSCO

= International Organization of Securities Commissions

CRD

= Capital Requirements Directive

BIS

= Bank for International Settlements

RWA

= Risk Weighted Asset

CESR

= Committee of European Securities Regulators

MiFID

= Markets in Financial Instruments Directive

FSAP

= Financial Services Action Plan

ESMA

= Europese Autoriteit voor Effecten en Markten

EURIX

= European Rating Index

BW

= Burgerlijk Wetboek

LIJST VAN AFKORTINGEN | IV

Inleiding

In deze master paper zal onderzoek gedaan worden naar de rol van externe rating bureaus bij de berekening van minimum kapitaalvereisten voor financiële instellingen. Vanwege de grote hoeveelheden risicovol passief dat een financiële instelling dagelijks aanhoudt en dat vaak op korte termijn moet terugbetaald worden, is het essentieel om vanuit solvabiliteitsoogpunt aan deze entiteit bepaalde minimumvereisten inzake kapitaal op te leggen.Aan de hand van ratings wordt bepaald hoe risicovol een bepaalde transactie is, hierbij wordt onder meer rekening gehouden met de solvabiliteit van de kredietnemer. Sinds de introductie van Basel II, kan een financiële instelling opteren om deze berekeningen intern te doen dan wel beroep te doen op een extern gegev rating. Externe credit rating agentschappen (later in de paper cra’s genoemd) zijn ondernemingen die op zelfstandige basis en tegen betaling de kredietwaardigheid van een entiteit of een transactie berekenenen en hier een rating voor bepalen. Aangezien het niveau van zulke ratings van groot belang is voor de bepaling van het minimum kapitaal, is het essentieel om na te gaan of er eventueel preverse incentieven optreden bij het gebruik van externe credit ratings en of er een verhoogde regulering nodig is om hieraan tegemoet te komen. In het verleden vertrouwde men erop dat de betalingen die de cra’s van de issuers ontvingen in ruil voor hun ratings en die hen konden stimuleren om te raten in het voordeel van de issuer niet aanzienlijk genoeg waren om hen ook daadwerkelijk die incentieven te laten volgen. Men dacht echter dat het hebben van een goede reputatie en een waarheidsgetrouw beeld naar de buitenwereld, van groter belang waren voor de cra’s. 18 Deze visie wordt ook wel de ‘reputational capital view’ genoemd.19 Hierbij is het van groot belang dat de rater bepaalde doelstellingen naleeft om zo het vertrouwen van de clienten voor zich te kunnen winnen. Verder zal het belangrijk zijn dat de cra aangeeft dat ze een meerwaarde kan betekenen en 18

A. W.A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financiel studies

2006, 18-20. 19

D. KERWER, “Holding Global Regulators Accountable: The Case of Credit Rating Agencies”, An international journal of policy, administration and institutions 2005, 453-475.

INLEIDING | 5

dat ze dus toegang heeft tot informatie die de issuer van die informatie niet heeft. 20

Dit alles zou volgens de reputational capital view de ideale basis zijn voor een

zelfregulerend systeem want wanneer clienten het gevoel niet meer hebben dat de cra waarheidsgetrouwe ratings oplevert, zullen ze geen geloof meer hebben in hun waarderingen en zullen ze minder willen beleggen in die gewaardeerde producten. 21 Hierdoor zal een rating van zulke cra geen meerwaarde meer opleveren voor de issuer en zal hij niet langer bereid zijn te betalen voor hun diensten.22 Deze benadering gaat er dus van uit dat de kwaliteit van een rating gewaarborgd zal worden door het potentiële verlies van reputatie eerder dan door een verhoogde mate van regulering. Nochtans heeft het gebruik van externe ratings in het verleden een enorm grote invloed gehad en een aanzienlijke rol gespeeld in het verloop van de financiële crisis. Die is namelijk in hoofdzaak ontstaan doordat banken handelden en transacties verrichtten zonder dat ze hiervoor een voldoende buffer hadden. Hun reserves waren onvoldoende om te voldoen aan de opvragingen van geld. Dit kwam doordat credit raters vaak een vertekend beeld van de werkelijkheid gaven als gevolg van de perverse incentieven die gepaard gaan met het gebruik van cra’s. Zo ontstond de nood om de regulering van cra’s ook op Europees niveau te optimaliseren. Deze master thesis zal een analyse inhouden van de tekortkomingen van cra’s enerzijds en de oplossingen die hiervoor geboden worden in de Europese regulering anderzijds. De paper bouwt voort op de economische paper, waarin verhoogde regulering als een alternatief gezien werd om tegemoet te komen aan de conflicts of interest met betrekking tot cra’s. De onderzoeksvraag die hierbij aan bod zal komen luidt als volgt: ‘Heeft het gebruik van externe credit rating agentschappen een toegevoegde waarde bij de berekening van het minimum kapitaal of treden er perverse incentieven op? En zo ja, kan hiervoor een oplossing 20

F. PARTNOY, “The Siskel and Ebert of Financial Markets? Two Thumbs Down for the Credit Rating Agencies.”, Washington University Law Quarterly 77 1999, 619-712. 21 J. MATHIS, J. MCANDREWS, J.C. ROCHET, “Rating the raters: Are reputation concerns powerful enough to discipline rating agencies?”, Journal of monetary economics 2009, 657- 674. 22 J.P. HUNT, “Credit Rating Agencies and the 'Worldwide Credit Crisis': The Limits of Reputation, the Insufficiency of Reform, and a Proposal for Improvement”, Columbia Business Law Review 2009, 109-209.

INLEIDING | 6

gevonden worden in de Europese regulering?’ Bij het onderzoek naar deze vraag zal eerst besproken worden met welke tekortkomingen een cra geconfronteerd wordt. Vervolgens zal de werking van externe rating kantoren verder worden toegelicht en zal deze methode vergeleken worden met een intern rating systeem. Verder zal de paper een overzicht weergeven van de evulotie in de regulering van cra’s als antwoord op de tekortkomingen.

INLEIDING | 7

Tekortkomingen credit rating agentschappen

Cra’s zijn ondernemingen die op zelfstandige basis ratings verlenen aan andere ondernemingen. 23 Hierbij wordt een extern rating systeem toegepast waarbij de credit agentschappen tegen betaling van de kredietnemer,de kredietwaardigheid van deze entiteit of het risicoprofiel van een financiële transactie berekenen.24 Hierbij komen echter vaak allerlei problemen kijken die het getrouwheidsgehalte van de ratings naar beneden halen.25 Enkele belangrijke

belangenconflicten

zullen

hier

verder

besproken

worden:

Gebrekkige

ratingmethoden van ratinginstituten (1.1), de exclusiviteit van de ratingagentschappen (1.2), gebrek aan onafhankelijkheid van CRA’s

(1.3), geen civielrechtelijke aansprakelijkheid

(1.4).26 Hierbij moet vooral aandacht besteed worden aan de drievoudige relatie tussen de emittent of uitgever, de belegger en de credit rater. De emittent vraagt de diensten van de credit rater in ruil voor betaling en de belegger doet beroep op de ratings die de credit rater heeft gegenereerd bij haar beslissing om te beleggen. Tegelijkertijd moet echter in acht genomen worden dat de reputatie voor een CRA uiterst belangrijk is en dat dit grotendeels een rem zou moeten zetten op deze perverse incentieven.27

4.1

Gebrekkige ratingmethoden van ratinginstituten

De oligopoliepositie die CRA’s verkregen dankzij het gebrek aan voldoende concurrentie, gaf aanleiding tot een passieve houding ten opzichten van veranderende marktomstandigheden. 28

De agentschappen speelden niet langer in op die veranderende situtatie door haar modellen

en methodes niet, of vertraagd aan te passen. Dit zorgde ervoor dat de kwaliteit van de ratings 23

S. RICHARD, Rating, rating agencies and the global financial system, New York, Publisher US, 2002, 39-41. Artikel 3(1) (a) Regulation (EC) 1060/2009 of the European Parliament and of the Council of credit rating agencies 25 T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 7598. 26 A. BOOT “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 11-18. 27 J. MATHIS, J. MCANDREWS, J.C. ROCHET, “Rating the raters: Are reputation concerns powerful enough to discipline rating agencies?”, Journal of monetary economics 2009, 657- 674. 28 R. CANTOR and F. PACKER , “The credit rating industry”, Federal Reserve Bank of New York Journal Quarterly review 1994, 1-26. 24

TEKORTKOMINGEN CREDIT RATING AGENTSCHAPPEN | 8

stapsgewijs afbrokkelde.29 Zo zal de nauwkeurige zorgvuldigheid bij het controleren van de objectiviteit van beschikbare informatie in de praktijk niet aanwezig zijn, dit louter en alleen om het feit dat het niet tot de wettelijke verplichtingen hoort van rating agentschappen om hun informatie te verifiëren.30 Ook zal omwille van het gebrek aan innovatie in de modellen die de CRA’s hanteren, een overmatig beroep worden gedaan op historische informatie die vaak niet meer up to date is en een vertekend beeld van de realiteit geeft.31

4.2

Exclusiviteit van de ratingagentschappen

De exclusiviteit of de beperkte mate van concurrentie op de markt van ratingagentschappen is grotendeels te wijten aan de hoge toetredingsbarrières.32 Om te beginnen moet gezegd worden dat niet zomaar iedereen kan toetreden, de opstartkosten zullen aanzienlijk hoog zijn, 33 bedragen die niet zomaar voor handen zijn.

34

Verder zal het behouden van een goede

reputatie, door een zo waarheidsgetrouw mogelijk beeld te geven, veel inspanningen vragen.35 Dit alles zorgt voor een uiterst geconcentreerd beeld van de ratingindustrie en weerspiegelt zich in het marktaandeel, 90%, van de drie belangrijkste CRA’s: Moody’s, Fitch en Standard & Poor’s. 36 Deze oligopolie positie zorgt ervoor dat cra’s in mindere mate kwaliteit van hun ratings gaan nastreven. De kans dat een onderneming voor de ratingdiensten naar een andere cra zal gaan is immers toch beperkt, waarom dan inspanningen leveren om de kwaliteit op pijl te houden? 37 Wanneer de nodige aanpassingen niet gebeuren, zal automatisch de kwaliteit

29

A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 12-18. T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 7598. 30

31

NEIJS, P., De toegevoegde waarde van ratinginstituten ter discussie, working paper universiteit van Tilburg, 2010, 75p.

32

A. BOOT, “Credit ratingbureaus en de stabiliteit van de financiële sector”, Maandblad voor accountancy en Bedrijfseconomie 2008, 227-234.

33


34

A. BOOT, “credit ratingbureaus en de stabiliteit van de financiële sector”, the economist 2008, 63-72. PARTNOY, F., How and why credit rating agencies are not like other gatekeepers, University of San Diego School of Law, 2006, 45p. 36 SMANT, D., VAN DER ENT, J.P., Credit rating en kapitaalmarkt: Feiten,verklaringen en ontwikkelingen van rating in Europa, Financiele en Monetaire Studies Rotterdam, 1994, 70p. 37 EUROPEAN COMMISSION, Public consultation on credit rating agencies, 2010, 32p. 35

TEKORTKOMINGEN CREDIT RATING AGENTSCHAPPEN | 9

drastisch achteruit gaan, 38 dit zal contradictioneel geen effect hebben op de vraag naar de diensten omwille van de oligopolie structuur van de cra’s.39

4.3

Gebrek aan onafhankelijkheid van de CRA’s

Een idealse situatie zou zijn dat een cra een onafhankelijke onderneming was die niet gebonden was door haar opdrachtgever, de emittent. In de praktijk is dit echter vaak niet het geval. Volgens het issuer pays model, is het de emittent die de cra’s uitbetaalt. Dit kan aanleiding geven tot allerlei tegengestelde belangen.40 Verder in deze paper zullen enkele van die belangenconfilicten besproken worden. Zo zal de functie van rating agentschappen uitgebreid worden naar een adviserende rol bij complexe financiële instrumenten. Ondernemingen zullen naast het raten, ook beroep gaan doen op het advies van CRA’s bij de aanwending van gecompliceerde en erg technische instrumenten, dit zorgt voor een extra inkomstenbron voor de cra’s.41 omwille van deze extra inkomst die de agentschappen niet willen verliezen, zullen ze er alles aan doen om hun client tevreden te stellen.42 Een tevreden client is een client die een hoge rating ontvangt, dit zal aanleiding geven tot ratings die de waarheidstoets niet altijd doorstaan43 en eerder naar boven herzien zijn geworden.44

4.3.1

Credit rating agentschap oefent druk uit op de onderneming

De cra zal de mogelijkheid hebben om de onderneming onder druk te zetten doordat zij de exclusiviteit heeft om de rating van de financiële producten te bepalen.

45

Wanneer de

46

onderneming niet voldoende biedt in ruil voor deze diensten, heeft de cra de optie om te dreigen de rating op een negatieve wijze te herzien waardoor het financiële product lager 38


39

L. WHITE, “The credit rating agencies”, Journal of economic perspectives 2010, 211-226. R. KUDPA, “issuer-pays model ensures ratings are available to the entire market”, The economic times 2010,68. 41 K. DONKERS, P.J.W. DUFFHEUS EN W.C.T. WETERINGS, " Credit Rating Agencies: Informatie-asymmetrie en civiele aansprakelijkheid", Maandblad voor Accountancy en Bedrijfseconomie 2010, 506-515. 42 R. CANTOR and F. PACKER, “The credit rating industry”, Federal Reserve Bank of New York Journal Quarterly review 1994, 1-26. 40

43


44

COVITZ, D., HARRISON, P., Testing conflicts of interest at bond rating agencies with market anticipation: evidence that reputation incentives dominate, working paper, Division of Research and Statistics Washington, 2003, 37p. 45 A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 11-18. 46

NEIJS, P., De toegevoegde waarde van ratinginstituten ter discussie, working paper universiteit van Tilburg, 2010, 75p. TEKORTKOMINGEN CREDIT RATING AGENTSCHAPPEN | 10

gewaardeerd wordt dan de echte waarde. In de studie van Goh en Ederingtion (1993) is onderzoek gedaan naar het effect van een downgrade. Dit is het naar beneden herzien van een rating. Resultaten duiden aan dat het effect op het beleggingsgedrag aanzienlijk negatief is bij een downgrade, in tegenstelling tot de effecten van een upgrade die nauwlijks merkbaar zijn. 47

4.3.2

De onderneming oefent druk uit op het credit rating agentschap

Niet enkel de CRA, maar omgekeerd kan ook de onderneming druk uitoefenen op de ratingagentschap. Zo zal de uitbetaling die een ratingagentschap ontvangt voor haar diensten, afhankelijk kunnen zijn van de grootte van de emittent die de diensten aanvraagt en betaalt.48 Op deze manier kunnen grotere ondernemingen een druk leggen op de schouders van de agentschappen omdat ze bij een ongevraagde negatieve rating de kans lopen om de uitkering te verliezen en zal er in de toekomst op hen geen beroep meer gedaan worden.49

Deze tegengestelde belangen van de onderneming en anderzijds de CRA zorgen voor een aanzienlijke vermindering van de kwaliteit.50De CRA is voor haar ratings en dus ook voor haar financiële inkomsten immers afhankelijk van de emittent, die eveneens de betaler is.51 Het is in het belang van de emittent dat de rating zo hoog mogelijk is 52 om zoveel mogelijk beleggers te kunnen aantrekken.53 De CRA heeft er dan weer alle belang bij dat de rating zo waarheidsgetrouw mogelijk is, zodat dit kan worden toegeschreven aan haar goede reputatie.

4.4

Geen civielrechtelijke aansprakelijkheid

De reputational capital view benadering heeft doorheen de jaren moeten wijken voor een meer strikte regulering. Hierbij werden de toetredingsbarrières verhoogd, waardoor de ratingmarkt 47

L. CRABBE, M. POST, “The effect of a rating downgrade on outstanding commercial paper”, The journal of finance 2012, 39-56. 48 R. KUDPA, “issuer-pays model ensures ratings are available to the entire market”, The economic times 2010,68. 49 F. PARTNOY, “The Siskel and Ebert of Financial Markets? Two Thumbs Down for the Credit Rating Agencies.”, Washington University Law Quarterly 77 1999, 619-712. 50 K. BHANOT, A. MELLO, “Should corporate debt include a rating trigger?, Journal of financial economics 2006, 69-98. 51 STROBL, G., XIA, H., The issuer pays rating model and ratings inflation: evidence from corporate credit ratings, working paper Business School University of North Carolina, 2011, 41p. 52 A. BOOT., “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 11-18. 53 R. KUDPA, “issuer-pays model ensures ratings are available to the entire market”, The economic times 2010,68. TEKORTKOMINGEN CREDIT RATING AGENTSCHAPPEN | 11

een oligopolische vorm kreeg en de drie grote cra’s zo een toezichtrechtelijke afhankelijkheid konden verwerven. 54Dit verleende aan cra’s de exclusiviteit om vergunningen inzake handel van financiële instrumenten te verkopen.55 Het is net die oligopolie positie die ertoe geleid heeft dat de kwaliteit van de ratings van afnemend belang was 56 , louter en alleen omdat zij de enigen waren waarbij zulke vergunningen konden verkregen worden.57 Het logische gevolg was dan ook dat de reputational capital view vanaf toen niet meer opging en dat er nood was aan een striktere regulering. 58 Een mogelijke verklaring hiervoor is dat de civielrechtelijke aansprakelijkheid ervoor zou zorgen dat CRA’s de kwaliteit van hun ratings zouden garanderen om zo civielrechtelijke claims te vermijden. 59 In de praktijk was dit echter helemaal anders, de aansprakelijkheidsmogelijkheden van ratingbureaus zijn namelijk zeer beperkt en er is zelfs de mogelijkheid tot civielrechtelijke immuniteit van de ratingbureaus. 60 Een verdere uiteenzetting van deze aansprakelijkheid valt echter buiten het bestek van deze thesis.

54

T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2007-2008, 75-98. 55 NEIJS, P., De toegevoegde waarde van ratinginstituten ter discussie, working paper universiteit van Tilburg, 2010, 75p. 56 A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 9-18. 57 www.law.yale.edu/documents/pdf/cbl/Partnoy_Overdependence_Credit.pdf 58 SMANT, D., VAN DER ENT, J.P., Credit rating en kapitaalmarkt: Feiten,verklaringen en ontwikkelingen van rating in Europa, Financiele en Monetaire Studies Rotterdam, 1994, 70p. 59 T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 7598. 60 K. DONKERS, P.J.W. DUFFHEUS en W.C.T. WETERINGS, " Credit Rating Agencies: Informatie-asymmetrie en civiele aansprakelijkheid", Maandblad voor Accountancy en Bedrijfseconomie 2010, 506-515. TEKORTKOMINGEN CREDIT RATING AGENTSCHAPPEN | 12

5

Minimum kapitaal vereisten voor financiële instellingen

5.1

Berekening minimumkapitaal

Voor het toezicht op de solvabiliteit van banken is het van groot belang dat de risico’s waaraan banken dagelijks blootgesteld worden, gemanaged worden.

61

De kapitaalstructuur

kan hier een belangrijke rol spelen door extra reserves aan te leggen per risicodragende transactie die wordt uitgevoerd. Dit aangehouden kapitaal moet functioneren als een buffer die de bank moet dekken tegen insolvabiliteit bij geldopvragingen. 62 Deze minimum kapitaalvereisten zijn richtlijnen en voorwaarden die aan banken worden opgelegd met als hoofddoel de bescherming van deposito’s van het cliënteel. De hoeveelheid minimum kapitaal zal bepaald worden in functie van het risicogehalte van de financiële instelling.63 Om ervoor te zorgen dat de regulering van de kapitaalstructuur gevolgd wordt door alle banken wereldwijd, werd de Basel Commissie voor het internationaal toezicht op financiële instellingen opgericht en kwam er in 1988 het eerste Basel-akkoord. 64 Deze richtlijn die diverse standaarden en solvabiliteitsvereisten oplegt, wordt gevolgd door 100 landen wereldwijd.65 Binnen deze richtlijnen van Basel, wordt aan banken de optie gegeven om voor de berekening van hun minimum kapitaal, beroep te doen op interne dan wel externe ratings op basis van het risicoprofiel. De Capital Requirement Directive (CRD) dateert van 14 juni 2006 en is een implementatie van de aanbevelingen die in de Basel akkoorden werden gegeven in Europese Unie wetgeving. De richtlijn is gericht op de eenmaking van de interne markt inzake financiële diensten.66 Zo zal er gestreefd worden naar een harmonisering van de belangrijkste wetgeving inzake kredietsinstellingen. 67 Gezien de centrale rol die minimum kapitaal speelt binnen de functionering van financiële instellingen, is de harmonisering van deze minimum vereisten van cruciaal belang. De hoeveelheid kapitaal dat een financiële

61

H. LEENAARS, “Risicomanagement van banken”, Maandblad voor Accountancy en Bedrijfseconomie 2003, 28. 62 C.A. HAZEU, “De crisis, de bank en de wet”, Tijdschrift voor openbare financiën 2011, 9-11. 63 TAYTAS, F., Basel II en de gevolgen voor het management van de banken, working paper of the University of Twente, 2008, 28p. 64 VAN ROY, P., The impact of the 1988 Basel Accord on banks’ capital ratios and credit risk taking : an international study, working paper National Bank of Belgium, 2005, 12p. 65 H.A. BENINK, “Issues inzake Basel II”, Maandblad voor accountancy en bedrijfseconomie (MAB) 2005, 33-40. 66 DIRECTIVE 2006/48/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 14 June 2006 relating to the taking up and pursuit of the business of credit institutions 67 DIRECTIVE 2006/48/EC OF THE EUROPEAN PARLIAMENT AND OF THE COUNCIL of 14 June 2006 relating to the taking up and pursuit of the business of credit institutions MINIMUM KAPITAAL VEREISTEN VOOR FINANCIELE INSTELLINGEN | 13

instelling verplicht is om aan te houden zal bepaald worden in functie van het risicoprofiel van deze instelling.68 Dit risicoprofiel is afhankelijk van de rating die aan deze entiteit werd gegeven. Zoals eerder reeds vermeld kan een financiële instelling voor de berekening van het minimum kapitaal zowel gebruik maken van haar eigen inschattingen om te komen tot interne ratings, dan wel beroep doen op externe ratings, uitgegeven door erkende rating agentschappen. Beide methodes zullen verder in de paper uitgebreid besproken worden.

5.1.1

Interne rating systeem

Wanneer men de ‘Internal rating based approach’ (IRB) van Basel II hanteert, wordt aan financiële instellingen de mogelijkheid geboden om gebruik te maken van hun eigen interne maatstaven en systemen voor het bepalen van een rating.69 Bij dit systeem is het de bedoeling dat de financiële instellingen zelf het kredietrisico van de leningportefeuilles gaan berekenen en de kapitaalhoeveelheid die ze hiervoor moeten aanhouden hiervan laten afhangen. 70Artikel 84 van de Capital Requirement Directive bepaalt dat deze interne systemen van classificatie die de financiële instellingen zelf zullen ontwikkelen, zullen moeten gevalideerd worden door de toezichthouders.71 72Enkel financiële instellingen die aan een aantal minimum voorwaarden voldoen en die toestemming hebben gekregen van de nationale toezichthouder komen in aanmerking voor de toepassing van het IRB systeem.73 De financiële instelling zal haar activa groeperen in verschillende subcategorieën van haar vermogen, zoals uiteengezet in art 86 van de richtlijn en vervolgens aan elke categorie een gewicht toekennen afhankelijk van het risico dat eraan verbonden is. 74 Hierdoor wordt het RWA “Risk Weighted Asset” gevormd dat de basis uitmaakt van de berekening.

7576

Het

verplicht aan te houden minimum kapitaal wordt vervolgens bepaald door een bepaald vast percentage te nemen van het RWA, meestal 8%. 77

68

H. LEENAARS, “Risicomanagement van banken”, Maandblad voor Accountancy en Bedrijfseconomie 2003, 28. 69 HOLMQUIST, J., Implementation of Basel II: Challenges and opportunities, Institute of International Bankers 2007, 21p. 70 stats.oecd.org 71 www.fsa.gov.uk/pubs/international/exposures.pdf 72 Art. 84 CRD 73 M. GORDY, “ A risk factor model foundation for ratings-based bank capital rules”, Journal of financial intermediation 2003, 199-232. 74 Art 86 CRD 75 H.A. BENINK, “Issues inzake Basel II”, Maandblad voor accountancy en bedrijfseconomie (MAB) 2005, 3340. 76 DAS, S., SY, A., How risky are banks’ risk weighted assets? Evidence from the financial crisis, IMF working paper Cornell University, Department of economics, 2012, 38p. 77 Art. 87 CRD MINIMUM KAPITAAL VEREISTEN VOOR FINANCIELE INSTELLINGEN | 14

5.1.2

Externe rating systeem

De financiële instelling kan ook opteren om voor de waardering van financiële instrumenten en transacties gebruik te maken van externe ratings, uitgegeven door externe cra’s. 78 De kredietwaardigheid en de kans op ‘default’ worden berekend en er wordt een rating code gegeven die reikt van AAA (beste kwaliteit) tot D (zeer slechte kwaliteit, kleine kans op terugbetaling) 79 . Hiertoe voeren ze een financiële analyse

80

uit waarbij zowel actuele als

historische gegevens worden gebruikt om te bepalen hoe kredietwaardig de onderzochte entiteit is. 81 De kredietwaardig van een investering wordt bepaald door de grootte van de rating.82 Deze methode wordt ook wel ‘Standardized approach method’ genoemd aangezien er gewerkt wordt met standaard ratings opgesteld door de BIS, een samenwerkingsverband genaamd ‘Bank for International Settlements’.83

78

Art 80-81 CRD Waarbij in acht moet genomen worden dat de ratingschaal afhankelijk is van het credit rating agentschap dat de ratings uitgeeft. 80 KRONWALD, C., Credit rating and the impact on corporate structure, working paper Duitsland 2009, 27p. 81 T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 7598. 82 OSTER, J., The Commission Proposal to Amend the Regulation on Credit Rating Agencies, Working paper on recent legal developments, 2012, 6p. 83 CARPENTER, S., WHITESELL, W., ZAKRAJ, E., Capital Requirements, Business Loans, and Business Cycles: An Empirical Analysis of the Standardized Approach in the New Basel Capital Accord , working paper of the Board of governors of the federal reserve system, 2001, 48p. 79

MINIMUM KAPITAAL VEREISTEN VOOR FINANCIELE INSTELLINGEN | 15

6

Evolutie in de regulering

Het toegenomen belang dat sinds de intreding van de Standardized approach methode werd gegeven aan het gebruik van cra’s, heeft aanleiding gegeven tot veel discussie en een verhoogde aandacht van de nationale toezichtsorganen. Zo is veel onderzoek gedaan naar hoe het gebruik van zulke cra’s meer efficiënt kon gemaakt worden. 84 Dit mede omwille van het feit dat banken nu zelf kunnen bepalen op welke credit rater ze beroep zullen doen, aanleiding geeft tot een verhoging van de impact van de eerder aangehaalde tekortkomingen waarmee cra’s te kampen hebben.

6.1

IOSCO-gedragscode

In 2004 nam de commissie de beslissing om de regelgeving met betrekking tot cra’s strenger te maken, dit door een gedragscode op te leggen aan rating agencies die actief zijn op het Europese grondgebied, genaamd IOSCO gedragscode.

85

Dit staat voor Internatonal

Organisation of Securities Commissons en heeft als hoofddoel credit ratings meer waarheidsgetrouw te maken en het vertrouwen in de cra’s te verbeteren. De gedragscode legt een aantal hooggegrepen doelstellingen op die credit agencies moeten naleven bij het uitvoeren van de ratings om zowel potentiële investeerders te beschermen, transparantie te verbeteren als de risico’s te verlagen.86 De gedragscode was bedoeld om het zelf-regulerend gedrag van de credit rating agencies verder uit te bouwen en meer efficiënt te maken.87 Ook werd het op deze manier mogelijk gemaakt om credit raters te behoeden voor perverse incentieven en andere factoren die de analyse en de ratings op een negatieve manier kunnen beïnvloeden. 88 De gedragscode was gebaseerd op een vrijwillige naleving door de rating agencies, waarbij de commissie de evolutie nauw in het oog hield en zich openstelde voor een een nieuwe regelgeving wanneer zou blijken dat de vrijwillige gedragscode niet voldoende

84

A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 11-18. Code of Conduct Fundamentals for Credit Rating Agencies, The Technical Committee of the International Organization of Securities Commissions, December 2004 86 A.SOMMER, “IOSCO, Its mission and achievement”, Northwestern Journal of International Law & Business 1996, 15-27. 87 www.iosco.org 88 STEELE, S.,SECURITIES AND EXCHANGE COMMISSION, A review of implementation of the IOSCO code of conduct fundamentals for credit rating agencies, paper of the technical committee of the international organization of securities commissions 2009, 17p. 85

EVOLUTIE IN DE REGULERING | 16

zou zijn.89Die vrijwillige gedragscode was gekenmerkt door het ‘comply or explain’- principe waarbij rating agencies verplicht werden om hun gedragscode publiek bekend te maken en moesten aangeven hoe hun code de IOSOC-principes met betrekking tot cra’s implementeert. Indien dit niet het geval was, waren de CRA’s verplicht om uit te leggen hoe en waarom hun gedragscode afwijkt van de IOSCO – principes.90 De bedoeling was om doelstellingen op te leggen die het marktvertrouwen en de bescherming van beleggers zou garanderen door te zorgen voor kwalitatieve ratings die niet beïnvloed waren door perverse incentieven van de credit raters. De functie van CRA’s moest door deze regel op verschillende niveau’s verbeterd worden.91 Een eerste aanbeveling van het IOSCO betreft de kwaliteit en integriteit van het rating process, dit zou geïmplementeerd, geformaliseerd en vooral gehandhaafd worden om ervoor te zorgen dat er een uitgebreide analyse plaatsvindt van de relevante informatie.92 Wanneer een rating gegeven wordt aan de issuer, mag hierbij nooit een garantie gegeven worden dat de bevinding een volledige weerspiegeling is van de waarheid, er moet immers op een eerlijke en zorgvuldige

manier

met

de

contractanten

worden

omgegaan.

93

94

Eventuele

belangenconflicten moeten vroegtijdig worden opgespoord en vervolgens zo snel mogelijk geneutraliseerd worden. De IOSCO gedragscode moest een oplossing zijn voor de verminderde kwaliteit van de ratings en de almaar toenemende belangenconflicten. Die poging is achteraf echter niet efficiënt gebleken, grotendeels omwile van het gebrek aan handhavingsmechanismen. Hierdoor is de gedragscode niet omgezet in een afdwingbare regelgeving maar heeft ze louter dienst gedaan als richtlijn bij het genereren van interne regels.95Hoewel de regels niet afdwingbaar zijn, hadden ze toch tot effect dat het vertrouwen in de rating agentschappen toenam. Dit kan verklaard worden in het licht van de reputational capital view, die eerder in deze paper aan bod kwam.96

89

AMADOU, N., The systemic regulation of credit rating agencies and rated markets, Senegal, 2009, 36p. HELLEINER, E., Global finance in crisis, the politics of international regulatory change, Canada, Routledge, 2010, 194p. 91 www.europa-nu.nl 92 A. 1.1-1.8 IOSCO code of conduct fundamentals for credit rating agencies. 93 N.E.D. FABER ,De kredietcrisis,Nederland, Kluwer, 2010, 246. 94 www.fsma.be 95 Werkdocument van de commissie bij het voorstel voor een verordening van het Europees Parlement en de Raad inzake credit rating agencies, SEC(2008), 9. 96 T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 7598. 90


6.2

Nood aan strengere regulering?

In het Resolutieplan van 2004 werd door het Europese Parlement het belang en de verantwoordelijkheid van credit rating agentschappen aangehaald en de vraag gesteld naar de eventuele noodzaak voor strengere regulering.97 Dit gebeurde mede naar aanleiding van het Enron schandaal in 2001. Een onderzoek naar de mate van regulering die aan credit raters wordt opgelegd, werd gevoerd door de committee of European Securities Regulators (CESR). 98

De Europese commissie was er echter van overtuigd dat de regulering van credit rating

agentschappen voldoende gedekt werd door enerzijds het zelfregulerende effect van de IOSCO gedragscode en anderzijds de drie richtlijnen inzake financiële diensten, namelijk de Market Abuse Directive 99 , de Capital Requirements Directive (CRD) 100 en de Markets in Financial Instruments Directive (MiFID)101 die voortvloeien uit het Financial Services Action Plan (FSAP) van de commissie. Hierdoor kwam de Europese commissie tot het besluit dat er op dat moment geen nood was aan sterngere regulering inzake credit rating agentschappen. 102

6.2.1

Capital Requirements Directive

De Capital Requirements Directive dateert van 14 juni 2006 en vormt een directe implementatie van de Basel akkoorden in de Europese Unie. 103 De richtlijn heeft als doelstelling de coördinatie van kredietinstellingen overheen de verschilende lidstaten van de Europese Unie. Zulke geharmoniseerde financiële vereisten voor kredietinstellingen laten eerlijke competitie toe tussen entiteiten die gelijkaardige veiligheidsgaranties waarborgen. De CRD speelt een belangrijke rol in de bankenwereld bij de regulering van de minimum kapitaal vereisten. Deze zijn namelijk gebaseerd op het risicoprofiel van de kredietinstelling, wat een functie is van de gegeven rating.104 Het gevaar van de verwijzing in de richtlijn naar het gebruik van externe ratings schuilt zich in het feit dat hierdoor het vermoeden zou kunnen gewekt worden dat een uitgegeven externe rating betrouwbaar en juist is, wat niet steeds het geval is. 105 Deze hoge mate van 97

SY, A., The systemic regulation of credit rating agencies and rated markets, IMF working paper, 2009, 33p. Communication from the commission on credit rating agencies (2006/C 59/ 02) 99 Market Abuse Directive (MAD 2003/6) 100 Capital Requirements Directive (CRD 2010/76/EU) 101 Markets in Financial Instruments Directive (MiFID 2004/39/EU) 102 Communication from the commission on credit rating agencies (2006/ C59/ 02) 103 Directive 2006/49/EC of the European Parliament and of the Council of 14 June 2006. 104 Art 78 en 84 CRD 105 Public consultation on credit rating agencies (05/11/2010) 98


afhankelijkheid en vertrouwen in credit raters kan gemilderd worden door aan relatief grote instellingen de verplichting op te leggen om voor de berekening van hun minimum kapitaal, gebruik te maken van interne ratings106 of door simpelweg beroep te doen op minimum twee verschillende cra’s.107

6.2.2

Market Abuse Directive

Verder is de commissie ervan overtuigd dat de Market Abuse Directive voldoende is om de activiteiten van credit rating agencies vrij te stellen van enige vorm van markt misbruik. De Market Abuse Directive dateert van 2006 en is gericht op het voorkomen van marktmisbruik om zo de goede werking van de financiële markten overheen de Europese Unie te garanderen. 108 Zo zal de richtlijn een waarheidsgetrouwe presentatie eisen wanneer er een aanbeveling wordt gegeven met betrekking tot een belegging.

Er moet echter worden

opgemerkt dat credit ratings opinies zijn, eerder dan aanbevelingen, en dat deze verplicht onderworpen zijn aan specifieke procedures om de waarheidsgetrouwheid ervan te garanderen.109 Verder kan uit de richtlijn worden afgeleid dat een credit rater een daad van markt manipulatie begaat wanneer deze een foute of misleidende rating uitgeeft waarbij hij kennis had of behoorde te hebben van deze foutieve rating.

110

Het is echter behoorlijk

moeilijk om te achterhalen of een credit rater al dan niet te kwader trouw was bij het uitgeven van een foutieve rating, wat de toepassing van de Market Abuse Directive op credit raters doet wankelen.111

6.3

Europese regulering inzake credit rating agentschappen

Credit rating agentschappen zijn op heden niet meer weg te denken uit de financiële markten. Dit grotendeels omwille van de transparantie functie die ze vervullen ten aanzien van investeerders.

112

Ook voor de bepaling van de minimum kapitaaleis en voor allerlei

solvabiliteitsdoeleinden worden ratings gebruikt door gereguleerde entiteiten. Dit heeft tot 106

Art 84 en annex VII CRD Communication from the commission on credit rating agencies (2006/ C59/ 02) 108 Directive 2003/6/EC of the European Parliament and of the Council of 28 January 2003 on insider dealing and market manipulation 109 Art 1 MAD 107

110

Paragraaf 11 MAD Art 6.5 MAD 112 A.BOOT, “Credit ratingbureaus en de stabiliteit van de financiële sector”, Maandblad voor accountancy en Bedrijfseconomie 2008, 227-234. 111


gevolg dat zulke cra’s een significante impact hebben op de functionering van de markt en op het marktvertrouwen van consumenten en investeerders. Hierdoor is het van essentieel belang dat credit rating agentschappen onderworpen worden aan een voldoende strenge regulering. 113 Het probleem hierbij is echter dat deze agentschappen zich ook buiten de Europese Unie bevinden en elk van hen onderworpen is aan de wetgeving die van toepassing is in dat land.114 Bijkomend is de katalyserende rol die credit raters speelden in de recente financiële crisis van 2007 115 waarbij deze credit rating agentschappen er niet in slaagden om de veranderende economische realiteit en verslechterende marktcondities weer te geven in hun ratings. 116 117 Als antwoord hierop kwam ‘Regulation 1060/2009 on credit rating agencies’ tot stand,118een geharmoniseerde regelgeving op Europees niveau inzake de werking van credit rating agentschappen. 119 Hierdoor voorkomt men het risico dat de interne markt verstoord wordt doordat de lidstaten verschillende nationale regelgevingen hebben, wat leidt tot verschillende niveaus van investeerders- en consumenten bescherming.

120

De verordening is tot stand

gekomen op 1 september 2009 en vond haar uitwerking op 7 december 2010. 121 De bedoeling was om te zorgen dat credit ratings die in de Europese Unie gebruikt werden, onafhankelijk, objectief en kwalitatief waren door meer transparantie te creëren en door het wegwerken van de bestaande conflicten. Voorheen vernoemde problemen die credit raters veroorzaakten, tonen aan dat de toen bestaande Europese regulering, namelijk de IOSCO code, MAD en CRD, die enkel in gelimiteerde gebieden van toepassing waren op cras’s, niet voldoende waren om de stabiliteit van de financiële markten te garanderen.122 De regulering stelt dat cra’s die credit ratings uitgeven in de Europese Unie, zich moeten registreren bij de autoriteit van de lidstaat waarin ze actief zijn.123 Verder wordt aan gereguleerde entiteiten, als

113

J.OSTER, ‘“Who Rates the Raters? The Regulation of Credit Rating Agencies in the EU”, The economist 2010, 353–76. 114 UTZIG, F., The financial crisis and the regulation of credit rating agencies: A European banking perspective, working paper Asian Development Bank Institute 2010, 26p. 115 M.MULLARD, “The credit rating agencies and their contribution to the financial crisis”, The policical quarterly 2012, 77-95. 116 A.BOOT, “Credit ratingbureaus en de stabiliteit van de financiële sector”, Maandblad voor accountancy en Bedrijfseconomie 2008, 227-234. 117 BAHENA, A., What role did credit rating agencies play in the credit crisis?, University of Iowa, Center for international Finance and Development 2010, 21p. 118 Regulation (EC) No 1060/2009 of the European Parliament and of the Council of 16 September 2009 on credit rating agencies 119 Paragraph 9 Regulation N° 1060/2009 on credit rating agencies 120 Paragraph 11 Regulation N° 1060/2009 on credit rating agencies 121 Art 41 Regulation N° 1060/2009 on credit rating agencies 122 Paragraph 2 Regulation N° 1060/2009 on credit rating agencies 123 Paragraph 43 Regulation N° 1060/2009 on credit rating agencies EVOLUTIE IN DE REGULERING | 20

kredietinstellingen, investerings- en verzekeringsondernemingen etc., de verplichting opgelegd om voor regelgevende doeleinden, enkel gebruik te maken van credit ratings die uitgegeven werden door Europese credit rating agentschappen. Dit zijn agentschappen die zowel geregistreerd zijn als hun hoofdzetel hebben binnen de Europese Unie. 124 De voornaamste doelstelling van de regulering is om de kwaliteit van de credit ratings te verbeteren door potentiële conflicten te voorkomen. 125 Verder in de paper zullen de opgestelde maatregelen besproken worden en zal toegelicht worden hoe ze elk bijdragen aan het voorkomen van de, in deze paper reeds vernoemde, conflicten.

A. gebrekkige ratingmethoden Zoals eerder reeds vermeld, zorgde de oligopoliepositie van cra’s ervoor dat deze vaak niet inspeelden op veranderende marktsituaties door hun modellen en methodes niet aan te passen, wat aanleiding gaf tot vertekende ratings. De Europese regulering tracht hieraan tegemoet te komen door de verplichting op te leggen om methodes te gebruiken die objectief en systematisch waren en die de mogelijkheid gaven om gevalideerd te worden.126 Verder zal de cra alle relevante informatie met betrekking tot de gebruikte modellen en methodes om tot de ratings te komen, verplicht moeten bekend maken aan het publiek. Hierdoor wordt het voor de gebruiker van de rating mogelijk gemaakt om een eigen analyse te voeren inzake de betrouwbaarheid van de credit rater.127 Ook intern zal het credit rating agentschap regels en procedures moeten voorzien die de kwaliteit van de ratings garanderen door, onder anderen, controle mechanismes op te leggen aan haar medewerkers en personen die mee helpen bij het tot stand komen van de ratings.128

B. Exclusiviteit van de ratingagentschappen Aan de beperkte mate van concurrentie op de ratingmarkt, wat leidt tot een hoge mate van afhankelijkheid van de credit raters, tracht de regulering tegemoet te komen door ervoor te zorgen dat de gebruikers van de ratings, zich er niet blindelings op vertrouwen maar eerder

124

P. STAIKURAS, “A theoretical and empirical review of the EU regulation on credit rating agencies: In search of truth, not scapegoats”, Financial markets, institutions and instruments 2012, 71- 155. 125 Paragraph 11 Regulation N° 1060/2009 on credit rating agencies 126 Paragraph 22 Regulation N°1060/ 2009 on credit rating agencies 127 Art 2 Regulation N°1060/ 2009 on credit rating agencies 128 Paragraph 26 Regulation N°1060/ 2009 on credit rating agencies EVOLUTIE IN DE REGULERING | 21

overgaan op een eigen analyse om de betrouwbaarheid van de credit rater in te schatten.129 Een eerste poging om deze doelstelling te bereiken, is te vinden in de verplichting die eerder reeds besproken werd, waarbij de credit rater informatie met betrekking tot de gebruikte methodes en modellen moet bekend maken aan het publiek. Deze laatste kan op die manier haar eigen ‘due diligence’ voeren om zo te bepalen of ze al dan niet gebruik zal maken van een bepaalde rating in plaats van er zonder evaluatie op te vertrouwen. 130 Verder zal de Europese toezichtshouder, de CESR een centrale database bijhouden waarin informatie te vinden is inzake de historische performantie van elke credit rater, die beschikbaar gemaakt wordt aan het publiek.131 Op deze manier beschikt de potentiële investeerder over voldoende informatie om te komen tot een persoonlijke beoordeling van de credit rater en wordt voorkomen dat deze zijn investering blindelings zal afstemmen op de uitgegeven rating.

C. Gebrek aan onafhankelijkheid van de credit rater Het feit dat de onderneming die gerate wordt, eveneens degene is die de credit rater betaalt voor het uitgeven van de ratings, geeft vaak aanleiding tot een grote mate van afhankelijkheid van elkaar. 132Zo zal de emittent afhankelijk zijn van de credit rater voor het verkrijgen van goede ratings en zal de credit rater voor het bekomen van premies, afhankelijk zijn van de emittent. 133 Een mogelijke oplossing hiervoor wordt door de regulering geboden door aan credit raters de verplichting op te leggen om elke mogelijke situatie waarbij de onafhankelijkheid in het gevaar komt, te registreren en bekend te maken bij het publiek.134 De onafhankelijkheid van de credit rater zal verder gegarandeerd worden door te zorgen voor de onafhankelijkheid van ten minste een derde, maar niet minder dan twee van de leden van de administratieve- of toezichthoudende vergadering. 135 Omdat een lange termijn verhouding tussen eenzelfde credit rater en een entiteit die gerate moet worden, de onafhankelijkheid van de credit rater in het gevaar zou kunnen brengen, heeft de regulering voorzien in een rotatie principe waarbij er een graduele verandering zal plaatsvinden binnen de verschillende analytische teams. Deze graduele verandering houdt in 129

Art 10 Regulation N°1060/ 2009 on credit rating agencies Paragraph 25 Regulation N°1060/ 2009 on credit rating agencies 131 A. DUFF, S. EINIG, “Credit ratings quality: the perceptions of market participants and other interested parties”, The British accounting review 2009, 141- 153. 132 A. BOOT, “De toegevoegde waarde van credit ratings”, the review of financial studies 2006, 1-18. 133 R. KUDPA, “issuer-pays model ensures ratings are available to the entire market”, The economic times 2010,6-8. 134 Paragraph 27 Regulation N°1060/2009 on credit rating agencies 135 Paragraph 29 point 13 section III commission recommendation 2005/162/EC on the role of non-executive or supervisory directors of listed companies 130


dat de samenstelling van een team continu zal aangepast worden, waardoor de individuen moeten verwisselen van team om er zo voor te zorgen dat een entiteit gedurende een lange termijn zal gerate worden door een team dat qua samenstelling exact hetzelfde is.136

D. Geen civielrechtelijke aansprakelijkheid De aansprakelijkheidsmogelijkheden van credit raters zijn in de praktijk zeer beperkt waardoor deze weinig of geen incentives hebben om de betrouwbaarheid van informatie na te gaan die ze gebruiken

bij het berekenen van een rating.

137

De regulering biedt de

mogelijkheid om credit raters aansprakelijk te stellen wanneer deze, of een derde betrokken partij informatie gebruikt waarvan deze weet of behoorde te weten dat ze foutief of misleidend was. Hierdoor kan van de credit rater afgedwongen worden dat deze enkel gebruik maakt van betrouwbare informatie voor het komen tot een rating.138

Het toezicht over de credit raters en het afdwingen van de Europese regulering inzake credit rating agentschappen werd gedaan door de competente autoriteiten van de lidstaat waarin de credit rater is gevestigd. Mechanismes om de goede naleving van de regulering te garanderen, werden dan ook uitgewerkt per lidstaat en vielen onder de bevoegdheid van de competente autoriteit van die lidstaat. Deze kon indien nodig een sanctie bepalen die proportioneel was met de schending van een bepaling van de regulering.139 De Verordening inzake credit rating bureaus is op 1 juni 2011 gewijzigd en de Verordening (EU) nr. 513/2011 is hiermee in werking getreden. 140 Hierdoor werd aan de Europese Autoriteit voor effecten en markten (ESMA) de exclusieve toezichtbevoegdheden gegeven inzake geregistreerde cra’s.

141

Op deze manier kan het hele toezicht- en registratieproces

vereenvoudigd en gecentraliseerd worden overheen de Europese Unie. 142 De algemene doelstelling hierbij was echter nog steeds om de risico’s op financiële instabiliteit te verminderen en het vertrouwen van investeerders te verbeteren door meer kwalitatieve ratings te garanderen. 136

Art 7 N°1060/2009 on credit rating agencies T. BRUYNINCKX, “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2007, 7598. 138 Paragraph 35 Regulation N°1060/2009 on credit rating agencies 139 Paragraph 58 Regulation N°1060/2009 on credit rating agencies 140 Eur-lex.europa.eu 141 Paragraph 6 Regulation N° 513/2011 on credit rating agencies 142 Regulation (EU) N° 513/2011 of 11 May 2011amending Regulation (EC) N° 1060/2009 on credit rating agencies 137


6.4

Voorstel tot vernieuwde Europese verordening

In oktober 2011 kwam de Europese Commissie tot het besluit dat de voorgaande Europese regulering inzake credit rating agencies tekortkomingen vertoonde en hierdoor niet volledig kon tegemoet komen aan de heersende conflicten. 143 De ruime overschatting van het kredietrisico in Europa, de heersende oligopolische vorm van de drie grote cra’s en het gebrek aan aansprakelijkheid hebben bijgedragen tot het besluit dat er nood was aan een gespecifieerde herziening van de bestaande regulering.144 Hierdoor is op 15 oktober 2011 het voorstel tot vernieuwing van de regulering tot stand gekomen.145 De doelstelling van het voorstel is om op een efficiëntere manier tegemoet te komen aan de tekortkomingen met betrekking tot cra’s door het invoeren van meer strikte maatregelen, gebaseerd op adviezen van aandeelhouders die gedurende 5 november 2010 tot 7 januari 2011 tot uiting zijn gekomen in een publieke consultatie. 146 Het afdwingen van de nieuwe regulering, de controle en bestraffing, zullen gebeuren op unie niveau, in overeenstemming met de principes van subsidiariteit. 147De invulling van deze maatregelen en de manier waarop ze bijdragen tot een vermindering van de potentiële conflicten, zullen verder in de paper uitgebreid aan bod komen. Aan het probleem van de gebrekkige rating methoden zal in het voorstel op een verbeterde manier trachten tegemoet gekomen te worden door de regels inzake het publiek kenbaar maken van de methodes te verstrengen. Zo zal er een verplichting opgelegd worden waarbij de credit rater het advies van de aandeelhouders moet raadplegen bij het vormen van nieuwe of het aanpassen van reeds bestaande methodes.148 De grote mate van afhankelijkheid van een credit rater wordt in het voorstel behandeld door aan financiële instellingen de verplichting op de leggen om hun eigen krediet risico analyse te voeren. Over de manier waarop deze analyse zal gebeuren, zal vervolgens toezicht gehouden worden door een competente autoriteit van de lidstaat waarin de rater actief is.149 Een andere maatregel die hetzelfde doel nastreeft, is te vinden in het voorstel om een issuer te verplichten wanneer hij een rating aanvraagt, beroep te doen op twee credit rating agentschappen die 143

F. AMTENBRINK and J. DE HAAN, “Regulating Credit Rating Agencies in the European Union: A Critical First Assessment of Regulation 1060/2009 on Credit Rating Agencies”, Common market law review 2009,1915–49. 144 J. BRUNSDEN, “EU proposes tougher regulation for credit rating companies”, Bloomberg Business week 2011,2-8. 145 Regulation N° 2011/0361 amending regulation N°1060/2009 on credit rating agencies 146 De consulatie is te vinden op: http://ec.europa.eu/internal_market/consultations/2010/cra_en.htm. 147 Art 5 proposal N° 2011/ 0361 148 Art 8(5a), 8(6)(aa), 22a(3) proposal N°2011/0361 149 Art 5a proposal N° 2011/ 0361 EVOLUTIE IN DE REGULERING | 24

onafhankelijk van elkaar opereren om zo tot een objectiever resultaat te komen.150 Verder zal het voorstel een verbetering van het rotatiesysteem inhouden door over te stappen op een agentschap gerichte verandering in plaats van een persoons gerichte verandering, zoals deze bestond in de voorgaande regulering. Meer concreet bestaat het systeem erin dat een credit rater nooit meer dan drie jaar credit ratings mag uitgeven voor eenzelfde issuer. 151 Waar de voorgaande regulering de onafhankelijkheid van credit raters vooral trachtte te behandelen op het niveau van het agentschap zelf, zal het voorstel tot een nieuwe regulering eveneens gericht zijn op de onafhankelijkheid van betrokken personen, zoals de aandeelhouders van het agentschap. Van deze zal verlangd worden dat wanneer zij 5% of meer aanhouden in een agentschap, zij het verbod opgelegd krijgen om meer dan 5% aan te houden in een ander credit rating agentschap. Ook zal een credit rater een verbod opgelegd krijgen om een rating uit te geven wanneer een persoon betrokken bij het rating proces 10% of meer van de stemrechten in dat agentschap bezit en eveneens een lid is van de administratieve- of toezichthoudende vergadering van de onderneming die een rating ontvangt.152 Waar de voorgaande regulering weinig bijbracht aan het probleem van gebrek aan competitie op de ratingmarkt, zal het voorstel een meer gerichte aanpak inhouden om concurrentie te stimuleren. Dit mede omwille van de grote impact van dit probleem.

153

Zo zullen de

voorgaande maatregelen om onafhankelijkheid te verbeteren nutteloos zijn wanneer er onvoldoende keuze is om over te schakelen op een andere credit rater. Een voorstel tot verbetering van de competitie bestaat erin een European Rating Index (EURIX) te creëren, waarin de ESMA elke uitgegeven rating door een geregistreerd agentschap zal publiceren en waartoe een investeerder toegang heeft.

154

De ratings verschijnen hierdoor in een

gestandardiseerde vorm, waardoor de onderlinge vergelijkbaarheid toeneemt, wat in het voordeel is van de investeerder. 155Verder zal het reeds besproken rotatiesysteem een gunstig effect hebben op de toetredingsmogelijkheden van de rating markt door aan opstartende agentschappen de mogelijkheid te bieden om uit te breiden in verschillende gebieden in de markt. 156

150

Art 8b proposal N° 2011/0361 Art 6b proposal N° 2011/ 0361 152 Paragraph 13 proposal N° 2011/ 0361 153 H. LANGOHR ,P. LANGOHR, The rating agencies and their credit ratings: What they are, how they work and why they are relevant, England, John Wiley and sons, 2010, 524. 154 www.finance-watch.org 155 Paragraph 23 proposal 2011/0361 156 Paragraph 7 proposal 2011/ 0361 151


Ook het feit dat issuers verplicht worden om een minimum aan informatie met betrekking tot hun financiële instrumenten bekend te maken, zal op positieve wijze bijdragen tot het creëren van competitie doordat de investeerder op deze manier makkelijker kan komen tot een eigen onderzoek over de de kredietwaardigheid van een financieel instrument. De link tussen deze publieke informatie en de verhoogde mate van competitie kan gesitueerd worden in het feit dat het vrijgeven van informatie met betrekking tot financiële instrumenten, de mogelijkheid om ongesolliciteerde ratings uit te geven verhoogt.

Ook de mogelijkheid tot civielrechtelijke aansprakelijkheid van credit raters zal in het voorstel behandeld en geoptimaliseerd worden. Dit is een belangrijk punt aangezien het de enige manier is waarop de investeerder van de credit rater kan afdwingen dat deze ratings uitgeeft die in overeenstemming zijn met de regels uit regulering 1060/2009 inzake credit rating agentschappen. Dit zou een grote stap vooruit betekenen, aangezien er tussen de investeerder en de credit rater meestal geen contractuele relatie zal bestaan en deze dus onder het heersend aansprakelijkheidsrecht, deze laatste enkel aansprakelijk kan stellen op basis van artikel 1382 BW, buitencontractuele aansprakelijkheid, in geval van samenloop. 157 Het voorstel houdt in dat wanneer een rating wordt uitgegeven die in strijd is met regulering 1060/2009 en een investeerder heeft zich op deze rating gebaseerd bij het maken van een investeringsbeslissing, deze investeerder de credit rater aansprakelijk kan stellen voor de schade die hieruit is voortgekomen.158 Van aansprakelijkheid is enkel spraken in het geval waar de inbreuk op de regulering een effect had op de berekening en het resultaat van de rating en de schade dus een rechtstreeks gevolg is van die vertekende rating. Het voorstel komt gedeeltelijk tegemoet aan het leveren van het moeilijke bewijs hiervan door een gedeeltelijke omkering van de bewijslast in te lassen inzake het bewijs van het bestaan en de impact van een inbreuk op de regulering. Dit omdat de investeerder vaak minder kennis heeft over de gebruikte procedures en methodes van de credit rater en het voor hem dus aanzienlijk zwaarder is om dit bewijs te leveren. Het bewijs inzake de veroorzaakte schade en het oorzakelijke verband, blijven echter liggen bij de investeerder.159

157

Art 1382 BW Paragraph 24-27 proposal 2011/0361 159 Art 35a proposal 2011/0361 158


Het voorgaande voorstel tot regulering van credit rating agentschappen zal geïmplementeerd worden, na volledige goedkeuring van Parlement en overheden en na publicatie in ‘The Official Journal of the European Union’.160

160

Art 2 proposal 2011/0361 EVOLUTIE IN DE REGULERING | 27

7

Besluit

De rol die cra’s spelen in de financiële wereld, is niet beperkt tot het wegnemen van informatieassymetrie tussen belegger en uitgever, maar reikt zich uit tot het mede bepalen van de minimum kapitaaleis waaraan financiële instellingen moeten voldoen. Deze minimum kapitaaleis is een elementair aspect bij het toezicht op banken, wat maakt dat cra’s eveneens een aanzienlijke effect kunnen hebben op de stabiliteit van financiële markten. De recente financiële crisis bracht aan het licht dat het gebruik van cra’s gepaard gaat met allerlei belangenconflicten, wat kan leiden tot een vertekend beeld van de economische realiteit. Uit het verleden is gebleken dat de regulering van cra’s onvoldoende was om aan deze tekortkomingen tegemoet te komen waardoor de Europese Unie pogingen heeft ondernomen om de wetgeving met betrekking tot cra’s te harmoniseren overheen de verschillende lidstaten. Belangenverstrengeling, gebrek aan transparantie en allerlei mogelijke anticompetitieve gedragingen blijven echter uiterst sterk geassocieerd met credit rating agentschappen. De financiële wereld is een verweven en complex gebeuren waardoor er geen eenduidige of allesomvattende oplossing is om tegemoet te komen aan voorvernoemde problemen. Meerdere maatregelen moeten hierbij in acht genomen worden, waarbij het wegwerken van toetredingsbarrières een cruciaal element is. Zo zal het mogelijk gemaakt worden om verhoogde competitie te genereren waardoor, via het marktmechanisme, een systeem van zelfregulatie kan ontstaan. In de economische thesis zal dan ook onderzoek gedaan worden naar het effect van verhoogde competitie op de kwaliteit van de credit ratings.

BESLUIT | 28

Bibliografie

WETGEVING Europese wetgeving Directive 2003/6/EC of the European Parliament and of the council of 28 January 2003 on insider dealing and market manipulation Directive 2006/48/ec of the European Parliament and of the council of 14 June 2006 relating to the taking up and pursuit of the business of credit institutions

Directive 2006/49/EC of the European Parliament and of the council of 14 June 2006 on the capital adequacy of investment firms and credit institutions (recast) Paragraaf 9 Regulation N° 1060/2009 on credit rating agencies Paragraaf 11 Regulation N° 1060/2009 on credit rating agencies Artikel 41 Regulation N° 1060/2009 on credit rating agencies Paragraaf 2 Regulation N° 1060/2009 on credit rating agencies Paragraaf 43 Regulation N° 1060/2009 on credit rating agencies Paragraaf 11 Regulation N° 1060/2009 on credit rating agencies Paragraaf 22 Regulation N°1060/ 2009 on credit rating agencies Artikel 2 Regulation N°1060/ 2009 on credit rating agencies Paragraaf 26 Regulation N°1060/ 2009 on credit rating agencies Artikel10 Regulation N°1060/ 2009 on credit rating agencies Paragraaf 25 Regulation N°1060/ 2009 on credit rating agencies Paragraaf 27 Regulation N°1060/2009 on credit rating agencies Paragraaf 29 point 13 section III commission recommendation 2005/162/EC on the role of non-executive or supervisory directors of listed companies Paragraaf 30 Regulation N° 1060/2009 on credit rating agencies Artikel 7 N°1060/2009 on credit rating agencies Paragraaf 35 Regulation N°1060/2009 on credit rating agencies Paragraaf 58 Regulation N°1060/2009 on credit rating agencies Paragraaf 6 Regulation N° 513/2011 on credit rating agencies Regulation (EU) N° 513/2011 of 11 May 2011amending Regulation (EC) N° 1060/2009 on credit rating agencies Regulation N° 2011/0361 amending regulation N°1060/2009 on credit rating agencies Artikel 5 proposal N° 2011/ 0361 on credit rating agencies Artikel 8(5a) proposal N° 2011/ 0361 on credit rating agencies Artikel 8(6)(aa) proposal N° 2011/ 0361 on credit rating agencies Artikel 22a(3) proposal N°2011/0361 on credit rating agencies Artikel 5a proposal N° 2011/ 0361 on credit rating agencies Artikel 8b proposal N° 2011/0361 on credit rating agencies Artikel 6b proposal N° 2011/ 0361 on credit rating agencies Paragraaf 13 proposal N° 2011/ 0361 on credit rating agencies Paragraaf 23 proposal 2011/0361 on credit rating agencies BIBLIOGRAFIE | 29

Paragraaf 7 proposal 2011/ 0361 on credit rating agencies Paragraaf 24-27 proposal 2011/0361 on credit rating agencies Artikel 35a proposal 2011/0361 on credit rating agencies Financial service action plan Capital Requirements Directive (CRD 2010/76/EU) Artikel 84 CRD 2010/76/EU Artikel 86 CRD 2010/76/EU Artikel 87 CRD 2010/76/EU Artikel 80 CRD 2010/76/EU Artikel 81 CRD 2010/76/EU Artikel 78 CRD 2010/76/EU Artikel 84 CRD 2010/76/EU Artikel 84 CRD 2010/76/EU Annex VII CRD 2010/76/EU Market Abuse Directive (MAD 2003/6) Artikel 1 MAD Paragraaf 11 MAD Artikel 6.5 MAD Markets in Financial Instruments Directive (MiFID 2004/39/EU) Code of Conduct Fundamentals for Credit Rating Agencies, The Technical Committee of the International Organization of Securities Commissions, December 2004 Artikel 2 van Besluit ECB/2011/17 van de Europese Centrale Bank houdende de tenuitvoerlegging van het tweede programma voor de aankoop van gedekte obligaties Rechtsleer Handboeken en verzamelwerken FABER, N.E.D.,De kredietcrisis,Nederland, Kluwer, 2010, 246p. HELLEINER, E., Global finance in crisis, the politics of international regulatory change, Canada, Routledge, 2010, 194p. LANGOHR, H., LANGOHR, P., The rating agencies and their credit ratings: What they are, how they work and why they are relevant, England, John Wiley and sons, 2010, 524p. RICHARD, S., Rating, rating agencies and the global financial system, New York, Publisher US, 2002, 41p. WHITE, L.J., Ratings, rating agencies and the global financial system, New York, Springer US, 2002, 63p.

BIBLIOGRAFIE | 30

Bijdragen in tijdschriften AMTENBRINK, F. and DE HAAN, J., “Regulating Credit Rating Agencies in the European Union: A Critical First Assessment of Regulation 1060/2009 on Credit Rating Agencies”, Common market law review 2009,15–49. BENINK, H.A., “Issues inzake Basel II”, Maandblad voor accountancy en bedrijfseconomie (MAB) 2005, 33-40. BHANKOT, K., MELLO, A., “Should corporate debt include a rating trigger?, Journal of financial economics 2006, 69-98. BOOT, A., “De toegevoegde waarde van credit ratings”, the review of financiel studies 2006, 9-18. BRUYNINCKX, T., “Rating agencies: inhoud, regelementering en aansprakelijkheid”, Jura Falconis 2008, 75-98. CANTOR, R. and PACKER, F., “The credit rating industry”, Federal Reserve Bank of New York Journal Quarterly review 1994, 1-26. BOOT, A., “Credit ratingbureaus en de stabiliteit van de financiële sector”, Maandblad voor accountancy en Bedrijfseconomie 2008, 227-234. CRABBE, L., POST, M., “The effect of a rating downgrade on outstanding commercial paper”, The journal of finance 2012, 39-56.

DONKERS, K., DUFFHEUS, P.J.W. EN WETERINGS, W.C.T., " Credit Rating Agencies: Informatie-asymmetrie en civiele aansprakelijkheid", Maandblad voor Accountancy en Bedrijfseconomie 2010, 506-515. DUFF, A., EINIG, S., “Credit ratings quality: the perceptions of market participants and other interested parties”, The British accounting review 2009, 141- 153.

BIBLIOGRAFIE | 31

BRUNSDEN, J., “EU proposes tougher regulation for credit rating companies”, Bloomberg Business week 2011, 2-9. GORDY, M., “ A risk factor model foundation for ratings-based bank capital rules”, Journal of financial intermediation 2003, 199-232. HAZEU, C.A., “De crisis, de bank en de wet”, Tijdschrift voor openbare financiën 2011, 211. HUNT, J.P.,” Credit Rating Agencies and the 'Worldwide Credit Crisis': The Limits of Reputation, the Insufficiency of Reform, and a Proposal for Improvement”, Columbia Business Law Review 2009, 109-209. KUDPA, R., “issuer-pays model ensures ratings are available to the entire market”, The economic times 2010,6-8. LEENAARS, H., “Risicomanagement van banken”, Maandblad voor Accountancy en Bedrijfseconomie 2003, 8.

MATHIS, J., MCANDREWS, J., ROCHET, J.C., “Rating the raters: Are reputation concerns powerful enough to discipline rating agencies?”, Journal of monetary economics 2009, 657- 674. MULLARD, M., “The credit rating agencies and their contribution to the financial crisis”, The policical quarterly 2012, 77-95. OSTER, J., “Who Rates the Raters? The Regulation of Credit Rating Agencies in the EU”, The economist 2010, 353–76. PARTNOY, F., “The Siskel and Ebert of Financial Markets? Two Thumbs Down for the Credit Rating Agencies.”, Washington University Law Quarterly 77 1999, 619-712.

BIBLIOGRAFIE | 32

SOMMER, A., “IOSCO, Its mission and achievement”, Northwestern University School of Law Northwestern Journal of International Law & Business 1996, 15-27. STAIKURAS, P., “A theoretical and empirical review of the EU regulation on credit rating agencies: In search of truth, not scapegoats”, Financial markets, institutions and instruments 2012, 71- 155. WHITE, L., “The credit rating agencies”, Journal of economic perspectives 2010, 211-226.

Wetenschappelijke literatuur

AMADOU, N., The systemic regulation of credit rating agencies and rated markets, Senegal, 2009, 36p.

BAHENA, A., What role did credit rating agencies play in the credit crisis?, University of Iowa, Center for international Finance and Development 2010, 21p.

CARPENTER, S., WHITESELL,W., ZAKRAJ, E., Capital Requirements, Business Loans, and Business Cycles: An Empirical Analysis of the Standardized Approach in the New Basel Capital Accord , working paper of the Board of governors of the federal reserve system, 2001, 48p.

COVITZ, D., HARRISON, P., Testing conflicts of interest at bond rating agencies with market anticipation: evidence that reputation incentives dominate, working paper, Division of Research and Statistics, Washington, 2003, 37p. DAS, S., SY, A., How risky are banks’ risk weighted assets? Evidence from the financial crisis, IMF working paper, Cornell University, Department of economics, 2012, 38p.

HOLMQUIST, J., Implementation of Basel II: Challenges and opportunities, Institute of International Bankers 2007, 21p. NEIJS, P., De toegevoegde waarde van ratinginstituten ter discussie, working paper universiteit van Tilburg, 2010, 75p.

BIBLIOGRAFIE | 33

PARTNOY, F., How and why credit rating agencies are not like other gatekeepers, University of San Diego School of Law, 2006, 45p.

KRONWALD, C., Credit rating and the impact on corporate structure, Duitsland, 2009, 27p.

SMANT, D., VAN DER ENT, J.P., Credit rating en kapitaalmarkt: Feiten,verklaringen en ontwikkelingen van rating in Europa, Financiele en Monetaire Studies, Rotterdam, 1994, 70p.

STEELE, S.,SECURITIES AND EXCHANGE COMMISSION, A review of implementation of the IOSCO code of conduct fundamentals for credit rating agencies, paper of the technical committee of the international organization of securities commissions 2009, 17p.

STROBL, G., XIA, H., The issuer pays rating model and ratings inflation: evidence from corporate credit ratings, working paper Business School, University of North Carolina, 2011, 41p.

SY, A., The systemic regulation of credit rating agencies and rated markets, IMF working paper, 2009, 33p.

UTZIG, F., The financial crisis and the regulation of credit rating agencies: A European banking perspective, working paper Asian Development Bank Institute 2010, 26p.

TAYTAS, F., Basel II en de gevolgen voor het management van de banken, working paper of the University of Twente, 2008, 28p.

VAN DER ENT, J.P., SMANT, D., Credit rating en kapitaalmarkt: feiten, verklaringen en ontwikkelingen van rating in Europa, Financiele en monetaire studies 1993, 70p. VAN ROY, P., The impact of the 1988 Basel Accord on banks’ capital ratios and credit risk taking : an international study, working paper National Bank of Belgium, 2005, 12p.

BIBLIOGRAFIE | 34

websites http://www.eur-lex.europa.eu [08/11/2012] http://www.law.yale.edu/documents/pdf/cbl/Partnoy_Overdependence_Credit.pdf [08/11/2012] http://europa.eu [12/02/2013] http://stats.oecd.org [28/02/2013] http:// www.fsa.gov.uk/pubs/international/exposures.pdf [12/02/2013] http://www.bis.org [08/11/2012] http://www.eba.europa.eu [5/03/2013] http://www3.ambest.com [22/03/2013] http://www.iosco.org [08/11/2012] http://www.europa-nu.nl [05/03/2013] http://www.fsma.be [04/04/2013] http://ec.europa.eu/internal_market/consultations/2010/cra_en.htm.[12/04/2013]

BIBLIOGRAFIE | 35

Overkoepelend besluit In mijn master thesis wordt een kritische analyse gedaan van de rol die credit rating agentschappen spelen in de economische wereld en de manier waarop ze informatie verschaffen. Het feit dat credit raters enerzijds voor hun diensten betaald worden door de aanvragers van deze diensten zelf, namelijk de issuers, en het beperkt aantal spelers in de ratingmarkt anderzijds, geeft aanleiding tot heel wat belangen conflicten. Als antwoord hierop zijn in Europa strenge reguleringen ontstaan inzake het beleid en toezicht op cra’s en zijn zelfs nieuwe wetsvoorstellen reeds in aanmaak. Hierbij moet echter de vraag gesteld worden of deze recente wetsvoorstellen een effectieve oplossing kunnen bieden voor de problemen die gepaard gaan met credit ratings. Zo zou het namelijk kunnen dat we afstevenen op een situatie van overregulatie waarbij het net de complexiteit van de regulering is die aanleiding geeft tot problemen. Een systeem van zelfregulering, dat via het marktmechanisme mogelijk gemaakt wordt, kan een alternatief bieden. De ‘reputational capital view’ benadering gaat er van uit dat credit raters in voldoende mate zelfregulerend functioneren doordat ze streven naar een goede reputatie. Om zulke reputatie te bereiken, is het belangrijk dat de credit ratings betrouwbaar en waardevol zijn, waardoor de credit rater het vertrouwen van de issuer voor zich kan winnen. In de economische paper zijn we echter tot de conclusie gekomen dat deze benadering niet steeds consistent is en geen verklaring kan bieden voor bepaalde marktgedragingen. Zo zal het belang van een goede reputatie afnemen naarmate er meer spelers op de markt zijn en verhoogde competitie mogelijk gemaakt wordt. Dit kan verklaard worden door het feit dat in een situatie van toegenomen competitie, de verwachte toekomstige inkomsten zullen afnemen doordat de totale winst moet gedeeld worden over een groter aantal spelers. Ter compensatie zullen de rating agentschappen opteren om meer inkomsten op korte termijn te genereren door ratings uit te geven die gunstig zijn voor de issuer. Deze theorie werd in onze economische thesis getoetst door een reeks testen op te stellen die nagaan wat het effect is van verhoogde competitie op de kwaliteit van credit ratings. De resultaten hiervan zijn in lijn met de bestaande literatuur. Ze tonen aan dat verhoogde competitie, wat in het economisch model overeenstemt met de toetreding van Fitch, een significant negatieve impact heeft op de kwaliteit van credit ratings. Het niveau van de ratings zal stijgen, uitsluitend als gevolg van een verhoogd aantal concurrenten in de markt, en de voorspellingskracht ten aanzien van een toekomstige ‘default’ zal afnemen. Deze resultaten

zijn echter gedeeltelijk in strijd met de recente wetsvoorstellen inzake credit rating agencies. Zo bevat het Europese voorstel van 15 oktober 2011 voor een nieuwe cra regulering een aanal maatregelen om de competitie in de rating markt te verhogen.

Meer spelers in de rating markt geeft aanleiding tot een daling van het belang dat credit raters hechten aan een goede reputatie, waardoor de kwaliteit van de credit ratings zal afnemen. Belangrijk hierbij is echter te weten dat verhoogde competitie ook voordelen met zich meebrengt. Zo zal er meer informatie beschikbaar zijn op de markt, waardoor investeerders een bredere keuze hebben en het mogelijk gemaakt wordt om ratings van verschillende agentschappen te vergelijken.

The effect of increased competition on the quality of credit ratings

Recommend Documents