10754962 Max Philippo
The role of credit rating agencies and regulatory effectiveness in the
financial crisis
Programme:
Economie en Bedrijfskunde: Economie en Financiering Supervisor: Ioana Neamtu
In this thesis, the credit rating agencies are discussed and examined to what extent they have had an impact on the financial crisis. It also denounces the policy of the US and examines the effectiveness of legislation on Credit Rating Agencies. A literature review and supporting charts have been used to find a possible answer to these questions. This study has shown that the Credit Rating Agencies have
had a significant impact on the financial crisis and that the change in regulation regarding Credit Rating Agencies has not yet had the desired effect.
2
Table of contents:
Contents
Introduction: ... 3
Methodology: ... 4
Section 1.1: Development Credit Rating Agencies over time: ... 5
Section 1.2: Change in business model: ... 7
Section 1.3: Conflict of interest and market failures: ... 8
Section 2.1: Ratings and information ... 9
Section 2.3: CRA involvement analysis in financial crisis ... 10
Section 2.3: Rating impact on financial crisis ... 13
Section 3.1: Regulation before financial crisis ... 15
Section 3.2: Analysis of liability Credit rating Agencies ... 16
Section 3.3: Regulatory effectiveness and recommendations: ... 17
Section 4.1: Conclusion & Discussion: ... 19
Section. 4.2: Limitations and suggestions: ... 21
Reference list: ... 22
3
Introduction:
In 2007, the subprime mortgage market collapsed. Many households were no longer able to repay the mortgages. The public lost confidence in the financial world, which led to the government being forced to prevent the next economic crisis by means of new legislation. Economists investigated which factors contributed to the emergence of the 2007 financial crisis. According to many studies, the Credit Rating Agencies have played a role in the collapse of the subprime-mortgage market.
Credit rating agencies serve as an important player in the financial world. The rating agencies assign a rating to a financial product that reflects the probability that a debtor will pay its debt back and the likelihood of default. These ratings are used by investors to help them with their decision-making in the financial field. If a financial product receives a low rating, it reflects that it would probably be a risky investment and the investor should, therefore, be compensated with a higher interest. The business model that is used by the CRA's have changed over time. Nowadays, issuers pay the credit rating agencies for rating their financial products, the so-called ‘'issuers-pay'' model. The market of rating agencies in the USA is dominated by three companies; Moody's, Fitch Group and Standard & Poor. These companies hold more than 95% of the market. In this thesis, the focus will be on these major companies.
There is a lot of debate about the exact role of the CRAs in the financial crisis. Harper (2011) concludes in his research that the credit rating agencies did play a role. However, Nagy (2009) puts this conclusion into perspective by stating that ratings are only opinions and do not entirely blame the CRAs for the origins of the crisis. In addition, the government would have failed in its policy to promote competition and transparency in this market (Dimitrov, 2015). However, White (2010) states that there are deeper causes that cannot be fully resolved by new legislation. Looking at the various propositions above, it can be said that the role of CRAs in the emergence of the financial crisis cannot be denied, but the extent to which they played a role should be considered. In addition, the government's intervention should also be highlighted in order to see whether it actually
produces the desired effect. This discussion raises two questions which are dealt with in this thesis: -To what extent have the role of Credit rating agencies influenced the financial crisis in the US?
- How did the US government respond and what has changed in the regulation with respect to Credit rating agencies after the financial crisis?
4
Methodology:
In order to obtain well-founded answers to these main questions, this study is divided into three sections. Each section covers a number of circumstances that support the answers to the above main questions. Based on written research and own statistical analyses, a clear picture is created in each section in order to be able to substantiate conclusions properly.
The first section examines the development of Moody's, Fitch and Standard & Poor in the credit rating agency market. This market development of the big three gives an insight into the distribution of the CRA market and their dominance. Moreover, a statistical analysis of the evolution of market shares within the CRA industry is used to map out the market distribution. In addition, this section highlights the market failures and the conflicts of interests within the CRA industry. With all written academic studies and their findings, a clear picture is created of the problems that arise in the CRA market. Furthermore, the change in the business model is denounced in this section. As result of, the way in which the CRAs do business is shown, and possible reasons for this change are given on the basis of previous academic studies and is supported by statistics.
The second section criticizes the information value of ratings in the run-up to the crisis and its reliability on the basis of previous findings. In addition, a brief introduction is given about a number of factors that have contributed to the emergence of the financial crisis and highlighted the securitization process. Following this, these factors are linked to the CRAs in order to obtain a clear picture of their influence in the run-up to the crisis. In this analysis of involvement, different works are put against each other out of one which are complemented by statistical data. Furthermore, the last part of this section examines the impact of ratings on the financial crisis. With the help of statistics and supporting studies, credit ratings are used to assess the impact on the financial crisis it provides possible explanations for this impact and highlights the aftermath of the crisis. This creates a clear insight into the extent to which CRAs have had an impact on the financial crisis.
In the last section, the regulation of CRAs is reviewed in order to obtain an answer to the second main question. Firstly, the regulation of the financial crisis is briefly mapped out in order to be able to observe a change in the regulation of CRAs. This is followed by an analysis of the (civil) liability of the Credit Rating Agencies. In addition, four problems, which according to the government have played a role with regard to credit rating agencies, are investigated on the basis of academic research. As a result, the government's view of the financial crisis will also be examined, which lead to a greater understanding of the rationale behind the new regulations. Lastly, this section outlines the objectives of the new legislation regarding CRAs, which have been set by the government, are examined on the basis of statistical data and earlier academic findings. This ensures that the role of
5
the government is properly evaluated and the effectiveness of the new legislation is investigated. Using all these findings, a possible answer to the main question is given.
Section 1.1: Development Credit Rating Agencies over time:
The credit rating market have started in the beginning of the 20th century. Moody’s3 began to assign ratings to US railroad bonds in 1909. After some years Poor’s publishing company1 (1919) and Fitch2 (1926) followed this lead and entered the market for credit rating agencies in de early 20’s (White, 2010).
The market of credit rating quickly began to adopt forms of oligopoly in the 20’s of the previous century. According to Partnoy3, the market for credit rating agencies is a reputation-driven and competitive business. This theory says that credit rating agencies will only survive within the industry if they assign reliable and accurate ratings. The so-called reputational capital view states that the value of the company depends on its reputation. As result of, the rating agencies were continuously motivated to maintain their reputation in the market at this time (Partnoy, 1999). High-quality and accurate rating results in higher profits and thus in more market power3. Because the incumbents4 had acquired a high reputation over the past decades, it became more difficult for potential rating services to enter the market. In addition, there was no regulator at this time to monitor the market of credit rating agencies (Veldkamp & Skreta, 2009).
After the great depression regulation regarding rating agencies changed5. The Securities and Exchange Commission was founded by the US government in the early 30’s. This autonomous entity had to supervise financial transactions and was allowed to implement laws itself. The SEC required banks to invest in safe securities in order to minimize the risk. In 1936, banks were allowed to invest only in bonds valued by legally recognized rating services. Although banks were previously able to obtain information from any source, it was now legally established that financial institutions have to invest in securities with certain rating. In this decennium, financial transactions became increasingly dependent on the rating of securities assigned by Moody’s, Fitch and S&P5. Insurance companies were also required by law to maintain a certain minimum level of capital stock in order to spread the risk. In the 1970s, the SEC introduced the National recognized statistical rating Organization
1Merged to S&P in 1942 retrieved from: https://www.spglobal.com/who-we-are/our-company/our-history
2Fitch merged with IBCA in 1997 retrieved from: http://www.nytimes.com/1997/08/11/business/fitch-ratings-agency-in-merger-talks.html 3See; Partnoy, F. (1999). The Siskel and Ebert of financial markets: Two thumbs down for the credit rating agencies. Wash. ulq, 77, 619. 4Moody’s, Fitch and Standard & Poor
6
(NRSRO)6. This was a criterion requirement for potential entrants for the market of credit rating agencies. A rating agency wishing to enter this market should first pass through the SEC’ criteria in order to obtain official recognition (Levine, 2010). As a result of this new legal circumstance, the ratings of the credit rating agencies were central to the bond market according to White (2011). The ratings became very valuable to financial institutions under this law, even though the information was already publicly available (Partnoy, 2002).This created a high entry barrier and enabled the large credit rating agencies to develop further on the market. Moreover, the SEC had not given a clear picture of what criteria the potential entrants had to comply with. The two figures below show the market distribution within industry. As there were hardly any other credit rating agencies than the big three before 2010, the data for that period is not taken into account in this analysis. From these two figures it can be deduced that there has hardly been any change in market shares. These figures suggest that the credit rating agency market is still dominated today by Moody's, Fitch and S&P despite the fact that a number of agencies7 have joined the market in 2010.
Figure 1: Relative Market shares NRSROs in 20108. Figure 2: Relative Market shares NRSROs in 20168
6SeeSEC (2011) retrieved from: https://www.sec.gov/rules/final/2014/34-72936.pdf
7A.M. Best, DBRS, EJR, JCR, KBRA, Morningstar, R&I, see; SEC. (2011). Annual Report on Nationally Recognized Statistical Rating
Organizations (As required by Section 6 of the Credit Agency
8SEC. (2011). Annual Report on Nationally Recognized Statistical Rating Organizations (As required by Section 6 of the Credit Agency
reform of Act of 2006). Data collected from https://www.sec.gov/divisions/marketreg/ratingagency/nrsroannrep0312.pdf; Moody's Corporation. (2010-2016). Annual report 2010.
http://s21.q4cdn.com/431035000/files/doc_financials/annual/10880386.pdf. Fitch 17.61% Moody's 37.23% S&P 42.38% Others [PERCENT AGE]
Fitch Moody's S&P Others (6 in total)
Fitch, 13.28% Moody's, 34.17% S&P, 48.92% Others; [VALUE]
7
Section 1.2: Change in business model:
After NRSRO's implementation by the SEC, the business model of rating services also changed. The agencies switched from the investors-pay model to the issuers-pay model. In the old business model, an investor pays for a rating service. The issuer-pay model states that the issuer now has to pay for a rating. There is a lot of speculation about the motives behind this policy change in the credit rating sector. White (2010) gives in his survey four possible explanations for this change:
The first statement contains the rise of the photocopying machines in the mid-1970s. This allowed one to copy rating manuals which would lead to free-ride behavior. The issuer-pay model rules out this free-riding behavior.
The bankruptcy of Penn-central railroad9 station in 1970 includes the second explanation. The financial system was shocked by this bankruptcy and lost confidence in the bond market. Issuers had to regain confidence. They did so by paying CRAs for a rating in order to convince investors that it would be a safe investment. However, Fridson10 claimed that after the bankruptcy investors were also more willing to pay for a less risky bond. This observation of Fridson (1999) is confirmed by figure 3. This figure shows the yield volatility of an American government bond in the period
surrounding the bankruptcy of Pen railroad station. In order to investigate the change in prices for a less risky bond, a US Government bond was assumed, as the probability of default for this financial product is very low. From this figure, it can be deduced that around the bankruptcy a decrease in yield can be seen. Therefore, it can be inferred that the US bond prices rose and that people were more willing to pay for a less risky bond in 1970.
The third possible explanation is the new regulation on rating agencies. Because the SEC required issuers to include a legally recognized rating on their financial products, the rating services obtained a certain privilege.
The fourth statement states that the market for rating services has a feature of a two-sided market. A two-sided market is characterized by the fact that payments can come from both sides. White (2011) based this statement on the study by Raymann (1999). This survey suggests that the quality of a product can be achieved if one or both parties pay for the product, in this case, the issuers and investors.
9PC went bankrupt in June 1970,Retrieved fromhttps://www.american-rails.com/penn-central.html 10
8 Figure 311: Yield analysis of US Government Bond in the period of 1961-1975.
The conclusion that white (2011) draws from his research and possible explanations is that the change in the business model has led to a conflict of interest within the credit rating agency market.
Section 1.3: Conflict of interest and market failures:
After the change in the business model, the integrity of the credit rating agencies was called into question. Many studies revealed that a conflict of interest had arisen in the credit rating agency market. Bolton (2009) argues that there two sources of market failure in the market of credit rating agencies. The first source was that rating agencies gave underexposed ratings to attract more business activity. Bolton's study argued that when an issuer does business with a particular CRA more often, the rating service tends to give a more favorable rating in order to retain the issuer and to prevent that the issuer will ‘’shop’’ at a competitor. This resulted in incorrect ratings assigned by rating services. Hirshleifer & David (2003) confirms this result by demonstrating through its study that the issuer also has a greater tendency to continue doing business with a particular CRA when it also issues the expected favorable rating. The second source was derived from the market structure. Bolton (2009) investigated two market structures where rating inflation is more likely to occur. Bolton concluded from his research that a monopoly would lead to higher market efficiency than if the market structure were a duopoly. According to Bolton12, this current market is inefficient leading
11Federal Reserve Bank of St. Louis. (2017, 1 November). Long-Term Government Bond Yields: 10-year: Main (Including Benchmark) for the
United States [Dataset]. Retrieved on 15 January 2018, https://fred.stlouisfed.org/series/IRLTLT01USM156N
12 Bolton, P., Freixas, X., & Shapiro, J. (2012). The credit ratings game. The Journal of Finance, 67(1), 85-11
3.00 4.00 5.00 6.00 7.00 8.00 9.00 1961/ 01 1961/ 10 1962/ 07 1963/ 04 1964/ 01 1964/ 10 1965/ 07 1966/ 04 1967/ 01 1967/ 10 1968/ 07 1969/ 04 1970/ 01 1970/ 10 1971/ 07 1972/ 04 1973/ 01 1973/ 10 1974/ 07 1975/ 04 Per cen t
Long-term US Government Bonds Yields: 10 year
maturity
9
to more rating inflation. He recommends a duopoly structure in which an issuer must be legally bound to a certain Credit Rating Agency. As a result, the CRA has no incentives to give underexposed ratings and rating shopping would be ruled out in this case. Skreta & Veldkamp (2009) confirm this result by stating that there is a potential bias present in the ratings if more than one company operates in the market of credit rating agencies.
Section 2.1: Ratings and information
Last decades, ratings have been indispensable to our financial system. Many scientists and
economists question the degree to which CRAs influence the financial market and the relevance of ratings. The economist Friedman once said:
"There's the United States and there's Moody's bond rating service. The US can destroy you by dropping bombs, and Moody's can destroy you by downgrading your bonds. And believe me, it's not clear sometimes who's more powerful."13
Partnoy (2002) examined the second part of Friedman's quote. He has been able to show a paradox through his research: ‘’Continuing prosperity of the credit rating agencies in the face of the declining
informational value of credit rating’’. The explanation of the first part of the paradox includes that
the ratings assigned by the CRA are valuable and essential. Because CRAs have a high market value, they have a strong influence on the financial market and, according to Partnoy (2002), could even control the financial market. On the other hand, practice proved the opposite. Large companies such as Enron Orange and Pacific were not able to meet their obligations in the mid-1970s while an AAA-rating14 was assigned by Moody's. This is remarkable because around this time the CRAs also changed their business models. Partnoy (2002) study finally found that when the public considers a rating to be valuable, this will result in ratings also becoming more valuable to them. The role of the CRA in assigning ratings to securities changed over time. Initially, the CRA's service consisted only of assigning solicited ratings to financial products. Over time, CRAs also began to assign unsolicited ratings. These ratings were assigned without any form of payment being made. These unsolicited ratings were used by CRAs to punish an issuer when doing business with a competitor (Partnoy, 2002). From this it can be concluded that CRAs had considerable power in the financial market. There was much criticism of the new way in which the CRA operated on the financial market. The financial institutions receiving the unsolicited ratings claim that these ratings often reflect poor knowledge of
13 As cited In: Thomas L. Friedman on 28 March 1999, in the New York Times Magazine
14See Appendix A.1: Timoschenko, A. (2015, 3 October). Rating agencies: Moody's, S&P and Fitch. Retrieved From: 15 January 2018, from;
10
the entity (Poon & Firth, 2010). In addition, they accuse the CRAs that these unsolicited ratings are lower than the solicited ratings. These statements were examined by Poon & Winnie (2009). The investigation has shown that unsolicited ratings differ significantly in comparison with solicited ones. Poon & Winnie (2009) have shown that there is a negative bias in unsolicited ratings. Poon & firth (2010) endorse this result by saying that companies that pay for the service are given a significantly higher rating than companies that receive an unsolicited rating.
Kliger & Sager (2009) examined the extent to which information has an impact on the price of a financial product. The survey questioned the way in which CRAs provided their information. CRAs claimed that they have consistent sources from which they drew the information. Ratings will be downgraded if the CRA considers that they will adjust the rating in the next two years (SEC, 2006). Kliger & Sager (2009) research has shown that if positive information is released by the CRA on a particular bond, the price of this bond will rise. Levine (2010) did, however, comment on his conclusion. He claimed that there is an exchange of knowledge between the CRA and the issuer before the rating is even assigned. Looking at the above results, it can be assumed that CRAs tend to assign a more optimistic rating to entities paying for a credit rating. These results show that CRAs are acting in a questionable way.
Section 2.3: CRA involvement analysis in financial crisis
The landscape of structured finance has expanded over the last decade. Securitization processes turned out to yield a positive result for the financial entities and this led to a complicated financial system (Acharya & Schnabl, 2010). At the beginning of 2007, the value of various structured finance products decreased. There was uncertainty about the decline in mortgage-backed-securities
(Tavakoli, 2008). According to Tavakoli15, Securitization technology allowed banks to move loans, MBS and other financial products from their balance sheets, thereby shifting risk and allowing banks to do more business. CDOs were an important factor in the emergence of the credit crisis. A CDO is a financial product in which collateral guarantees a possible default, usually this collateral includes obligations, loans and mortgages (Mishkinn, 2011). Duffie & Garlenanu (2012) explain the securitization process in their research in two steps. In the first instance, a portfolio will be developed in which collateral is selected. After that, the bank tries to structure an optimal capital combination in order to value a portfolio.
The CDO market developed rapidly into a large financial submarket. The market value of CDOs was estimated at 3 Billion dollars in 2005 (SEC, 2007). CDOs enabled banks to develop
15Tavakoli, J. M. (2008). Structured finance and collateralized debt obligations: new developments in cash and synthetic securitization (Vol. 509). John Wiley & Sons.
11
prime mortgages. These were certain mortgages where the risk of repayment is transferred to the mortgage borrower. This new financial technology enabled the bank to lend people a mortgage that was not initially qualifying (Mishkinn, 2011). These subprime mortgages were securitized and resold to third parties.
According to Barnett-Hart (2010), there are two factors that make the securitization process attractive to investors and investment banks. Firstly, process made it easier for banks to meet regulatory capital requirements. Banks were able to free up the balance sheets through this process. This allowed them to comply with the capital requirements of BASEL II16 and to comply with their own risk requirements. Secondly, securitization made it possible to reallocate the risk. By combining different uncorrelated financial products into a portfolio, banks were able to diversify risk. This enabled portfolios with a higher risk to be rated better according to Barnett-Hart (2010).
The bundled portfolios were divided into three tranches. The tranche with the lowest
probability of default was marked with AAA status, named senior tranche. By the end of 2006, almost all CDOs had obtained AAA status17. Because most CDOs were assigned AAA status, these financial products became very attractive to investors. Research by Wojtowicz (2012) has shown that CDOs generated a higher yield compared to corporate bonds. This led to an increase in demand for CDOs, but investors were unaware that the portfolios contained an underlying risk. By combining high-risk and low-risk bonds, a tranche can be composed that can qualify for e. g. AAA status. This is a crucial difference when CDOs are compared to corporate bonds, as it can explain that when bonds are bundled, the risk associated with the combination of bonds will be estimated to be lower. This while the effect of a default on a risky bond has a major impact on the loss of a CDO tranche. As a result, the CRAs' risk model would underestimate the risk compared to corporate bonds. Credit ratings thus gave investors an erroneous indication of the actual risk. Defaults led banks and investors to lose their investments, causing the entire sub-prime mortgage to collapse (Wojtowicz, 2012).
In figure 4, CDOs issues are presented in the immediate run-up to the collapse of the mortgage market and its aftermath. In 2006, there was a substantial increase in the issuance of CDOs. After the collapse of the subprime mortgage market, a very significant decrease of
approximately 85% was taking place on the issuance of CDOs in 200715. From this, it can be deduced that the CDOs fell sharply in value so that the current ratings gave an erroneous picture of actual risk. Furthermore, the fall in value will raise the question of whether it has also had an impact on the CRAs' revenues.
16 Basel Committee on Banking Supervision. (2004). International Convergence of Capital Measurement and Capital Standards (A Revised
Framework Comprehensive Version). Retrieved from: https://www.bis.org/publ/bcbs128.pdf
12 Figure 418: effect of the financial crisis on CDOs issuance
In order to investigate the impact, Figure 5 shows Moody's revenues. The figure shows a small decrease in revenues in 2007, but this is not significant. From this, it can be concluded that the decrease in value of structured finance products had a moderate impact on Moody's revenues. Remarkable because the previous figure shows a significant decrease in 2007. This may be explained by the fact that, even in financially difficult times, CRAs are important to maintain public confidence in the economy. In addition, it can be deduced from this that the credit rating agencies do not bear the ultimate risk and that the issuers and investors receive the impact of the financial crisis.
Figure 5: Moody's revenues analysis in the financial crisis19.
18Asset-backed alert. (n.d.). Rating-Agency Shares of CDOs issuance [Dataset]. Retrieved January 8, 2018, from
https://www.abalert.com/rankings.pl?Q=102
19 Source: Moody's Corporation. (2011). Annual Report. Retrieved from
http://s21.q4cdn.com/431035000/files/doc_financials/annual/12862427.pdf $0 $5,000 $10,000 $15,000 $20,000 $25,000 2 0 0 6 2 0 0 7 2 0 0 8 2 0 0 9 2 0 1 0 2 0 1 1
MOODY'S REVENUES IN FINANCIAL
CRISIS ($MILL.)
$0 $1,000 $2,000 $3,000 $4,000 $5,000 $6,000 $7,000 $8,000 $9,000 $10,000 2006 2007 2008 2009 2010 2011 2012 2013 2014 2015 Tho us andsCDOs issuance by Credit Rating Agencies
13
In Rajan & Griffin's (2010) research, four problems are highlighted that would have contributed to the inconsistent rating of CDOs.
Firstly, it argued that investors in CDOs with a higher risk premium should be compensated. This is because the investors considered comparable investments, such as corporate bonds, to be a good measure for investing, while the underlying risk of CDOs is very different from the risk of a corporate bond as already concluded by Wotjowicz (2012).
Subsequently, Griffin & Tang (2008) argue that it is likely that there will be an economic crisis, only the results of the Rajan & Griffin (2008) study prove that this is highly unlikely.
The third problem describes that investors in CDOs are likely to have unrealistic or biased assumptions regarding ratings. Coval et al. (2009) confirmed this conclusion by proving that bundling a portfolio with uncorrelated bonds could lead to a decrease in the probability of default, while the individual probability of a bond was higher. This gave investors an erroneous indication of actual risk due to the misleading ratings of the CRAs.
Finally, Griffin & Tang (2008) concluded that fraud, laxity, and selfishness of the banks would have led to a decrease in the quality of the collateral. This decrease in quality would then be
inconsistent with the CDO's rating.
Section 2.3: Rating impact on financial crisis
Looking at the above results, it can be said that CRAs assigned misleading and inconsistent ratings. The banks combined a certain capital structure in the portfolio in order to obtain a higher rating. Figure 6 shows that approximately 80% of the CDOs present in the market contain an AAA rating between 2000 and 2007. Figure 6 shows also that credit rating agencies systematically assigned AAA ratings to CDOs from the beginning of this century (Benmelech & Dlugosz, 2010). This is not
remarkable either because research by Barnett heart (2010) has shown that the profits of the big three rose significantly in the same period. From these findings, it can be noted that the influence of the CRA was considerable in the previous decades.
14 Figure 6: Share of CDOs with AAA status to total in the run-up to the financial crisis (%)20.
The collapse of the mortgage market has led CRAs to downgrading structured finance products. Research by Benmelech & Dlugosz (2010) shows that around 30 % of all downgrades in 2007 and 2008 were highly rated (AAA). Figure 7 shows the number of downgrades of CDOs during the crisis. The figure shows that there was a significant increase in the number of downgrades of CDOs in the year 2007 by the big three. In the years during the crisis, this number dropped moderately to its previous level before the crisis. Taking into account the above-mentioned result of Benmelech & Dlugosz (2010), it can be concluded that the credit rating agencies played a role in the collapse of the Subprime Mortgage market because the best rated financial products suddenly became less valuable due downgrading.
Figure 721: CDO downgrades during the Financial Crisis
20 Retrieved from: Benmelech, E., & Dlugosz, J. (2010). The credit rating crisis. NBER Macroeconomics Annual, 24(1), 161-208.
0 2000 4000 6000 8000 10000 12000 2007 2008 2009 2010 2011 2012 NU MB ER O F D OW NGR AD ES
CDO Downgrades from AAA to BB in the U.S.
15
Bennemlech & Dlugosz (2010) described two possible explanations that have contributed to the significant downgrading of ratings in the structured finance market. First, it is suggested that CRAs deliberately overrated structured financial products. This statement was based on the
phenomenon of rating shopping. This phenomenon would have resulted in CRAs writing overvalued ratings, leading to an inflationary rating in the structured finance market. The second potential explanation for the significant downgrading of AAA tranches includes the invalid rating model of the CRAs. According to this study22, the models would have led to an underestimation of the probability of default. In addition, they claim that if there is public knowledge about an error in the model, issuers will look for a model that assigns the most favorable rating. From this conclusion, it can be deduced that one statement encourages another.
Section 3.1: Regulation before financial crisis
The regulation on credit rating agencies was very limited before the crisis. According to White (2011), it was only when Enron went bankrupt that more supervision of the credit rating agency market came into being. Research had shown that the CRAs did not downgrade their ratings until a few days before Enron's declaration of bankruptcy. This erroneous action by the credit rating agencies came to the attention of the media and policymakers. This also brought the disputable NSRSO system to public attention. It demanded legislative reform in the area of CRA legislation. In response to the dissatisfaction of late intervention by the CRAs in Enron's bankruptcy, the American Congress came up with a new legislation. This law required the SEC to send a report to the US Congress about the credit rating agency market and the terms and conditions for becoming a recognized CRA (NRSRO system). This law allowed the US Congress to gain a better understanding of the structure of the market and the non-transparent NRSRO system. This law put the SEC under pressure, resulting in the SEC allowing two new CRAs (A. M. Best and Domino bond rating servic6es)23 to enter the market. However, the SEC had still not clearly defined the NRSRO system, which meant that the desired effect was lacking according to White.
The SEC was seen as a barrier to entry into the CRAs market and in response, the US
21The Association of for Financial Markets in Europe. (2007-2012). AFME Securitization Data Reports. Data obtained from
https://www.sifma.org/search/?aq=esf%202014&hPP=10&idx=prod_wp_searchable_posts&ap=0&is_v=1
22 Benmelech, E., & Dlugosz, J. (2010). The credit rating crisis. NBER Macroeconomics Annual, 24(1), 161-208.
23SEC. (2011). Annual Report on Nationally Recognized Statistical Rating Organizations (As required by Section 6 of the Credit Agency
16
government implemented a new Credit Rating Agency Reform Act (Amadou, 2009). This law obliged the SEC to establish a new system which clearly sets out the conditions to be met by a potential entrant in order to obtain legal recognition. This stimulated competition in the credit rating agency market. In addition, the influence of the SEC on the market was partly limited by this act24
Section 3.2: Analysis of liability Credit rating Agencies
Although the involvement of the CRAs is undeniable, there are different views on the liability of rating agencies in the financial crisis. In 2008, at the insistence of the US Congress, the SEC evaluated the role of credit rating agencies in the financial crisis. This evaluation report was part of the Credit Rating Reform Act mentioned above. This report identified 4 factors which, according to the SEC, had had an impact on the financial crisis with regard to the credit rating agencies25.
Firstly, it emerged that CRAs were no longer able to manage the complexity and size of structured finance products since 2002. This observation is supported by Benmelich & Duglozs (2010), who stated in his research that the structured finance market was the largest and the most complex financial market in the world.
Secondly, the CRAs did not have a transparent methodology for rating structured finance products. However, the CRAs were not legally obliged to publicly disclose their methodology (SEC, 2002). The conclusion of the SEC is partly contradicted by Benmelech & Dlogosz (2010), by saying that there was indeed public knowledge that errors were present in the rating model, which eventually would create rating shopping.
The third factor describes the inadequacy of the rating process. The SEC concluded that the CRAs did not use clear guidelines in the process and acted inconsistently with regard to the rating of a structured financial product. Due to the complexity of the CDOs and MBS, the likelihood of an error in the process was greater.
Fourthly, the SEC exposes the integrity of the CRAs. The SEC concluded that the current market structure and conflicts of interest have led to a lack of integrity on the part of the CRA. The SEC, therefore, wanted to limit these undesirable effects through the CRR-Act (SEC, 2006). However, the results of Bolton's study (2009) prove the opposite. His research has shown that increasing competition in the credit rating agency market would increase the likelihood of a conflict of interest.
24 See; Sy, M. A. N. (2009). The systemic regulation of credit rating agencies and rated markets (No. 9-129). International Monetary Fund. 25SEC. (2006). Annual Report on Nationally Recognized Statistical Rating Organizations (As Required by Section 6 of the Credit Rating
17
In addition, he proved that a monopoly structure would be the most efficient and would avoid a conflict of interest.
At the beginning of the financial crisis, the CRAs were frequently sued by investors claiming that the CRAs were liable for the collapse of the mortgage market (Nagy, 2009). Harper (2010) describes one argument put forward by the CRAs that can explain the success. In the liability cases, the CRAs often invoked the first amendment, because ratings were only opinions and the
assignations of ratings were not legally binding. In addition, the ratings contained disclaimers in which they ruled out civil liability. Although it is difficult to hold the CRAs legally liable for the
financial crisis, it can be concluded that the CRAs have some degree of liability looking at de different arguments mentioned above.
Section 3.3: Regulatory effectiveness and recommendations:
As already mentioned, CRR-act was implemented in 2006. According to Sagers & Fitzpatick (2010), this law, which focused on increasing transparency, liability and competition in the market, did not have the desired effect. According to Sagers & Fitzpatick (2010), this law would have created legal immunity for the CRAs and would only have decreased SEC's influence.
Because the results were not in line with the goals set, the US Congress came up with a new law in 2009: the Dodd-Frank Act26. Dimitrov (2015) examines the effectiveness of the DF-act in relation to the credit rating agencies. He examines the effect of a provision derived from this law. The provision increased the SEC's ability to legally punish CRAs for assigning inaccurate ratings. This provision should lead to an increase in the credibility and quality of ratings and should limit credit rating inflation (SEC, 2010). Dimitrov's study (2010) has shown that increasing the risk of civil liability leads to less accurate and informative ratings. Dimitrov27 bases a possible explanation for this phenomenon on an earlier study by Morris (2001). In this study, it is argued that if an issuer suspects that ratings will be biased, in an industry where reputation is considered important, the advisor (CRA) has a reason to lie in order to prevent reputation damage. Partnoy (1999) complements this research by stating that the CRA industry is driven by reputation. Based on rational thinking, the provisions should improve the quality of ratings. However, Dimitrov's (2015) research has shown that increasing
26United States. (2010). Dodd-Frank Wall Street Reform and Consumer Protection Act: Conference report (to accompany H.R. 4173).
Washington: U.S. G.P.O.
18
the regulatory oversight of CRAs does not necessarily lead to better quality ratings. One of the goals of the Dodd-Frank act is to increase competition. According to the US Congress, this should lead to more informative and accurate ratings. In the period 2010-2016, seven new NRSROs entered the market (SEC, 2016). Assuming that increasing the number of CRAs should lead to a more balanced distribution of the market shares.
Figure 8: Analysis of NRSROs industry28
Market share29 2010
(%)
Market share29
2016 (%)
Change market share (%) S&P 42,38% 48,92% 6,53% Moody's 37,23% 34,17% -3,06% Fitch 17,61% 13,28% -4,33% A.M. Best. 0,23% 0,39% 0,16% DBRS 0,23% 1,78% 1,55% EJR 0,03% 0,81% 0,77% LACE 0,63% 0,70% 0,07% R&I 0,03% 0,28% 0,24% Realpoint 0,30% 0,45% 0,15% JCR 0,03% 0,16% 0,13% HHI-index30 0,349 0,314
To investigate this objective, a Herfindahl-index31 can be seen in figure 8. This Index indicates the degree to which a market is concentrated. The HHI index of 2016 shows a value of 0.314, according to the U. S. legal guidelines32 regarding market concentration, a market is considered to be highly
28SEC. (2011). Annual Report on Nationally Recognized Statistical Rating Organizations (As required by Section 6 of the Credit Agency
reform of Act of 2006). Data collected from https://www.sec.gov/divisions/marketreg/ratingagency/nrsroannrep0312.pdf; Moody's Corporation. (2010). Annual report 2010. http://s21.q4cdn.com/431035000/files/doc_financials/annual/10880386.pdf.
29Outstanding ratings to total outstanding ratings according to Form NRSRO made public by NRSROs (government bonds excluded). 30
Herfindahl-index: 𝐻 = ∑ 𝑆𝑖𝑛 2, 𝑤ℎ𝑒𝑒𝑒 𝑠𝑖 = 𝑀𝑀𝑒𝑀𝑒𝑀 𝑠ℎ𝑀𝑒𝑒 𝑖𝑖 %, 𝑖 = 𝑖𝑛𝑛𝑛𝑒𝑒 𝑜𝑜 𝑜𝑖𝑒𝑛𝑠 𝑖=1
31A measure of firm size relative to market size used as an indicator of market power.
32Department of Justice and The Federal Free trade Commission. (2010). Horizontal Merger Guidelines. Retrieved from:
19
concentrated if the Herfindahl index is above 0.25. The table shows that the HHI has increased over the last 5 years, which means that market concentration has decreased. Although there has been a decrease in concentration according to the HHI-index, no significant change can be observed. In addition, the government bond category is not included in the index calculation. This was, in fact, the rating category dominated by the big three (SEC, 2016) which resulted in a highly concentrated sub-market. Since this concentrated sub-market has not been taken into account, the indexes do not give a representative impression of reality. Although figure 7 shows a slight decline in market
concentration, there is still a high degree of concentration in the market. It can, therefore, be concluded that the Dodd-Frank act has not yet led to more balanced market shares. A possible explanation for this may be that ratings from Moody's, Fitch and S&P are more valuable to investors, as they have been operating in this market for decades. This advantage in reliability makes it difficult for small CRAs to manage themselves in the market.
Although civil liability and market transparency were increased by the Dodd-Frank act, many economists questioned the effectiveness of the law. The above results show that the market is still unevenly distributed and those market failures have not yet been resolved. For this reason, many economists called for a reform of the regulation. White (2011) criticizes the approach of the American government and argues for less regulation. He gives two arguments why the current approach, more regulation, would have little effect. By increasing transparency, there is a risk of the intellectual property being eroded and, in the long run, discouraging its development. In addition, increasing legal requirements will increase the costs for potential entrants, which in principle increases the entry barriers. White (2011) argues for more prudential supervision. He argues that financial institutions should seek their own information instead of the CRAs making safety
judgments. This allows investors in the market to choose which' advisor' is credible. As a result, the regulation of CRAs is no longer necessary.
Section 4.1: Conclusion & Discussion:
The first section highlighted the development of the CRA market. The results show that the Credit Rating Agency market is dominated by three large companies. In addition, it can be concluded that the SEC, with its NRSRO guidelines on becoming a legally recognized rating agency, has increased the barriers to entry in the CRA market for potential entrants. These guidelines were rather unclear, allowing the incumbents of the CRA market to develop freely without interfering with the entry of new NRSROs into the market. Looking at the statistical results, it can be concluded that today there is still an unfair distribution in the CRA market. In addition, this section highlighted the change in the
20
business model. The change from the investor-pay model to the issuer-pay model, has called the integrity of the CRAs into question. In his research, White (2011) gave four reasons for this change. Bolton's (2009) investigation then concluded that this change of business model would have created conflicts of interest. High ratings would have a significant impact on the value of financial institutions (Kliger & Sager, 2011), which would benefit from a high optimistic rating. Bolton’ study (2009) showed that if an issuer would do business with the same CRA on a frequent basis, this would lead to an optimistic rating. As a result, issuers would be less inclined to' shop' with the competitor for a more favorable rating. Bolton (2009) claimed that the current market form would lead to inefficiency and inflation of ratings within the CRA industry. On the basis of Bolton's results (2009), he argued for a market form in which issues are legally bound by a certain CRA. As a result, CRAs would no longer be inclined to issue underexposed ratings, which would lead to the elimination of conflicts of interest.
The second section examined the role of the CRAs to what extent they have been involved in the emergence of the financial crisis. This section has shown that ratings have played an important role in the structured financial market. This submarket turned out to be favorable for the revenues of the financial entities according to Barnett-hart (2010). However, these financial institutions were aware that they were heavily dependent on the ratings of the CRAs. This dependence constantly increased the power of the CRAs in the financial world. As a result, the information value and
reliability of ratings deteriorated (Poon & Firth, 2010). It can also be concluded from this section that the influence of the CRAs on the collapse of the subprime reporting market was significant.
Ultimately, the structured financial portfolios turned out to give an incorrect representation of the actual risk, resulting in reputational damage to the CRAs. This resulted in a massive downgrading of structured finance products by the CRAs, so that these products turned out to be worthless. It can therefore be concluded that these underexposed ratings derive from the conflict of interest between the CRA and the issuer. These ratings were a favorable form of business for the CRAs. This allowed them to retain the issuers, while the issuers ultimately carried the risk.
In third section, the government's regulation of credit rating agencies was denounced. The American Congress questioned the role of the CRAs in the financial crisis. In response, the SEC concluded in its evaluation report that the credit rating agencies did indeed play a role in the emergence of the financial crisis (SEC, 2008). This report showed that the lack of transparency, competition and liability within industry would have strengthened the role of the CRA in the emergence of the financial crisis. In response, the Dodd-Frank Act was implemented in 2009 to counter these market failures. Although the DF Act has promoted civil liability and transparency, the policy to increase competition has failed. The results of the last section clearly show that the
21
hardly changed in 2016 compared to 2010 when the DF-act was implemented. It can therefore be concluded that the DF-act has not yet had any effect in this area. However, only a period of 5 years was considered. The future will have to determine whether this law will ultimately produce the desired effect.
Looking at all the findings of the three sections, a possible answer to the raised main question will be given. Looking at all the findings of the three sections, it can be said that the influence of the credit rating agencies on the financial crisis has been significant. The statistical findings and written academic works confirm this conclusion. In addition, there has indeed been a change in the legislation on CRAs. However, the question is whether this is the best approach to minimize the market failures. In my opinion, the objectives have still not been achieved. The market is still dominated by three large companies and the controversial issuer-pay model is still used today by the CRAs, so that conflicts of interest continue to exist. As it still has no effect on current
regulation, we should look at where the core of the problem lies. In my opinion, there are two problems that should be addressed. Firstly, there should be a law prohibiting the issuer-pay model. The CRAs should return to the old model in which the investor pays for a rating instead of the issuer. This minimizes the conflict of interest, thus excluding the assignment of underexposed ratings. In addition, the opaque NRSRO process must be reformed to become a legally recognized agency. This would promote transparency within the market and should increase CRAs' liability. Addressing these two problems would, in my opinion, lead to fair competition, thus limiting the role of CRAs in possible future economic crises. However, it is still difficult to say how the government should deal with this circumstance. The dominance in the market of CRAs remains a stumbling block for the government and it might take a very long time until fair market sharing.
Section. 4.2: Limitations and suggestions:
The reader should bear in mind that the scope of this thesis may be too broad. The findings of this research are based solely on previous literature that has been supplemented with some statistics. Although the conclusions are confirmations of earlier academic works, this thesis can serve as a theoretical basis for further empirical studies.The conclusions would be even more well-founded if this were confirmed by empirical evidence. In future empirical studies, we could look at a market form in which the market for Credit Rating Agencies is entirely in the hands of the government to see to what extent this has an effect on the quality of ratings. In addition, an empirical follow-up study could also investigate the extent to which market failures occur if work is still carried out under the investor-pay model in order to determine whether conflicts of interest have actually been minimized. These suggestions for further research could be a good addition to this thesis.
22
Reference list:
Acharya, V. V., Schnabl, P., & Suarez, G. (2013). Securitization without risk transfer. Journal of
Financial economics, 107(3), 515-536.
Baily, M. N., Litan, R. E., & Johnson, M. S. (2008). The origins of the financial crisis.
Barnett-Hart, A. K. (2009). The story of the CDO market meltdown: An empirical analysis (Doctoral dissertation, Harvard University).
Benmelech, E., & Dlugosz, J. (2010). The credit rating crisis. NBER Macroeconomics Annual, 24(1), 161-208.
Bolton, P., Freixas, X., & Shapiro, J. (2012). The credit ratings game. The Journal of Finance, 67(1), 85-11
Coval, J., Jurek, J., & Stafford, E. (2009). The economics of structured finance. The Journal of
Economic Perspectives, 23(1), 3-25.
Davidoff, S. M., & Zaring, D. (2009). Regulation by deal: the government's response to the financial crisis. Admin. L. Rev., 61, 463.
Dimitrov, V., Palia, D., & Tang, L. (2015). Impact of the Dodd-Frank act on credit ratings. Journal of
Financial Economics, 115(3), 505-520.
Duffie, D., & Singleton, K. J. (2012). Credit risk: pricing, measurement, and management. Princeton University Press.
Federal Reserve Bank of St. Louis. (2017, 1 November). Long-Term Government Bond Yields: 10-year: Main (Including Benchmark) for the United States [Dataset]. Retrieved on 15 January
2018, https://fred.stlouisfed.org/series/IRLTLT01USM156N
Fitzpatrick IV, T. J., & Sagers, C. (2009). Faith-based financial regulation: A primer on oversight of credit rating organizations. Admin. L. Rev., 61, 557.
Griffin, J. M., & Tang, D. Y. (2011). Did credit rating agencies make unbiased assumptions on CDOs?. The American Economic Review, 101(3), 125-130.
23
Harper, S. (2011). Credit Rating Agencies Deserve Credit for the 2007-2008 Financial Crisis: An Analysis of CRA Liability Following the Enactment of the Dodd-Frank Act. Wash. & Lee L. Rev., 68, 1925.
Hirshleifer, David (2003) and Siew Hong Teo (2003). "Limited attention, information disclosure, and financial reporting." Journal of accounting and economics 36.1-3 (2003): 337-386.
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. Colum. Bus. L. Rev., 109.
Kliger, D., & Sarig, O. (2000). The information value of bond ratings. The journal of finance, 55(6), 2879-2902.
Lacalle, D. Credit‐rating agencies. Life in the Financial Markets: How they really work and
why they matter to you, 95-98.
Mishkin, F. S., & Serletis, A. (2011). The economics of money, banking and financial markets. Toronto: Pearson Addison Wesley.
Morris, S. (2001). Political correctness. Journal of political Economy, 109(2), 231-265.
Nagy, T. (2009). Credit Rating Agencies and the First Amendment: Applying Constitutional Journalistic Protections to Subprime Mortgage Litigation. Minn. L. Rev., 94, 140.
Partnoy, F. (1999). The Siskel and Ebert of financial markets: Two thumbs down for the credit rating agencies. Wash. ulq, 77, 619.
Partnoy, F. (2006). How and why credit rating agencies are not like other gatekeepers. Poon, W. P. (2003). Are unsolicited credit ratings biased downward?. Journal of Banking &
Finance, 27(4), 593-614.
Poon, W. P., & Chan, K. C. (2010). Solicited and unsolicited credit ratings: a global perspective. Rousseau, S. (2005). Enhancing the accountability of credit rating agencies: The case for a disclosure-based approach. McGill LJ, 51, 617.
Skreta, V., & Veldkamp, L. (2009). Ratings shopping and asset complexity: A theory of ratings inflation. Journal of Monetary Economics, 56(5), 678-695.
Sterling, K., Fridson, M. S., & Kong, V. C. (2009). Return Dynamics of Distressed Bonds. The Journal of
24
Sy, M. A. N. (2009). The systemic regulation of credit rating agencies and rated markets (No. 9-129). International Monetary Fund.
Tarr, D. G. (2010). The political, regulatory, and market failures that caused the US financial crisis: What are the lessons? Journal of Financial Economic Policy, 2(2), 163-186.
White, L. J. (2010). Markets: The credit rating agencies. The Journal of Economic Perspectives, 24(2), 211-226.
Wojtowicz, M. (2014). CDOs and the financial crisis: Credit ratings and fair premia. Journal of Banking
25
Appendix:
A.1: Rating scale and definitions
Timoshenko, A. (2015, 3 October). Rating agencies: Moody's, S&P and Fitch. Retrieved From: 15 January 2018, from; http://www.billiontrader.com/post/106
