Banks, to get rid of illiquid assets they posses and to attain financial freedom in lending, searched for new innovative techniques. This innovative method of converting these illiquid assets in to liquid assets technique is called asset securitization. Banks pool up these illiquid assets like mortgage loans and sell it to agencies called as special purpose vehicle (SPV). These special purpose vehicles convert these loans in to securities and sold to investors. Before agencies sold these securities they got it rated from rating agencies. Asset securitization as a process reduced information asymmetries; increased financial slack; served as a lower cost of financing source; reduced regulatory capital; and reduced bank risk. The process of asset securitization as a whole has many advantages but by the end of year 2007 it started to crack with financial crisis. It is therefore necessary to study what went wrong in the process of asset securitization that lead to financial crisis.
The study analyzes the role of asset securitization in financial crisis by analyzing the economics of asset securitization process as whole. Then in depth analysis of credit rating agencies methodologies and economics of how they rate these securities is studied. As it is difficult to analyze the rating processes and methodologies of all rating companies in this thesis I have decided to analyze Moody’s investor service. Moody’s has been selected because its name is synonym with quality in the market.
The growth and fall of mortgage industry performance of mortgage industry have been analyzed. The factors that led to financial crisis have been analyzed. The study analyze the moody’s rating methodologies and rating models and updates to rating models. The short comes in rating methodologies and rating process has been discussed. The rating models updates effect on default rate of rating has been analyzed. Finally the effect of these default rates on financial crisis has been studied and analysis of role of asset securitization in financial crisis is studied.
The process of asset securitization started in the year 1870 when Government National Mortgage Association (GINNIE MAC) purchased pools of loans and converted in to securities and sold these securities to investors. In the year 1970 special innovative technique called tranching were used to distribute losses involved in these pools of loans backed by mortgages and sold to investors. Kaptan and Telang (2002) Asset securitization is the process of converting illiquid assets in to cash flows. Both financial intermediates such as banks and investors benefited from this process. Banks benefited with extra liquidity to lend more loans to able borrowers where as investors got opportunity to invest in capital market for more returns.
In the process of asset securitization, rating agencies rating securities is crucial because rating influence the marketability of the securities. There are many rating agencies which rate Residential Mortgage Backed Securities, of these three largest credit rating agencies with overseas market that are based in United States are Moody’s, Standard and Poor (“S&P”) and Fitch. These rating agencies use statistical models to analyze risk involved. Rating agencies constantly review performance of these securities and according to performance they upgrade or downgrade rating.
To lessen the effects of a mild recession in 2000, the Federal Reserve cut interest rates. This interest rate cut along with increasing housing price made people to invest in housing this helped to drive growing demand for nontraditional mortgages products. Banks have extra liquidity to lend more loans to borrowers and started to lend more and more loans to non prime borrowers, which led to poor performance of loans and in turn effected whole asset securitization.
This report will explore what is the role of asset securitization in financial crisis. In order to research what is the role of asset securitization in financial crisis the following have done
1. Analysis of asset securitization process
2. Analysis of Evolution of financial crisis
3. Analysis of Rating agencies methodologies and procedures in rating process.
The details of analysis techniques are explained in methodology chapter. And extensive literature review is done to get hold of the subject. Finally in depth analysis has been done to reach the goal of the report.
Kaptan and Telang (2002) defined Asset securitization as an innovative process which channelizes flow of funds from investors to issuers in efficient manner. In simple words, the process of asset securitization starts with financial institutions like banks which pools up individual loans and create securities against them. These securities are rated and sold to investors. In words of these authors, asset securitization is the process of converting assets in to securities and in turn in to liquid cash.
Origins of securitizations can be traced back to 1870`s where Government National Mortgage Association (GINNIE MAC) started selling securities that are backed by pool of mortgage loans. These securities were named as mortgage pass through securities.
This process of securitization has changed in 1970 where new innovative concept of tranching was introduced in issuing the securities (tranched securities). These tranched securities are sold to investors. Kaptan and Telang (2002)
(Uzun and Web, 2007) makes understanding of asset securitization more simple through an illustration of the process of asset securitization, banks which are financial intermediaries in capital market has various types of assets such as mortgage loans, car loans, leasing contracts etc on their balance sheets. These assets are not marketable so these are illiquid assets. Banks, to get rid of these illiquid assets and to attain financial freedom in lending search for new techniques. This innovative method of converting these illiquid assets in to liquid assets technique is called asset securitization. So asset securitization plays a major role in converting these illiquid assets in to cash flows (liquid assets).
Uzun and Web, also provide information on what kind of assets the banks securitize. These authors explain this as, the process of asset securitization starts with banks deciding which assets they want to securitize for example mortgage loans. Then bank pools these mortgage loans and sell it to trustee or separate entity which is called special purpose vehicle. (Uzun and Web, 2007)
Role of Special purpose vehicle (SPV) is explained by the Securities and Exchange Staff (2008) as, SPVs either government backed agencies or private agencies such as Fannie Mac, Friede Mac, Ginnie Mac buys these loan pools and are entitled to interest and principal of underlying loans in the pools. Then SPV issues different classes of securities known as tranched securities backed by pool of loans.
The role of SPV is to separate risk of newly created securities from the origin bank loans. If these SPV are not there it is very difficult to assess the risk involved with those securities underlying the loans. It is difficult to access risk because risk involved is closely related to origination bank practices. Information of origination bank practices such as how they lend loans what documentation they check before issuing loans and credit quality of loans. Securities and Exchange Staff (2008) conclude that these securities issued from this SPV isolates the risk involved from origination bank. Investors invest on these securities and investment risk is directly interrelated to credit quality of loan borrowers whose loans are offered as collateral for the securities. To boost the demand for these securities the SPV enhances credit quality by process called over collateralization and subordination.
Over collateralization, is the process in which credit quality is improved by giving payment guarantee by insurer. So if there is any principal or interest default it is insured there by making investors clear in mind that there is no risk involved in investing in these securities. Over collateralization is one way of credit enhancement but the principle way of credit enhancement is done by subordination.
In subordination process SPV issues different layers of tranches (securities) such as junior, mezzanine, senior tranches. If the trust experience any loss in interest or principal payment, lower most tranches, junior tranche absorb all the losses and then mezzanine tranche absorbs any more remaining losses that are left over by junior tranche leaving top most tranches, senior tranche safe from any kind of losses. So senior tranche is safe from all interest and principal default. So by process of tranching top most tranches (securities) get more demand from investors and demand reduces when it goes down the ladder up to junior tranches. Junior tranches are backed by over collateralization for its marketability in capital market. The process of tranching differentiates structured finance from normal securitization process.
In normal securitization process assets are converted into securities and sold. In structured finance these securities are tranched so that at least one class of securities gets better rating when compared to average rating of all securities.
The asset securitization makes calculation of risk more complex using technique called tranching. The calculation of risk is more complex because the risks involved in these pools are distributed.( Securities and Exchange Staff) (2008)
Asset securitization is the process of converting illiquid assets in to cash flows (liquid assets). Both financial intermediates such as banks and investors benefited from this process. Banks benefited with extra liquidity to lend more loans to able borrowers where as investors got opportunity to invest in capital market for more returns. Kaptan and Telang (2002)
In brief benefits of asset securitization are reducing information asymmetries; increasing financial slack; serving as a lower cost of financing source; reducing regulatory capital; and reducing bank risk (Greenbaum and Thakor, 1987)
(Kaptan, Telang (2002), (Uzun and Web, 2007) conclude that asset securitization is the process in which illiquid assets of banks are converted into cash flows or liquid assets. (Greenbaum and Thakor, 1987) conclude these techniques of asset securitization as benefits for banks as well as for investors in capital market. Securities and exchange staff concludes the process of credit enhancement using process called subordination distributed risk of loss in the whole tranche. And the process of over collateralization increased demand for these securities in capital market. Securities and Exchange Staff (2008) concluded that the process of tranching evenly distributed risk and assessing this risk is a complicated process.
The main role of rating agencies in capital market is to rate the bonds and securities in specific scale. Rating agencies use qualitative and quantitative methods to access cash flows of these bonds or tranched securities. These ratings are used by investors in capital market as bench mark in investing. Thus rating agencies helped the investors in making decision to invest in capital markets by reducing information asymmetries between issuers and investors. (Committee on the Global Financial System), (2005).
According to Ruth Rudden, the evolution of rating industry started when there was a big demand for the corporate bonds in USA. The investors interested to invest in these corporate bonds were very skeptical about risk involved as they were not provided with company’s credit information that issued these bonds. So there was a pressing need for an independent and third party institution to analyze credit risk of these bonds which helped the investors in making decision to invest according to their criteria. Thus credit rating agencies came into existence. (Ruth Rudden, 2007),
John moody was the first to introduce credit ratings in 1909. He used rating scale to rate the bonds. These ratings were useful for investors to understand credit risks. Credit rating agencies (CRAs) stressed more on expected cash flow generated by the issuer (special purpose vehicle) ongoing business in determining the rating. In general CRAs revenues were generated from subscribes who subscribed to receive rating on debt securities. Rating agencies from the start has been rating bonds on specific scale. Mason and Rosner, concluded that the rating doesn’t give information on whether particular bonds must be bought or sold. They give their opinion on relative safety of the bonds. (Mason and Rosner, 2007)
The main importance for the credit ratings rose in the capital market because of US treasury department. US treasury department said the quality of the bonds rated by rating agencies is appropriate. Ruth Rudden, concluded that the importance of credit rating agencies in the capital market became prominent and the investors relayed on these ratings to invest on the bonds. (Ruth Rudden, 2007)
Then with the introduction of new structure finance products, rating agencies started to rate these products as well. In one of the reports by the Committee on the Global Financial System, (2005), wrote about the Rating agencies, rated the structured finance products like asset backed securities, CDOs, RMSBs etc, same as the traditional bonds. Rating agencies performed the same function as with traditional bonds that was reducing information asymmetries between issuers and investors. Committee on the Global Financial System, (2005)
Issuers of structured finance products wanted these securities to be rated on the same scale as traditional bonds so that investors think structured finance has same kind of risk that of bonds. (Mason and Rosner, 2007) spoke about the structured finance as, for past few years with the introduction of newly formed structure finance products; these CRAs are chasing the agencies that issue these structured finance products instead of subscribers for revenue.
This lead to three fold increase in the revenues by CRAs and effected the integrity and base source of the aim on which rating industries are build. To meet the demand of these newly introduced structured finance products; CRAs have introduced many new models and approaches to access these products for ratings. (Mason and Rosner, 2007) The three largest credit rating agencies with overseas market that are based in United States are Moody’s, Standard and Poor (“S&P”) and Fitch.
According to (Rousseau Stephane, 2009), all the rating agencies methodologies are almost same for rating RMBS products.
First issuer of these securities approach rating agencies to rate their securities so that they can sell it in capital market. And issuer provide all the data information of the assets underlying the securities like loan data, proposed capital structure of SPV, proposed credit enhancement for each tranche of the securities.
Rating agency will assign an analyst to analyze the tranches for rating it. First probable looses incurred on all tranches are calculated. Rating agencies used complex statistical models for analyzing loss. The loss analysis gives rough idea of how much credit enhancement is required for each tranche to give particular rating.
Then analyst analyzes proposed capital structure of SPV to check whether it meets particular rating. Then finally analyst analysis the cash flow which gives information of interest and principal paid out of SPV and analyzes whether particular asset which is under tranche meets payment obligation. Analyst then rates each tranche and submits his rating to committee where they vote on the analyst view.
Once rating is confirmed they send the rating to issues rather than publishing it. If the issuer is satisfied with the rating he makes it public. If issuer makes rating public, rating Agencies get paid if not they get breakup fee. (Rousseau Stephane, 2009).
According to (Daníelsson J, 2002), Rating traditional bonds is much easier because of availability of historical data where as rating structural products like RMBS you need much more complex models than that of normal models.
As the financial system become more complex, the need for complicated statistical models becomes greater. More the complexity, lesser the reliability on these models, so does these models tends to be less reliable. It is clear from the credit crisis of 2007 that the rating agencies used over optimistic input data, inappropriate modeling and insufficient checking of data quality and permitting gaming of models. Despite of advanced models, stress tests, and all the numbers, risk models do have important role to play in modeling risk as long as its limitations are known. Risk models are good at managing particularly trading desk but when asked to model whole institution it fails. So relying on such folly statistical models to model risk is foolishness. And the numbers that these models give are inappropriate.
Financial models are not simple and do not have basic or fundamental thermos to build on. These models can easily make you believe the results are accurate, the reason for these are;
1. Endogenous risk: In finance we can only model aggregate behavior. Financial modeling changes the statistical laws governing the financial system in real-time, leaving the modelers to play catch-up. This becomes especially pronounced as the financial system gets into a crisis. This is a phenomenon is called endogenous risk.
2. Quality of assumptions: we can’t take it to consideration all parameters in to model so it is important to take it to consideration the main parameters that affect the outcome of the model. For example if we consider present situation of financial crisis the main parameter is liquidity which has been be ignored by modelers while modeling risk.
3. Data quality: data quality is the most and foremost important thing in statistics because the accuracy of these models depends up on quality of data. (Daníelsson J, 2002)
To prove what Daníelsson J, said Vanessa G. Perry proved, there is always dearth of data on subprime market. The data that is available is proprietary lender data. And this data had drawbacks on analysis of market trends. To analyze data properly we need property records which contain information on mortgagee and mortgager, transaction price, property location, credit score, foreclosure rate of neighborhood state.
This data was necessary for the rating agencies to analyze the market condition properly. Roughly to analyze loan performance, three sets of data was taken into consideration, that is the Borrower data, loan data, property data. Borrower data should contain income, FICO score, and demographics. The loan data should contain loan amount, LTV, loan type, interest rate/fee, terms such as FRM/ARM, payment history. Property data should contain location, prices, sales, foreclosure, and employment rate. One can predict the probability of default if and only if these data of loan is available. (Vanessa G. Perry, 2008).
According to (Committee on the Global Financial System, 2005), and (Mason and Rosner, 2007) there are many concerns on rating agencies which rated the RMBS, they are;
1. Transparency- Given the role that is played by rating agencies in removing Asymmetries, it is important that they be transparent on what they do. Rating agencies never disclosed completely their methodologies they use to rate RMBS and key assumptions and rating criteria. Credit rating agencies never accepted that the data provided by issuer of securities are not sufficient to rate. And rating agencies never provided historical performance data about their ratings.
2. Quality of rating process- there is a huge growth in RMBS market because of ease in lending loans. And at the same time these RMBS products started to get more complex. The rating agencies did not have enough staff to tackle increasingly complex products and huge volume of these products. Because of shortage of work force these rating agencies were not able to catch up with rating upgrades or downgrades accordingly with change in circumstances like issuers principal or interest short fall.
3. Conflict of interest- the rating agencies main role is to act as an intermediate between investors and issuers. This trust of being intermediate has been broken by rating agencies by charging issuers for rating products instead of getting paid by subscribers who subscribe for these ratings to invest in these products. Because of shift in the axis of being intermediate, these rating agencies got paid from issuer who in turn profited rating agencies by gaining millions of dollars. This process of issuer paying for his rating created conflict of interest. So considering profits they incur from this new role, rating agencies tend to rate products issued by these issuers a higher rating than they actually are. The issuer has ability to adjust deal structure to get desired rating. And issuer has influence on rating process. (Committee on the Global Financial System, 2005), (Mason and Rosner, 2007).
According Tom Bulford (2008), (Ruth Rudden, 2007) “The credit rating agencies like Moody’s, Standard and Poor’s and Fitch played a central role in growing the residential mortgage-backed securities, these credit rating agencies were titled to rate these securities on behalf of the huge investment banks to sell to the investors.
The ratings of these securities were to identify the risk involved in the securities, they followed a particular three main flow in calculating the risk rating for the investors, the first as to interest the investors on the securities, they provided portfolios of RMBS which highlighted a certain level of risk involved in it, this was done through tranches which means, the different level of risks involved securities were put into different groups called tranches. This helped the investors in deciding whether to stay first in line during the event of default or down the queue. This was one point where the investors relayed on the ratings to invest on the securities.
The other two things which they followed to rate the securities, one was data which was used in the financial models of the rating agencies to rate these securities, the data contained here are the information about the mortgage loans that are parceled by the investment banks. These mortgages came from the originators who provided all the information about the mortgagees like their credit history, income, etc. hence these originators provided information was historical.
The information given source was not sure about as they stood by the words of the originators. Using this information on the models they used in the rating would off course end up being inaccurate. This made the investors relaying on the high rating given by these rating agencies and hence invested confidently.
The rating agencies assured that the portfolios of mortgage backed securities were “stress tested” by ‘Monte Carlo simulation of macroeconomics variables to create a loss distribution’. The assumptions were not wide enough because the rating agencies relied upon historical data, and till now MBS were concerned ‘the performance history that did exist occurred under very benign economic conditions’. The reasons just don’t stand on rating agencies following the historical data for the calculations but also the workload and the conflicts when the interest rates rose which laid the investment bank concentrate on getting the best ratings on the securities that is laid for sale.
This increased the competition between the agencies; they did not want to lose deals and hence gave ratings as necessary with one initiative that was not to lose deals. One of the illustration proves the above comment, one of the member in an rating agency who did not want to lose a deal wrote a mail which said “I had a discussion with the team leaders here and we think that the only way to compete is to have a paradigm shift in thinking, especially with the interest rate risk”. Another said “We are meeting this week to discuss adjusting criteria for rating CDOs of real estate assets because of the ongoing threat of losing deals”. Tom Bulford (2008),
Tom Bulford (2008) concluded that the roles of these rating agencies in financial crisis are to be studied thoroughly. Rating agencies main duty is reducing information asymmetries between issuers and investors but with the introduction of structured finance products rating agencies deviated from their main role of reducing information asymmetries. In fact they started to favour security issuer as they are paid for rating.
Committee on the Global Financial System, (2005) concluded that role of rating agencies in capital market is to rate bonds or securities on specific scale. (Ruth Rudden, 2007), (Mason and Rosner, 2007) concluded that the importance of credit rating agencies in the capital market became prominent and the investors relayed on these ratings to invest on the bonds and the rating doesn’t give information on whether particular bonds must be bought or sold. They give their opinion on relative safety of the bonds. The rating agencies got paid by issuer of securities for rating structured finance products on same scale as normal bonds.
Tom Bulford (2008) concluded that change of role by rating agencies as information intermediation between issuer and investor got strained with the introduction of structured finance products. (Daníelsson J, 2002) concluded that to rate structured finance products rating agencies need more complex models. (Vanessa G. Perry, 2008) concluded that there is no enough historical data on subprime market and in turn this dearth of data affects accuracy of the rating process. Committee on the Global Financial System, (2005) concluded that there are some concerns relating rating process of RMBS. Tom Bulford (2008) concluded rating agencies had played their role in financial crisis and need to be blamed for their irresponsible behavior.
The roots of financial crisis are complex and obscure. The main culprits are mortgage banks brokers, rating agencies, to some extent federal reserve and government. Financial crisis started with Federal Reserve slashing interest rates to encourage spending and reduced 30-year bond issues to increase the prices. This along with American dream of home ownership triggered housing boom. This housing boom has been used by many mortgage lending banks. The introduction of FICO scores instead of traditional point based system and the off-balance sheet vehicle made lending loans easy. Loans were given to people with low credit history (sub-prime loans) Souphala, C and Anthony, P.C, (2006)
According to Souphala, C and Anthony, P.C , (2006), the introduction of FICO scores instead of more traditional “point based system” credit scoring. And the off balance sheet vehicle (OBSV) made banks to lend loans to people with low credit score. This type of lending is called subprime where these borrowers are who fail credit history requirements in the standard (prime) mortgage market. The subprime lending is known as high cost lending and primarily driven by credit history and down payment where as prime lending is driven by down payment only. People thought prime lending is complicated but have great promise and great peril.
The subprime lending provided opportunity for homeownership to those who haven’t passed credit history in the past. Lower credit history of subprime lending which could have resulted in more delinquent payments and defaulted loans.
US mortgage market, which for decades was dominated by fixed rate mortgages, included nontraditional mortgages, simultaneous second-lien mortgage, and no documentation or low documentation loans. Nontraditional mortgages allow borrowers to defer payment of principal and sometimes interest and include interest only mortgages (IOs) and adjustable rate mortgages (ARMs) with flexibility payment options. Interest rates are much higher than that of prime loans, is the main reason of risk for borrowers.
Strong home price appreciation and declining affordability have helped drive growing demand for nontraditional mortgage products that can be used to stretch home buying power. Souphala, C and Anthony, P.C , (2006).
National partners in home ownership in the largest private public partnership program whose solo aim is increasing home ownership rate to all time high by the end of decade by increasing creative financing methods for mortgage loans. In this program, retailer, home builders, Fannie Mac, Freddie Mac, mortgage bankers are the partners who came up with innovative ideas such as using FICO score instead of point based system is introduced to ease the requirements to lend loans to people whose credit history is not good to get mortgage loan. Another innovation is off balance sheet vehicle which made lending loans easy. (Mason and Rosner, 2007)
According to Souphala, C and Anthony, P.C , (2006),The government and the quasi-government agencies were main reason who influenced the US mortgage credit cycle by their legislative reforms and the mandates, the alternative mortgage transaction parity act in 1982 eliminated regulatory disparities between state and federal chartered mortgage by granting state chartered institutions the authority to issue alternative mortgage(sub-prime), including the use of variable interest rates and balloon payments, regardless of state mortgage lending law. The tax reform act 1986. Then stimulated demand for mortgage debt by retaining the deduction for home mortgage interest.
To lessen the effects of a mild recession in 2000, the Federal Reserve cut interest rates. Although the Fed has raised interest rates past year, mortgage rates have largely been unaffected. This interest rate cut along with increasing housing price made people to invest in housing. Home ownership is best way of making wealth in fact most households find it difficult to invest in anything but their homes. These factors helped to drive growing demand for nontraditional mortgages products that can be used to stretch buying power. Souphala, C and Anthony, P.C , (2006).
Due to poor standard of lending there has been raise in subprime loans, the delinquency rate increased in the year 2006-2007 because of subprime loans issued in previous years. The overall rise in delinquency rate is sudden and overwhelming. The market started to response to these high delinquency rates in the second half of 2006 and first half of 2007. In spite of high delinquency rate, market had confidence on highly rated tranches of subprime RMBS (senior tranches). In the second half of 2007 this confidence came to its low when credit rating agencies lowered their rating on highly rated tranches. These downgrades created uncertainty and doubt on quality of rating these rating agencies assigned. With more exposure to risk related to subprime debts, restricted liquidity of banks, the inter market for term loans was effected so there was a sharp increase in risk premium. These authors concluded that banks lost confidence and have less liquidity. This resulted in present financial crisis.
The result of this is freezing all structured finance products and cut down in non confirming mortgages. This is because of those agencies giving non confirming mortgages had lots of loans and RMBS which were not sold. Several financial institutions in the U.S. and abroad were hit with sizable losses owing to their exposures as sponsors of SPVs and underwriters of other structured credit, as well as their direct exposures to subprime-related debt.
Souphala, C and Anthony, P.C , (2006). concluded that lower standard of the way the mortgage loans were issued is the one of the main reasons of financial crisis. (Mason and Rosner, 2007) concluded that US federal government played its role in financial crisis by easing many laws. Souphala, C and Anthony, P.C , (2006) also concluded that degrading senior tranche ratings by rating agencies triggered financial crisis and banks lost confidence and have less liquidity to lend loans.
Through the journey of knowing the impact of asset securitisation, and its process we have seen the number of parties involved in this process. As mentioned earlier for the process of Asset securitisation it passes from the financial institutions to the special entity vehicle to the rating agencies, the investors relay on these ratings and invest on the securities. Hence, to understand the role played by all these parities in financial crisis. The methodology used in this report address the key research question of "role of asset securitization in financial crisis" and "role played by rating agencies in financial crisis". Both the key problems are one and same because rating agencies are part and parcel of asset securitization process. These chapters will address what methodologies best suit to address the problem.
Quantitative and qualitative research methodologies have been adopted for the purpose of this report. We use quantitative research to facilitate qualitative research.According to Dow (2002), quantitative methods are commonly used for analyze trends and measure. Quantitative research is done only when the person is having profound knowledge on the subject or to analyze the phenomenon which is under research. As to analyze the causes of financial crisis, quantitative methods are used to analyze secondary data on interest rates, housing prices, loan performances. According to Silverman (2000), qualitative data analysis involves gathering qualitative data and analyzing qualitative data. Qualitative data is the data collected from documents such as reports, emails, transcripts etc. And quantitative analysis is interpreting the quantitative data by understanding it and explaining it. To analyze the rating methodologies and rating methodologies updates of rating agencies we use qualitative analysis.
To be more precise, among the three main rating agencies who are the market holders, I have chosen Moody’s as they were the first to start up with rating the bonds and were regarded more reliable in the ratings provided. Moody’s don’t hold a first place in market when it comes to revenue but their existence in the field is high compared to any other rating agencies in the market. I believe in “experience is the profit to the company”. Though the models used in the rating agencies were identical with minor changes, the information I could find and reliable was on Moody’s which motivated me to analyse the models used in Moody’s which again drained the financial sector.
Taking into consideration on the models used by these Moody’s on rating the residential mortgage securities, it becomes necessary for me to relay on the models they have followed over the years and hence find the changes made and its impact on the ratings. Hence the methodology followed here is qualitative methods.
To find the impact of updates of moody’s rating methodologies on ratings; quantitative analysis is done on issuance and default rate.
According to Kinnear & Taylor (1996), data collection is very important step in any research. The type of data used reflects on quality of the research. To achieve the goal of report, Secondary data has been collected from different sources such as 1. Research papers, 2. Academic journals, 3. Working papers of banks, 4. Internet. Secondary data has been used in this report to reach research objective. Secondary data is the data that has been be published by other authors and organizations(Kothari, 2005).
Data has been collected from various government websites for the purpose of analysis of financial crisis. Data on federal interest rates has been collected from Federal Reserve statistical reserve website. 30 year fixed mortgage rate has been collected from Freddie Mac company website. Housing prices index (median), real housing price index has been collected from US Department of Housing & Urban Development website. Foreclosure, delinquency and loan performance data has been collected from US Government Accountability Office (GAO) website.
Moody’s investor service (Moody`s) has been selected for the analysis of role of rating agencies in the financial crisis. Moody`s has been selected because moody’s has been in the business of rating right from the starting and ratings of moody’s are synonym for quality and investors trusted these ratings. Moody’s rating methodologies, rating process and rating review process have been collected from moody’s website. These information has been published my moody’s as to prove transparency in the rating process. Default rate which is consists of principal impairments, interest impairments, total impairments has been collected from moody’s annual report of moodys Global Credit Ploicy on Default and loss rate of structured finance securities. Moodys revenues and expenditures has been collected from moodys annual reports. All these informantion has been collected from moodys website
Secondary data is considered to be more risky(Kothari, 2005) as the data is not reliable. For the purpose of financial crisis analysis I have used publications from top authors in the field, working papers of banks, and statistics from government websites. It is reliable to use this data as it has been published by top authors in the field and from top publications, government websites, and renowned banks. So keeping trust on these institutions and authors I can conclude that secondary data collected from this source is reliable to reach research objective.
Secondary data collected for the analysis of role of rating agencies in financial crisis is from Moody`s investor services company website. This data which has been collected from Moody`s website is reliable as it is official publications of the company.
Finally to conclude the data collected for the analysis part is reliable and validated as it is collected from official company website, government websites and from renowned journals from renowned publications.
In order to reach the objective of this report, the following analysis is done
1. Analysis of evolution of credit crisis: To reach at aim of analysis of evolution of credit crisis, the factors that lead to financial crisis which has been discussed in professional and academic context has to be studied. Federal interest rates, 30 year fixed mortgage rate, Housing prices index (median), real housing price index. Foreclosure, Delinquency and Loan performance.
These factors have been collected from various sources. Data on federal interest rates has been collected from Federal Reserve statistical reserve website. 30 year fixed mortgage rate has been collected from Freddie Mac company website. Housing prices index (median), real housing price index has been collected from US Department of Housing & Urban Development website. Foreclosure, delinquency and loan performance data has been collected from US Government Accountability Office (GAO) website. The validity and reliability of these data has been discussed in the previous section.
Data that is collected has been quantitatively analysed for trends by plotting line graphs. Looking at trends of line graphs and knowledge gained from study of professional and academic context, analysis is to be done. The effects of each factor on loan performance delinquency and foreclosure rate have to be studied. Finally conclusions of what is the evolution of credit crisis have to be written.
While analysis the financial crisis, there are many other factors which would have been reasons for financial crisis but in this research only main and important factors that led to financial crisis have been studied. The factors that are going to be used are important because various authors in professional and academic context have concluded these are main factors that effected financial crisis. There is no problem with the reliability and validity of the data collected as it is published by government. Only restriction in the analysis of financial crisis is analyzing different other factors.
2. Qualitative analysis of rating methodologies (moody’s mortgage metrics): yearly (1999-2008) Moody’s rating methodologies, rating process and rating review process have been analyzed. The data for the analysis have been collected from moody’s website. This information has been published by moody’s website as to prove transparency in the rating process. Rating methodologies, process of rating, review of rating procedures of Moody’s have to be studied qualitatively.
As the data have been collected from Moody`s Investor services website, reliability and validity of the data cannot be questioned. The limitation is moody`s never completely disclosed their methodologies and sensitive parameters in the rating process. So to completely rely on data available Moody’s website seems bit mistake but general idea of what is Moody’s methodologies and rating process can be understood from this analysis. Incomplete disclosure is also one of the concerns raised by many authors as stated in professional and academic context, so we don’t need to worry about it. Only aim of this analysis is to get in depth view of what is Moody`s rating methodologies and process. Few sensitive parameters which haven’t been published will not affect the analysis process of the research to be done.
3. Qualitative analysis of updates to moody’s mortgage metrics till 2008. The updates of Moody’s methodologies in rating have been collected from moody’s website. Qualitative analyses of these updates have to be done. The data collected from Moody’s investors service is reliable and valid as it is published on Moody`s website itself. There is only one limitation as analysis is done on the data published. If Moody’s have published all the updates the analysis would have been accurate. Immaterial of limitations the published updates can clearly be useful for the purpose of analysis.
4. Quantitative analysis of effects of these updates on default rate: Default rate which is consists of principal impairments, interest impairments, total impairments has to be analyzed whether it has effected by updates. Data for this analysis has been collected from moody’s annual report of moodys Global Credit Ploicy on Default and loss rate of structured finance securities 2008. the analysis have to be done what effect did rating updates have on default rate. Finally conclusion have to be drawn from the analysis. The data source is reliable and accurate as the default rate data has been collected from Moody`s publication. The only limtation is about updates which has been discussed above.
5. Addressing concerns rose on rating methodologies and rating process: After analysis have been done on rating methodologies and procedures, updates effect on default rate, concerns raised by rating methodologies and rating process which have been discussed in professional and academic context have to be addressed. To address these concerns qualitative, qualitative analysis which have been done on rating methodologies and process, effect of updates on default rate have to be used. Moodys revenues and expenditures has been collected from moodys annual reports from moodys website. Moodys revenue have been used to analyze revenues gained by moodys in the process of rating which is one of the consern rised by one of the author in profissional and acadamic context. As it is qualitative analysis the conclusiosn is from the point of view my research.
6. Analyzing role played by rating agencies in financial crisis: after analysis have been done on financial crisis and rating agencies analysis have to done on role played by rating agencies in financial crisis. This analysis is to done with the help of conclusions drawn from analysis of financial crisis as well as rating agencies. Final conclusions have to be written summarizing role of rating agencies in financial crisis. As above it is qualitative analysis the conclusiosn is from the point of view research.
Speaking through the paper about asset securitization and its effect on the financial sector and the economy as a whole it becomes very important to show the facts and s and how it affected the economy.
Mortgage market has increased in volume till 2006 for various reasons like low interest rate, increase in housing prices, lowering standards of lending. Let us analyze these factors and how these triggered financial crisis.
In response to detiorating economy in 2001 Federal Reserve slashed interest rates to expand money supply and encourage borrowing (-1). Federal Reserve slashed interest rates drastically and touched 1% in the year 2003. In central banking terminology it is zero interest rate. This slashing has encouraged more spending and investing. Low interest rate means banks can lend cheap loans to borrowers. Economy started to grow by the end of year 2002. At the start of year 2004 as economy started to grow and Federal Reserve started to increase the interest rate.
30- Year fixed mortgage rate(fig 2) is the rate borrower takes loan and is fixed throughout the loan period. The interest rate doesn’t change with change in the Federal Reserve interest rate. 30- Year fixed mortgage rates have reduced till 20003 and from then it’s almost constant (-2). Decrease in 30 year fixed mortgage rate means cheap loans.
So reduction in Federal Reserve interest rate as well as reduction in 30- year fixed mortgage rate made loans to be more affordable. In USA, the American dream is to have own home. This sentiment along with affordable loans made borrowers to investment in housing sector.
US federal government to boost this growing sentiment among borrowers to invest in housing sector started organization called National partners in home ownership. The main aim of this organization is to increasing home ownership rate to all time high. To achieve their aim they insisted banks to use FICO score instead of age old point based system. In this program, retailer, home builders, SPV, mortgage bankers are partners so they used FICO scores to accesses credit quality to lend loans. They have used off balance sheet vehicle to avoids placing owners’ equity, liabilities or assets on a firm’s balance sheet and accomplished by placing those items on some other entity’s balance sheet. All these techniques made mortgage banks to lend loans easily for people whom even don’t deserve loans (subprime).
So a perfect platform is laid for housing boom as lending become easy. And people who took mortgages invested in financing new home or refinancing existing homes. Because of this boom in housing prices started to increase ( -3) (-4).
Housing price index (median) is the measure of house prices. It gives better understanding of housing prices trend. If we look at (-3) we can see from the year 1999 till 2006 housing prices started to increase and from then it started to decrease.
Real home price index (-4) is the measure of residential houses price. Home price index includes only those houses under mortgages. So real home price index gives an good idea of how the housing price of those under mortgages increased. During 1999 till 2006 the prices increased rapidly and from there it started to decrease.
So housing price index (median) and real home price index both shows there is increasing pricing trend till 2006. Especially the increase in price of houses under mortgage loans made people borrow more loans and invest in housing sector for more profits.
Proper platform is laid for lending loans and rising housing prices forced many borrowers to take mortgages and invest in booming housing sector for profits.
The rise in mortgage loans can be seen from 5 and 6. The outstanding mortgages as the share of total GDP (percentage) and total mortgages outstanding, the trend is increasing. In 1999 the share of mortgages in total GDP is merely 47.59% where as in 2005 its 69.35% same with total mortgages outstanding i.e. in 1999 its 5055.4 billion dollars and in 2005 its 8660.1 billion dollars.
Non prime mortgages icreseased from 2000 till 2006 and started to drop drastically because of financial crisis. (-7). Analysis data about two thirds of total non prime mortgages consists of subprime mortgages. The subprime mortgages increased rapidly from 457,000 to 2.3 billion in just five years from 2000 to 2005. ALT-A loans too experienced increasing trend from 2000 till 2005. In fact ALT-A percentage increase is much higher than subprime. In 2000 ALT-A constitutes 78,000 loans where as in the year 2005 it’s around 1.2 million.
Ease in lending practices, demand for non prime mortgagees resulted in poorer loan performance. This can be seen from -8 and -9. Delinquency is when borrower defaulted one or two months of scheduled payments. When we see performance of loans from 2000 till 2008 mortgages associated with subprime and ARM has highest delinquency rate and this rate started to decline from 2000 till 2006 and with start of financial crisis it started to climb up. Whereas delinquency rate is almost constant for prime mortgages. Foreclosure is termed when borrower is defaulted and lengthy legal process where institutes repossess the property. Foreclosure rate subprime and ARM loans during the period 2000 till 2008 increased. There is a slight decrease in rate till last quarter of 2005 from there it started to increase. This is increase is due to financial crisis. Whereas foreclosure rate on prime loans is constant almost throughout the period.
The rise in delinquency and foreclosure rate is mainly due to poor practices in lending. And local unemployment rate, downfall in housing prices from 2006 played a major role in rapid increase in rise of foreclosure and delinquency rate.
According to Rousseau Stephane (2009) the loan to value ratio is one more parameter in accessing default risk of the loans. Higher the loan to value ratio higher the borrower likely to get defaulted. For the period of 2000 till 2008 there is the increase in the trend of loan to value ratio.
Easing in lending standards, demand for loans, and rising housing boom, lower interest rates, created perfect environment. Banks started to lend loans to people with poor credit histories. This practice showed it effect on higher credit to loan ratio. And poor performance in loans. At the start of 2006 the unemployment starts to rise, housing prices start to decline. As major share of mortgages consists of subprime borrowers who felt difficult to repay due to worsening economic environment. So delinquency and foreclosure rate increased. This gave way to financial crisis. So the main reasons for financial crisis are ease in lending mortgages, lower interest rates, and government policies.
Non prime mortgages are known to be more profitable than prime loans but are more risky. To get rid of this risk banks pool up these loans and sell it to agencies. These agencies are called SPV. SPV can be government backed or private. Banks pool up prime loans as well but majority of loans consists of non prime which are considered too risky. The SPV divided this pool of loans in tranches and sell to investors. Before selling these tranched securities to investor’s agencies need to get it rated from rating agencies.
Rating agencies main purpose is to rate these tranched securities depending upon the probability of default. Probability of default is calculated using quantitative modeling using statistical models.
Let us consider moody’s investors service. Moody’s are one of the three top rating industries in the world. It is also most respected for its commitment and its good history of good ratings.
The agencies before they want to sell tranched securities in the market for investors they approach moody are to rate these tranched securities. Moody’s use their quantitative model to calculate default of probability. The quantitative model used by moody’s is called moody’s mortgage metrics. Moody’s update their model periodically with new parameters accordingly with changing economic conditions. So let us see what moody’s mortgage metrics is and what updates they made till 2008.
According to (Moody`s Investors service, 2003), In the year 2003 moody’s mortgage metrics published and it is as following.
In this model moody’s incorporated up to date performance information which they got from loan performance Inc formerly known as mortgage Information Corporation. The loan performance information from loan performance Inc is used to calculate frequency of default, cause and effect relationship of macroeconomic variables and loan loss severity stress
Moody’s believe that there is shift in origination practices which are using technology to access borrowers quality, loan risk and appraisal accuracy. So the whole process is streamlined with improved technology, tighter control of lending. However the strict the lending practices are there is huge impact on the loan performance beyond prediction to any statistical model.
There are many refinements to rating approach with stronger emphasis on evaluating the direct sensitivity of loan and pool performance to specific economic stress. Advanced time series analysis increases predictive power and enables meaningful conclusions from even moderately lived vintages.
The rating process starts with the review of an originators and servicers policies and practices. Next comes the analysis of the characteristics of loans in the pool. Then management will allocate corresponding rating accordingly depending up on credit support levels.
A new model has been developed and new parameters have been introduced as inputs. This model is called moody’s mortgage metrics (MMM). Once pool is selected MMM run through each and every loan in the pool and simulates it over 1000 economies. MMM takes in to consideration interest rate and unemployment and real-estate movement to determine the probability of loan defaulting, prepaying in any quarter. Performance of the survivors is also simulated in future quarters. So loss analysis is time and history dependent.
Loan loss severity is the function of loan characteristics and local economy. And it is also known that market movement effects loan severity and the effect is complex.
The models result after simulation of loan performance is pool vector which is aggregate of individual loan results into pool losses for each economy. Now having pool loss distribution in hand moody’s then analyzes the cash flow to the securitization structure and determines expected loss to any supported tranche
Moody’s choose to analyze the likelihood of foreclosure status as a measure of default for loans. Analyzing only loans suffering a loss would have eliminated particularly for loans with deep mortgage insurance, which often generate no net loss to certificate holders even if the property is liquidated.
One means of measuring the predictive power of the model is to graph power curve. Comparing the models prediction of default frequency to cumulative observations of default.
Time series analysis increases predictive power and enables extraction of meaningful conclusions from even moderate short lived vintages and newer product types.
Loan pools have natural distribution of default over time. Initially the defaults are low but with time particularly years between 3-5 it is tricky. These are mainly because of influence of economy like unemployment rate declining market value. So these trigger the default rate to increase. So only by performing Time Series Analysis in the development phase and analyzing new pools using time series model to pool performance accurately reflect the interactions of economic influences on the natural default pattern. Exact shape of economic path is crucial element of loan performance.
Economic conditions such as unemployment rate have its effects on default rate so moody’s included this parameter in their analysis.
Moody’s mortgage metrics uses time series analysis as an appropriate mathematical tool as a factor which effects binary decision.
When analysis individual loan risk in the pool Moody’s stress more on default frequency and default severity. Default frequency is chance or probability that the borrower gets defaulted and will reach foreclosure and default severity is total amount of loss it incurs if the loan gets defaulted.
While evaluating both default severity and default frequency, moody’s consider many input parameters such as;
* Loan to value ratio is one important parameter which gives information on borrower’s equity. Borrower’s equity is very important because it gives information about borrower’s ability to withstand any economic situation. And loan to value ratio gives perfect idea of the borrower’s equity.
* Borrower quality is one main important parameter and plays main role in the outcome of default foreclosure and default severity calculations. Borrower quality is analyzed using borrowers credit score of FICO. Credit score alone doesn’t give borrowers quality so moody’s consider frequency, duration, and time since any mortgage delinquencies, performance on consumer credit loans, length of employment, and income history. The post-mortgage debt service to income ratio (DTI) as other parameters in analyzing borrowers quality.
* Originator and servicer practices are crucial in determining default frequency. Borrower quality alone doesn’t give perfect picture of borrower’s quality. So it is very important to analyze originator practices in lending mortgages such as Past performance of the originator’s loans; underwriting guidelines for the mortgage loans and adherence to them; Loan marketing practices; Credit checks made on borrowers; Appraisal standards; Experience in origination of mortgages; Collection practices; and Loan liquidation procedures.
* Loan term gives an idea principal retirement and borrower’s equity accumulation. Shorter the loan term higher will be monthly installments and longer the loan smaller will be monthly installments. If the borrower doesn’t have enough cash at disposal, it is more likely he get defaulted having a short loan term.
* Time to foreclosure and carrying cost means how much time it takes foreclosure on and dispose the property. The time to foreclose on a property varies depending on different states. At average it takes five to eighteen months for foreclosure to be complete. In calculation of severity it is important to capture all elements of loan losses. Accumulated interest carrying cost in the event of default is key parameters in calculating severity
* Interest rate has profound impact on the severity and foreclosure. Moody’s mortgage metrics uses different economic simulations for varied interest rates to find out probability of default of foreclosure and severity.
* Property type can be single house attached or detached, villas, bungalows etc. detached houses experiences experience less loss severity when compared to attached houses and apartments. So it is very important to add property type in the loss calculations.
* Home value can be important parameter because higher the home value it is more difficult to sell it in the market and to convert it in to liquid cast in case of worsen economic situation.
* Purpose of loan can be for new home or to refurbish existing home. Historical data suggests that mortgages taken to refurbish existing home are less likely to default because their equity is not too bad. Whereas those borrowers with mortgages used to buy new homes are more likely to default because of lowering equity.
* The presence of mortgage insurance covers principle as well as interest payments and expenses to cover to see the property such as legal and real estate expenses. So mortgage insurance will decrease loss severity. Moody’s mortgage metrics incorporated mortgage insurance in the calculation of loss severity. It is very complex calculation method to incorporate insurance in to loss severity calculations.
* Concentration issues like pool size have its effect on calculation of loss distribution. Smaller the size of the pool loss expectation can be calculated where as larger the pool size it becomes difficult to calculate the default.
Moody’s mortgage metrics uses all the above parameters in the calculation of probability of loss. To calculate loss using simulation of various economies (economic simulations). Moody’s simulate future economic stress rather than depending on historical economy as worst possible situation. Moody’s mortgage metrics simulates loan by loan quarter by quarter of all possible economies to calculate probability of default.
Once probability of default is calculated Moody’s then considers its credit enhancement to rate accordingly. If probability of default is low and credit enhancement is sufficient enough to absorb any principal or interest default it is rated top most. and if it does not matches then Moody’s contact originator and advice him to enhance credit. After all these steps Moody’s rate tranched securities and give the rating to originators. (Moody`s Investors service, 1998), These ratings are given in a specific scale so that investors can understand. The rating scale Moody’s followed as;
Aaa rating is supposed to be of the highest quality, with minimal credit risk.;
Aa rating is supposed to be of high quality and are subject to very low credit risk.;
A rating is supposed to be upper-medium grade and are subject to low credit risk.;
Baa rating is supposed to be moderate credit risk.
They are considered medium grade and as such may possess certain speculative characteristics;
Ba rating is supposed to have speculative elements and is subject to substantial Credit risk.;
B ratings are supposed to speculative and are subject to high credit risk.;
Caa rating is supposed to be of poor standing and are subject to very high credit Risk.;
Ca ratings are supposed to be highly speculative and are likely in, or very near, default, with Some prospect of recovery of principal and interest;
C ratings are supposed to be lowest rated class of securities and are in default, with Little hope for recovery of principal or interest.
According to (Moodys Investor service, 2003), Once the rating has been given based on the probability of default, moody’s needed to screen and monitor these securities which may deviate from expected rating because of originator performance or poor performance of loans. Monitoring these securities is second important task of moody. Moody’s use various quantitative methods to screen these outstanding securities (tranched securities) these are called qantitave screens. The qantitave screens are used to review appropriateness of current rating of the tranched securities using the measure of ratio of credit enhancement percent of tranche and expected loss percentage. There are different methods in calculations such as pipe line loss percentage method of calculation, this method is used to calculate losses incurred over a period of time and lifetime loss.
There are two steps in the review process and the first step is to check whether the tranched security should be upgraded or downgraded. And the second step is to check whether the upgrade or down grade action should be warranted.
In the process of review of whether to upgrade or degrade a tranche moody’s first review the accuracy of the data. In general originator review data. And originator dosent has any specific methodology of how they check accuracy of the data. For this reason Moody’s review the accuracy of data using web based analytic tools called performance data service (PDS). PDS is the online service which gives feedback on the moody’s structured finance information.
After the data has been reviewed for its accuracy, the next step is to review pool composition. In reviewing pool composition more moody’s compare current pool composition with original pool composition for the effect of any prepayment on the pool. Moody’s also review securitization historical data to check performance. Moody’s not only review performance of pool that is reviewed but also review other pools of trances issued by the servicer and in some cases other securities issued by other servicer as well. In the review process moody’s given more importance to cumulative losses, recent losses, delinquency percent, loss severity and pool factors. In the review process of performance of current pool moody’s concentrate more on current performance and current trends.
The pool loss analysis uses quantitative analysis techniques. After screening for accuracy of data, moody’s calculated estimated lifetime loss, deal structure, excess spread benefit. Different methodologies used calculate lifetime losses are committed loss, straight line loss, loss curve, annualized loss, pipeline loss, lifetime loss based on delinquency.
The committed loss is calculated when current cumulative loss percent is subtracted from original estimated life time loss percentage and divided by current pool factor. This committed loss receives lot of importance when the pool is relatively new. In straight line loss calculation, it is assumed that the pace of loss till now remains same in the future for remaining life of transaction. The loss curve calculation incorporates an estimated loss curve as well as a 12 or 18-month pipeline loss to determine an estimated loss percent. Annualized loss is calculated when previous year’s annualized loss is multiplied with projected average time left.
Once the lifetime loss is calculated then moody’s check if any credit enhancement is there to cover this losses. If so moody’s calculate the amount of excess spread benefit. Moody’s then simulates different cash flow scenarios and review excess spread benefit as well how good the pool is in absorbing excess spread.
Once all the review process is done, moody’s consulted the originator and credit enhancers to clarify any doubts in the process of calculation. Once it’s clear moody’s decide whether to upgrade or degrade the rating based on the lifetime loss.
Moody’s constantly update their models to capture risk involved in calculating probability of loss. These updates are continuous effort of Moody’s to keep their models updated for changes in the market. Moody’s update few parameters for the basic model to capture accurately the market trends and to accurately calculate probability of loss.
According to (Moody`s Investors service, 2005), Moody’s has updated few parameters to their model in December 2 2005. They have updated interest rate assumptions for the calculation of cash flows. And minor changes to the lifetime loss calculation methodologies. A volatility curve is used to assess the distribution of loss coverage estimates across the rating spectrum. Moody’s has made minor changes to the volatility curve used in the calculation of lifetime losses.
In the analysis of credit spread three new interest rate stresses are included. These interest rate stresses are;
1. Standard interest rate stress;
2. LIBOR curve;
3. A curve which is midway between LIBOR curve and standard interest rate curve.
According to (Moody`s Investors service, 2006), Moody’s continued to update their methodologies in rating RMBS. They have made number of refinements in the year 2007. These changes are made because of changing dynamics of the loan performances (bad performance of 2006 RMBS vintages) in the year 2006. To address the changing loan performances Moodys changed their methodologies to incorporate risk involved in the loans and its attributes. The changes made are increased transparency in the rating, expansion of loan characteristics, delinquencies and its triggers.
The new changes in the year 2007 are;
1. Default assumptions for highly leveraged loans;
2. Risk assumptions for stated income loans and low and no documentation loans;
3. Risk assumptions for newly originated loans;
4. Modified Delinquency triggers;
5. Modified excess spread compression;
6. Increase in loss expectation;
7. Expansion of loan attributes for subprime loans;
8. Increase in loss expectations for early delinquencies;
9. Higher originator specific loss coverage.
1. Default assumptions for highly leveraged loans: Moody’s generally included risk involved with borrowers with less or no equity in to loss expectation calculations. At the end of 2006 with the fall of housing boom, there is more risk involved with the borrowers with less or no equity. So moody’s kept in mind the changing economic situations and increased it risk assumptions for frequency of default of the loans by 25% and sometimes to 90% based on geography, FICO score, documentation type and occupancy type.
2. Risk assumptions for stated income and low or no documentation loans: loans associated with stated and low or no documentation loans have higher probability of default when compared to its counter partners. Moody’s generally differentiate loans with stated income and low and no documentation loans with full documentation loans in the calculation of probability of loss. In the year 2006 the performance of slated income and low and no documentation loans is poor. So moody’s had increased 20% to 25% for risk assessment of the loans consisting of slated income and low and no documentation. And this increase in risk assessment percent reflected in loss expectations.
3. Risk assumptions for newly originated loans: the loans that originated in the year 2006 performed very badly and have increased first or early payment default rate. Keeping previous year performance in mid moody’s increased loss assessment by 10% for the first payment. And this risk will decrease when borrower starts to pay properly the monthly installments. And once the borrower started to pay monthly payments timely this loss assessment will reduce to 0%.
4. Modified delinquency triggers: delinquency rate is one important parameter in calculating credit performance of the transaction. Delinquency rate decide how much credit enhancement is needed. Moody’s generally consider delinquency as loans with 60 days or more past the due. Moody’s believe that calculation of credit performance and credit enhancement to be more effective they need to add more triggers in the calculation of delinquency rate. Moody’s anticipate more loan modifications than in future in order to reduce losses. The originator should provide more refinancing options to mitigate the losses.
5. Modified excess spread compression: reduction in housing prices and high interest rate and tightening lending standards may affect subprime borrowers and are more likely to default and will experience payment shocks and have fewer refinancing options. Moody’s believe that originator of the loans should help these borrowers at the brink of default by changing their adjustable rate of interest to fixed rate of interest and helping them to refinance. This may reduce losses and have effect on excess spread. To reflect the greater expected levels of modifications, Moody’s will assume that 10% of the subprime loans will remain at their initial fixed rate over their life.
6. Increase in loss expectation: the poor performance of RMBS in the year 2006, moody’s has revised its methodology and increased average loss expectation by 25% and greater for certain servicers whose performance is poor. And each and every loan is analyzed in the pool in loan level basis if the information is not provided assumptions of 25% loss expectation is factored in to loss expectations.
7. Expansion of loan attributes for subprime loans: moody’s has added more parameter in the loan level analysis of probability of default. These new parameters are information of credit history of borrowers in detail, the level of reserves of the borrower, information of borrowers evading taxes and insurance, whether the borrower is first time home owner. Moody’s modified their expected loss of loans by first time buyers by 40% compared to seasoned borrower, first time ownership borrowers are 25% more riskier than their counter partners, and self employed borrowers are 10% more riskier than slated income borrowers. These changes in the loan level analysis are to make sure present economic market is captured properly and probability of default is as accurate as possible.
8. Increase in loss expectations for early delinquencies: the delinquency rate of loans originated in the year has more early delinquency rate than any year and maximum percent of these delinquencies are then foreclosures. Keeping in mind this poor performance, moody’s has adjusted their loss expectations of delinquency rate for new transactions. Moody’s have also incorporated report of lag in to the analysis.
9. Higher originator specific loss coverage: based on the previous performance of the originator securities moody’s has increased it loss expectations by 20% for poor performance originators when compared to better performance originators. This differentiation in the originator based loss expectations is because of poor performance of some originators of RMBS when compared to other originators.
Whether changes in the Moody’s methodologies have resulted in increase downgrades in the ratings. The analysis of effect of these changes in moody’s methodologies on default rate of securities is studies in next section.
In this section first let’s analyze issuance and default rate and later in the section analyze the effect of updates on default rate.
The servicer wants the rating from the rating agencies for their securities marketability. In this section we will analyze the issuance and default rate of these securities over a period of 1999 till 2007.
According to (Moody’s Global Credit Ploicy, august 2008), Issuance of new ratings for structured finance products has an increasing trend from 1999 till 2006 where it peaked with more than 25,000 issues. On the other hand it’s less than 5000 in the year 1999. Issuance of new rating for US RMBS sector too has the same trend and had its peak value in 2006 and there after it started to decrease.
In 2007 new rating issuance started to decline. 22,465 new structured finance ratings were issued. This number when compared to 2006, where it new rating issuance peaked is 20% less. This decline trend is mainly because of poor performance of mortgage backed sectors. US HEL AND us RMBS new issuance fell 44% and 19% respectively.
In 2008 issuance of new ratings has further went down by 87% when compared to new ratings issued previous year(2007) and down by 91% when compared to 2006 where new rating issuance peaked. Issuance fell sharply in US HEL and UD RMBS sectors.
According to (Moody’s Global Credit Ploicy, august 2008), There is increase in number of rating outstanding from 1999 till 2008. Till 2007 US RMBS has the largest share of rating outstanding. In 2007 new issuance fell 20% but rating outstanding increased. The main reason for the increase is the rapid issuance in the previous years. In 2008 the case is same where rating outstanding increased when compared to 2007. Highest single share of increase of 25% is recorded in US RMBS sector.
According to (Moody’s Global Credit Ploicy, august 2008), To differentiate default between corporate and structural finance, moody’s introduced material impairments concept in 2003. These material impairments are further classified as principal impairment and interest impairment. Principal impairments are those securities having principal write-downs or losses and the securities that are downgraded to CA or C even if they did not have any interest shortfall or principal write-down. Interest impairments are those securities that are not principal impairments and having outstanding interest shortfalls. (Moody’s Global Credit Ploicy, august 2008)
Total impairments have steadily increased from 2001 till 2004. In 2005 with the introduction of new structured finance products which has increased liquidity in to the market, and strong real estate market and low interest rate resulted in low impairments. These impairments are in par with those of 1999-2000 and lower than those observed during the period 2001-2004. In 2006 impairment rate increased slightly. In 2007 there are more impairments in fact these impairments are so huge such that when we sum of all impairments of previous 14 years it nearly equals impairments of year 2007. This is mainly because of home price decline, tightening credit standard followed by banks, increase in interest rate. Of the total impairments in the year 2007 USRMBS has the maximum share. In the year 2008 same trend continued because of collapse of housing market. Total impairments crossed 12000 marks. This impairment rate has effect on whole of the structure. This impairment rate is alarming such that 12 month impairment rate has increased to record high.
When we compare impairments on yearly basis, impairment rate is higher on the 2005 to 2007 vintages especially those of 2006. And largest number of impairments are those that of subprime first liens, and then closely followed by ALT-A, subprime seconds.
US RMBS sector played a major role in the total impairments from the starting. Initially it has huge share but at the starting of 2000 because of lowering of interest rate, favored conditions for housing market, the impairment rate in US RMBS sector decreased. This decrease was short lived for couple of years and kick started to all time highs in the years 2007 and 2008. The main reason are bad housing market, banks become stricter in lending loans.
Due to increase in housing prices and interest rate slashes and ease in lending of mortgages resulted lending more mortgages. Banks in turn securitized these loans and got extra liquidity to lend more loans. The extra liquidity in the hands of banks and demand for non prime mortgages made banks to lend more loans to non prime borrowers. All these mortgages are pooled and sold to agencies who are SPV. These agencies then convert these pools of loans in to securities and sold to investors. Before selling to investors these securities need to get rated from rating agencies for its marketability. Due to ease in lending practices there is sharp increase in issuance of these securities till 2006 and with the start of financial crisis lot of securities are frozen with agencies and that is why there is sharp decline in the issuance from the year 2007 till 2008. Whereas rating outstanding depends on previous year’s issuance it has slight increase. Due to poor lending practices, the impairments rose from the year 2007. These impairments are mainly due to poor lending of mortgages to subprime borrowers.
Moody’s have updated their model in the 2005 and later updated in the years 2006 and 2007 as well. The major updates started in the year 2006. When we compare the default rate it has sharp increase in the year 2007 and default rate sky lined in the year 2008. The effect of changes to the rating methodologies in the year 2006 has huge effect and can be seen in the default rate in the year 2007. Later in the year 2007 more updates has been introduced in the rating methodology which resulted in the highest default rate.
(Committee on the Global Financial System, 2005), and (Mason and Rosner, 2007) has argued that the rating methodologies used by rating agencies (in this case we consider moody’s) to rate RMBS products is in appropriate. Rating agencies rate both the products in the same scale. This can be analyzed by comparing the economics of both RMBS products (tranched securities) with corporate products (corporate bonds).
The main aim of rating is to show how much risk is involved to investor’s before they invest. Moody’s rate both RMBS and corporate bonds in same scale irrespective of risk involved in both are different.
If we consider corporate bonding structure its dynamic in nature where as RMBS structure is static in nature. The corporate bond structure is dynamic in nature because corporate agencies invest in the pools of investment projects. These projects can perform well and it can under perform as well depending up on the situation and conditions. Corporate agencies can change their strategy on how these investment projects are carried out. In case of any losses, the investment company can enhance the credit by adding more funds there by adding much need equity. The enhancement of credit can protect investors from increased risk. This flexible nature of corporate bonds made them dynamic in nature.
If we consider RMBS structure, it is static in nature. It is static in nature because it has fixed pool of assets. The performance of these pools depends on borrower’s ability to repay timely installments and principal. These pools can never outperform as there is no scope for outperform because it is based on payments by borrowers which are fixed. There are more chances of these pools to underperform because there are more chances of borrower getting defaulted in the payments due to social and economic reasons. So change in originator standards can highly influence the performance of these securities.
An investment in corporate bonds is an investment in a proven ongoing dynamic investment strategy. An investment in a mortgage pool is an investment in a static pool of unproven fixed-income investments.
Issuers of structured finance products wanted RMBS to be rated on the same scale astraditionalbonds so that investors think structured finance has same kind of risk that of bonds.(Mason and Rosner, 2007). This can be clearly proved here that despite of differences in the risk rating agencies rate RMBS and corporate bonds on same scale thus creating conflict of interest.
According to(Committee on the Global Financial System,2005), and(Mason and Rosner, 2007) rating agenciesNeverdisclosed completely theirmethodologies they use to rate RMBSand keyassumptions and rating criteria. This can be proved by looking at the moody’s models used for the rating and their methodologies. The methodologies were not clearly explained and there are many parameters those are not explainedin the models andare used by analyst according to situation in the rating process. This practicerise concerns about rating process and transparency of the rating process.
The rating agencies main role is to act as an intermediate between investors and issuers. This trust of being intermediate has been broken by rating agencies by charging issuers for rating products instead of getting paid by subscribers who subscribe for these ratings to invest in these products. Because of shift in the axis of being intermediate, these rating agencies got paid from issuer who in turn profited rating agencies by gaining millions of dollars.
If we look at 16 which gives information of net revenue of the moody’s over a period of eight years from 2000 till 2008, we can easily find the net revenue moody’s increased rapidly over years till 2007 where credit crisis has halted the growth. Because of crisis new issuance has decreased there by decrease in the net revenue. But due to past issuance and rating outstanding moody’s still get revenue from the servicers.
If we look at 17 which gives information of percentage share of revenues by structured finance products. The share of revenues by structured finance products slowly increased from above 25% and peaked 43% in the year 2006. This shows how much revenue moody’s earned with the help of structured finance products.
This process of issuer paying for his rating created conflict of interest. So considering profits they incur from this new role, rating agencies tend to rate products issued by these issuers a higher rating than they actually are. The issuer has ability to adjust deal structure to get desired rating. And issuer has influence on rating process.
(Daníelsson J, 2002) said the statistical model to be accurate it should not have Endogenous risk, Quality of assumptions should be good, and Data quality should is good. If we consider moody’s model for rating RMBS, there is Endogenous risk involved because moody’s mortgage metrics try to model RMBS which is static in nature and doesn’t have aggregate behavior. It is easy to model an institution using statistical model but when modeling structure moody’s play catch up game. RMBS performance is affected by the borrower’s ability to repay the loans so moody’s play catch up game to incorporate borrower’s performance for accurate results. (Vanessa G. Perry, 2008) proved, there is always dearth of data on subprime market. The data that is available is proprietary lender data. And this data had drawbacks on analysis of market trends. So moody’s model is not good enough for rating RMBS. The results which come are just catching up of present situation and it doesn’t capture any future risk.
Quality of assumptions in modeling risk is very important and moody’s have their own assumptions in moody’s mortgage metrics. Initial models of moody’s (2003) have limited assumptions on borrower’s performance. As RMBS performance is based on borrower’s performance it is important that borrowers performance need to be incorporated in the model. This mistake has been rectified by moody’s in the later updates in the year 2007. This update is too late as the crisis has already started and new assumptions on borrowers has increased downgrades in the ratings which is also one more reason for credit crisis.
Data quality is one more important element in statistical modeling. If the data quality is not good the result of the model is not accurate enough to be true. Moody’s generally get data from servicer about the underlying loans. Moody’s doesn’t cross check the accuracy of the data with any third party agencies. This draw back has been rectified later in 2007 when data quality has been rechecked using third party agencies.
According to Tom Bulford (2008), (Ruth Rudden, 2007) “The credit rating agencies like Moody’s played a central role in financial crisis. this can be analysed by looking at the rating models, updates they did till now to the rating models, and short comes in the modelling and rating process.
Servicer (agencies or SPV) buys loans from the banks and convert them in to securities. These securities to be sold in the market they need to be rated by these rating agencies like moody’s. And these rating agencies get paid for rating these securities. This process of earning money has created conflict of interest and change in the axis of rating agencies role from market intermediates. Rating agencies like moody’s been favouring these agencies and made millions of dollars of profits by rating these agencies.
Rating agencies like Moody’s relied on data provided by these agencies for their models. They have never verified the accuracy of the data provided by the agencies. And models used by rating agencies are not that accurate if the data is not accurate the results from these models are not accurate. And their methodologies lack transparency and had endogenous risk involved it. Finally these models predict probability of default for present and are not accurate for calculating future probability of risk. So rating agencies like moody’s started to play catching up game by updating their models for changing economic conditions.
As financial crisis started, Moody’s updated their models with more input parameters to capture the present economic situation and result is downgrading of many top rated tranches. These downgrades created uncertainty and doubt on quality of rating these rating agencies assigned. With more exposure to risk related to subprime debts, restricted liquidity of banks, the aftermarket for term loans has effected so there is sharp increase in risk premium. Thus banks lost confidence and have less liquidity. This resulted in present financial crisis.
From the chapter one which is introduction, motivation for choosing the research topic “role of asset securitization” and introduction to the research has been discussed.
From chapter two which is professional and academic context, lot of articles and publication, and working papers have been studied. The views of different authors on the research topic have been written down in to three sub chapters namely 1.Asset securitization; 2 Rating agencies; 3 financial crises. The study which has been written focused mainly on background and economics of asset securitization. Further, the chapter explains views of authors at evolution of rating agencies and its methodologies and rating procedures for rating and concerns of rating methodologies and practices and role of rating agencies in financial crisis. And finally different authors view on evolution of financial crisis and role of government in financial crisis and finally financial market turmoil.
From third chapter which is on methodology adapted to reach the aim of the research Quantitative and qualitative methods are employed and quantitative methods are employed to validate qualitative methods. Further in the chapter, data source has been discussed and limitations of the data used have been discussed as well. Finally analysis methods employed to reach the research question has been discussed.
Finally from chapter four which is data analysis, we can conclude that sub standard mortgage lending practices have highly influenced by the incentives of the government. The innovation in lending practices, involvement of government, the rising home prices, decrease in federal reserve interest rate create demand for nontraditional mortgages such as non prime mortgages.
The mortgage originators found new innovative methods in lending to cope up with the borrowers demand this could have been done only with the help of government intervention. The aggressive lending methods and increase in non prime mortgages led to increase in credit risk for the borrowers and investors as well.
As the interest rate increased, the borrowers found it difficult to repay monthly installments as many borrowers are non prime borrowers with low credit history. So with the rise in interest rate these borrowers found it difficult to cope up with increased monthly installments. This in turn increased delinquency and foreclosure rate. The final blow came when housing prices started to decrease. The condition got much worse and additional stress contributed for much higher delinquency and foreclosure rate.
Banks in turn securitized these toxic loans and got extra liquidity to lend more loans. The extra liquidity in the hands of banks and demand for non prime mortgages made banks to lend more loans to non prime borrowers.
All these mortgages are pooled and sold to agencies who are SPV. These agencies then convert these pools of loans in to securities and sold to investors. Before selling to investors these securities need to get rated from rating agencies for its marketability. Rating agencies charged the agencies to rate the RMBS thereby there is change in role played by rating agencies. This change is role of rating agencies earned them millions of dollars of money. To rate these securities Moody’s used statistical models.
Rating agencies like moody’s relied on data provided by these agencies for their models. They have never verified the accuracy of the data provided by the agencies. And models used by rating agencies are not that accurate if the data is not accurate the results from these models are not accurate. And their methodologies lack transparency and had endogenous risk involved it. Finally these models predict probability of default for present and are not accurate for calculating future probability of risk. So rating agencies like moody’s started to play catching up game by updating their models for changing economic conditions.
As risk involved is not clearly assessed by these banks they started to rate these tranches high enough. As the market started to meltdown because of financial crisis, moody’s updated their methodology accordingly to changing economy and reviewed the securities and downgraded many top rated securities. These downgrades created uncertainty and doubt on quality of rating these rating agencies assigned. With more exposure to risk related to subprime debts, restricted liquidity of banks, the aftermarket for term loans has effected so there is sharp increase in risk premium. Thus banks lost confidence and have less liquidity. This resulted in present financial crisis.
1. Joseph R.Manson and Joshua Rosner. (2007), “Where Did the Risk Go? How Misapplied Bond Ratings Cause Mortgage Backed Securities and Collateralized Debt Obligation Market Disruptions”, Working Paper Series, November 11, Social Science Research Network,< https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1027475>, viewed on 18th November, 2009
2. Ruth Rudden. (2007), “Evolution of Credit Ratings—Part 1”, Internal Presentation, March 14, CariCRIS, <https://www.caricris.com/pdfs/article/evolutionpart1.pdf>, viewed on 8th November, 2009
3. Daníelsson, J.(2002), “ Blame the models”, Journal of Financial Stability 4, PP. 321-328
4. Report submitted by Committee on the Global Financial System (2005). Bank of international settlements: “The role of ratings in structured finance: issues and implications”[online], < https://www.bis.org/publ/cgfs23.pdf>. Viewed on 14 November 2009.
5. Vanessa G. Perry. (2008), “The Dearth and Life of Subprime Mortgage Data: An Overview of Data Sources for Market Modeling”, Working Draft, January 8, The Hoyt Group, < https://www.hoyt.org/subprime/vperry.pdf>. viewed on 28th October , 2009
6. Rousseau Stephane. (2009), “regulating credit rating agencies after the financial crisis: the long and winding road toward accountability”, Working Paper Series, June 23, Capital Markets Institute, Université de Montréal, <https://papers.ssrn.com/sol3/papers.cfm?abstract_id=1456708 >. viewed on 8th November, 2009.
7. Federal Reserve Bank of San Francisco, (2007). The Subprime Mortgage Market, Annual Report 2007. [online]
< https://www.frbsf.org/publications/federalreserve/annual/2007/subprime.pdf>, Accessed 14 November 2009
8. Kaptan, S and Telang, C. (2002), “Innovative concepts in finance”, New Delhi, Sarup & Sons.
9. Greenbaum, S.I., Thakor, J.V. (1987), "Bank funding modes: securitization versus deposits", Journal of Banking and Finance, Vol. 11 No.3, pp.379-92.
10. Hatice Uzun, Elizabeth Webb, (2007) “Securitization and risk: empirical evidence on US banks”, The Journal of Risk Finance, vol-8, no 1, pp-11-23.
11. The Staff of the Securities and Exchange Commission, (2008), “Summary Report of Issues Identified in the Commission Staff’s Examinations of Select Credit Rating Agencies”, [online], < https://www.sec.gov/news/studies/2008/craexamination070808.pdf>, viewed on 14 November 2009.
12. Tom Bulford, (2008), who will rate the rating agencies? Fleet Street Invest, Sept 01.
13. Souphala, C and Anthony, P.C , (2006), “The Evolution of the Subprime Mortgage Market”, Federal Reserve Bank of St. Louis Review, January/February 2006, 88(1), pp. 31-56.
14. Dow, S (2002), Economic methodology: an inquiry, Oxford University Press Inc., New York.
15. Silverman, D. (2000), Doing Qualitative Research. A Practical Handbook. London, Sage.
16. Kinnear & Taylor (1996), ‘Focus versus nominal group interviews: a comparative analysis’, Journal on Marketing Intelligence & Planning
17. Kothari, C R (2005), Research Methodology: Methods and Techniques, New Age Publishers.
18. Moody`s Investors service. (2003). Moody`s Mortgage Metrics : A model analysis of residential mortgage pools. Moody`s Investors service.
19. Moody’s Global Credit Ploicy. (august 2008). Default and loss rate of structured finance securities:1993-2008. Moody’s .
20. Moody’s Global Credit Ploicy. (august 2007). Default and loss rate of structured finance securities:1993-2007. Moody’s .
21. Moody’s Global Credit Ploicy. (august 2006). Default and loss rate of structured finance securities:1993-2006. Moody’s .
22. Moodys Investor service. (2006). US Subprime-Overview of Recent Refinements to Moody’s Methodology: July 2007. Moodys investor service.
Mortgage market can be classified in to two categories based on credit quality, those are
1. Prime – prime is associated with borrowers with very good credit history and entitled to attractive interest rates and terms.
2. Non-prime- non prime is sub divided in to two categories
* ALT-A- ALT-A is associated with borrowers with credit history which is not good enough to be prime but enough to get loan with some restrictions on interest rate and monthly payment (higher interest rate and low monthly payment)
* Subprime—a subprime is associated with borrowers with very bad credit history and has to pay high interest rates.
3. There are two types of loans those are fixed rate mortgages which are has interest rate fixed throughout life of loan and second type is Adjustable Rate Mortgage (ARM) where interest rate is variable.
4. The following are the terminology used to describe performance of the loans.
* Current – borrower meets scheduled payments regularly.
* Delinquent – borrower defaulted one or two months of scheduled payments.
* Default – if the borrower haven’t repaid scheduled payment for more than 90 days.
* Foreclosure – if borrower is defaulted and lengthy legal process where institutes repossess the property.
5. The nonprime market segment featured a number of non-traditional products and characteristics:
* Hybrid ARM—Interest rate is fixed during an initial period then “resets” to an adjustable rate for the remaining term of the loan.
* Payment-option ARM—Borrower has multiple payment options each month, which may include minimum payments lower than what would be needed to cover any of the principal or all of the accrued interest. This feature is known as “negative amortization” because the outstanding loan balance may increase over time.
* Interest-only—allows the borrower to pay just the interest on the loan for a specified period, usually the first 3 to 10 years, thereby deferring principal payments.
* Low and no documentation loans—require little or no verification of a borrower’s income or assets.
* High loan-to-value (LTV) ratios—Borrower makes a small down payment, causing the ratio of the loan amount to the home value to be relatively high.
* Prepayment penalties—Borrower incurs a fee if he or she pays off the loan balance before it is due.
Federal Reserve Interest Rate
TABLE-1 Federal Reserve Interest Rate
30-Year Fixed-Rate Mortgages
Total Mortgages Outstanding
house onwership rate
Outstanding Mortgages as a Share of GDP (percent)
housing prices (median)
low or no documentation
loan to value
Delinquency Rates all loans
Delinquency Rates prime
Delinquency Rates sub prime
Delinquency Rates ARM
Started all loans
Started sub-prime ARMs
% change in principal impairments
% change in interest impairments
us rmbs +HEL/
% of us rmbs in total impairments
data on number of principal impairments, interest impairments, total impairments, percentage change in principal impairments, percentage in intrest impairments, number of impairments in US RMBS sector, percentage change of US RMBS impairments in total impairments.
We will send an essay sample to you in 2 Hours. If you need help faster you can always use our custom writing service.Get help with my paper