DOC

On the Design of Collateralized Debt Obligation-Transactions

By Wendy Bailey,2014-11-25 19:42
6 views 0
On the Design of Collateralized Debt Obligation-Transactions

    How does the market handle information asymmetries in

    securitizations?++

    Günter Franke*

    Markus Herrmann**

    Thomas Weber*

    October 2007

    Abstract

    The strong growth in collateralized debt obligation transactions raises the question how these transactions are designed. The originator designs the transaction so as to maximize her benefit subject to requirements imposed by investors and rating agencies. An important issue in these transactions is the information asymmetry between the originator and the investors. First Loss Positions are the most important instrument to mitigate conflicts due to information asymmetry. We analyse a model to study the optimal size of the First Loss Position and then a set of European collateralized debt obligation transactions. We find that the asset pool quality, measured by the weighted average default probability and the diversity score of the pool, plays a predominant role for the transaction design. Characteristics of the originator play a small role. A lower asset pool quality induces the originator to take a higher First Loss

    Position and, in a synthetic transaction, a smaller Third Loss Position. The First Loss Position

    bears on average 86 % of the expected default losses, independent of the asset pool quality. This loss share and the asset pool quality strongly affect the rating and the credit spread of the lowest rated tranche.

    JEL classification: G 10, 21, 24

    Keywords: Securitization, collateralized debt obligations, asset pool quality, First Loss Position, synthetic transactions, tranching.

    * University of Konstanz, department of economics, D-78457 Konstanz, POB D 147. E-mail: guenter.franke@uni-konstanz,de, thomas.a.weber@uni-konstanz,de.

    ** HSBC Bank, global research, London. E-mail: markus-herrmann@hsbcgroup.com

    ++ We are very grateful to valuable comments of Michel Habib, J Krahnen, Peter de Marzo, Nandu Nayar, Branko Urosevic, the participants of the European Skinance Workshop 2007,

    the European Financial Management Association Conference in Vienna and the European Finance Association Meeting in Ljubljana. In particular, Kjell Nyborg suggested many very

     helpful improvements.

    1. Introduction

    Over the last 20 years the volume of securitizations has grown tremendously. The global volume of securitization issuance was estimated to be roughly 270 bn USD for 1997 and about 2100 bn USD for 2006 (HBSC (2007)). The recent subprime-crisis depressed the issuance volume. Securitizations were accused of fostering intransparancy of bank risks which dried out the liquidity in the interbank market. Whether the intransparancy was generated by the securitizations or by the complexity of structured investment vehicles investing in securitization bonds, is an unsettled empirical question. It is also controversial whether securitizations have positive or negative effects on financial stability. In any case, many financial intermediaries use securitizations for their management of default risks. Given the importance of securitizations, there is amazingly little research on securitizations. A subset of these securitization-transactions are collateralized debt obligation (CDO)-transactions. They can be collateralized loan obligation (CLO)- or collateralized bond obligation (CBO)-transactions. In the former case a bank typically securitizes part of its loan portfolio. In the latter case the originator of the transaction, a bank or an investment manager, buys bonds, and sometimes in addition some loans, pools them in one portfolio and sells the portfolio to investors.

    This paper analyses an important aspect of CDO-transactions. Given information asymmetries between banks and investors about the quality of securitized loans or bonds, investors are concerned about buying lemons and, therefore, insist on credit enhancements in securitizations which mitigate potential problems of information asymmetry. If a bank, for example, securitizes the interest and principal payments of many loans granted to small and medium sized enterprises, then investors know little about these obligors, relative to the bank. This provides room for adverse selection and moral hazard of the bank. Since investors penalize the bank for information asymmetries, she therefore attempts to mitigate their effects. In a perfect capital market these problems would not exist. Therefore securitization research needs to focus on market imperfections to understand the design of securitization transactions. Information asymmetries, transaction and management costs, costs of financial distress, costs of equity capital, other regulatory costs and liquidity premiums appear to be important. Besides, management costs of investors and of the bank including those of involved third parties play a role in the securitization process. These costs include the costs of setting up the transaction (internal costs of the originator, fees of lawyers, rating agencies, custodians etc.) and the costs of managing the transaction after the setup. They are incurred by the originator and the investors buying the securities. Securitization, thus, induces various costs which pose

     2

    a barrier to securitization. Securitization makes sense only if these costs are overcompensated by some benefits. These may come from better risk allocation across agents, a reduction of the bank’s cost of required equity capital, other regulatory costs and funding costs. Moreover, the transfer of default risks in a securitization gives the bank the option to take other risks. The purpose of this paper is to add to the understanding of the design of securitization transactions by analysing credit enhancements which take care of the information asymmetries. The most important credit enhancements are contractual obligations of the originator to bear default losses of the asset pool underlying the transaction. In all transactions there exists a First Loss Position (FLP) which bears all default losses up to a given limit. The FLP is defined as a fraction of the volume of the securitized asset portfolio. Investors only bear default losses beyond the FLP. The higher the FLP, the more are investors protected against default losses and, hence, against problems of information asymmetries. In synthetic transactions, investors usually bear only part of the default losses beyond the FLP. They take a limited second loss position (SLP) and the originator takes the third loss position (TLP) by not selling the super-senior tranche. He may buy protection against the losses of the TLP through a senior credit default swap. Similarly, the originator need not retain the FLP, but may sell part or all of it. But is not publicly known to what extent the originator retains the risks of the FLP and the TLP.

    The market imperfections mentioned above pose a challenge to the originator of a transaction. How should she design the transaction so as to maximize her net benefit? There are many degrees of freedom in setting up a transaction. For example, what should be the quality of loans/bonds serving as collateral for a transaction? How many loans/bonds should be included? These choices determine the asset pool quality and the associated information problems. Given this choice, how large should the FLP be so as to mitigate problems of information asymmetry? Should the transaction be structured as a true sale- or a synthetic transaction so as to allow for a TLP? How large should be the TLP? These questions can only be answered taking into consideration not only the needs of the originator, but also the needs of investors. They insist on a solid design of the transaction so as to protect them against potential losses due to information asymmetries We try to answer these questions by, first, analyzing the optimization problem of the originator and deriving hypotheses about an optimal design. Second, we investigate a European set of securitization transactions to test these hypotheses. In the empirical analysis, we not only investigate the choice of the asset pool and of the FLP, but also the lowest rated bond tranche sold to investors. This tranche can be viewed as the mirror of the FLP since the FLP determines the protection of the lowest rated tranche against default losses. Therefore the characteristics of the lowest rated tranche help to understand the importance of the FLP from the perspective of the investors. To our best knowledge, this study is the first to analyse the interaction between the quality of the securitized asset pool, the other choice variables and originator characteristics.

     3

    The main findings of the paper can be summarized as follows. First, a theoretical model shows that the FLP should be inversely related to the quality of the securitized asset portfolio. The FLP should increase when the portfolio quality declines. The quality of the securitized asset portfolio is measured by its weighted average default probability (WADP) and by Moody?s diversity score (DS). A lower WADP and/or a higher DS improve the asset pool quality. The empirical evidence confirms that the FLP increases when the asset pool quality declines. We interpret this as evidence that a lower portfolio quality reinforces problems of asymmetric information which are mitigated by a higher FLP.

    Second, the qualitative finding that a lower asset pool quality raises the FLP does not tell us how this position is quantitatively determined. Therefore, we investigate two transformations of the asset pool quality into loss sharing characteristics, assuming a lognormal distribution for the default loss rate of the underlying portfolio. The first characteristic is the share of expected default losses absorbed by the FLP. The second characteristic is the probability that all default losses are fully borne by the FLP, i.e. investors are not hit. This probability equals the cumulative probability of the loss rate distribution of the asset pool at the FLP. We denote it as the support-probability of the FLP. (1-the support-probability) is the probability that investors are hit by default losses. In particular, it is the probability that the lowest rated tranche, i.e. the tranche with the lowest rating, is hit. Its rating is determined by this probability according to S&P.

    We analyse the loss share and the support-probability, assuming a lognormal loss rate distribution. Empirically, it turns out that the share of expected default losses, with a mean of 86 %, is independent of the asset pool quality. This indicates that a share of 86 % is the guideline for the market which may be influenced to some extent by other considerations. A constant share of the expected default loss implies for the lognormal model that the support-probability of the FLP depends inversely on the WADP and, surprisingly, also inversely on the DS. This is confirmed by the empirical findings. These findings gives us a rather precise understanding of how the market copes with information asymmetries in securitizations. Third, the loss share of the FLP and asset pool quality are quite powerful in explaining empirically the rating and the credit spread of the lowest rated tranche. But the credit spread of the lowest rated tranche is better explained by its rating, its maturity and the date at which the transaction is arranged. This underlines the important role of the rating agencies. Fourth, the attractiveness of a synthetic relative to a true sale transaction increases with the portfolio quality. Hence TLPs are more likely for transactions with better portfolio quality. Better quality implies a lower default risk of the super-senior tranche, making it less attractive for the originator to buy protection on this tranche through selling it or buying a super-senior default swap. The preference for synthetic transactions is stronger for originators with a better

     4

    rating. Presumably, for highly rated originators funding through standard bonds is cheaper than through true sale transactions. Retention of the super-senior tranche is in strong contrast to the literature which argues that the originator should sell the least information-sensitive tranche. The size of the super-senior tranche, i.e. the size of the TLP, increases with the

    portfolio quality, in contrast to the size of the FLP which is inversely related to portfolio

    quality. This indicates the different nature of the FLP and the TLP. The FLP appears to be important for investor protection while the TLP does not and , therefore, is driven by other considerations.

    Fifth, surprisingly, characteristics of the originator like her total capital ratio, Tobin?s Q and other variables which proxy for her securitization motives, add little to the explanatory power of the regressions. This indicates that the design of securitization transactions depends little on these characteristics. Essentially, rating agencies and investors appear to be the dominant forces.

    The paper is structured as follows. In section 2 the relevant literature is discussed. In section 3 we model the originator’s optimization problem and derive hypotheses about her choice of the transaction design. The empirical findings are presented in section 4 and discussed in section 5. Section 6 concludes.

    2. Literature Review

    The design of a CDO-transaction regarding the handling of information asymmetries is a complex task. In order to relate it to the literature, we first characterize CDO-transactions.

    1Depending on her motives, the originator selects a set of loans or/and bonds as the

    underlying asset pool of the transaction. In a static deal, this set is determined at the outset. In a dynamic (managed) deal, this set changes over time depending on the originator’s policy. In a true sale transaction, all loans/bonds are sold without recourse to the special purpose vehicle which issues an equity tranche (=FLP) and various tranches of bonds to investors. The originator can freely use the proceeds from issuing the tranches including the sold part of the equity tranche. In a synthetic transaction the originator retains ownership of the loans/bonds and transfers part of the default risk through a junior credit default swap to the special purpose vehicle. This swap covers default risks beyond a threshold. This threshold implies a FLP of the originator. The coverage of default risks by the swap is limited by the face value of the bonds issued by the SPV. Often the issued bond-tranches cover only a small fraction of the nominal value of the underlying portfolio so that the originator retains a large super-senior tranche and its associated default risk unless this risk is protected through a super-senior

    1 The bonds may include a few tranches of other securitization transactions or structured finance

    products.

     5

    credit default swap. Hence, the super-senior tranche represents a TLP which is only hit if the SLP of investors is fully exhausted by default losses. In contrast to a true sale transaction, the originator does not receive the issuance proceeds in a synthetic transaction. These need to be invested in AAA-securities or other almost default-free assets. In all transactions, the originator decides about the choice of the asset pool, the size of the FLP, the tranching of the bonds to be issued. If the originator opts for a synthetic transaction, he also decides about the TLP. All these decisions are taken by the originator in close collaboration with the involved rating agencies and leading investors.

    In the following we summarize the literature related to these issues. There exists a variety of papers modelling the optimal design of financial contracts. Several papers show the optimality of first loss positions (FLP). In the absence of information asymmetries, Arrow

    (1971) [see also Gollier and Schlesinger (1996)] analyse the optimal insurance contract for a

    setting in which the protection buyer is risk averse, but the protection sellers are risk neutral. If the protection sellers bound their expected loss from above, then a FLP of the protection buyer is optimal. This follows because optimal risk sharing entails an upper limit of the realized loss borne by the risk averse protection buyer. Townsend (1979) considers risk

    sharing between a risk averse entrepreneur and investors in the presence of information asymmetries about the entrepreneur’s ability to pay. If the entrepreneur fully pays the investors’ claim, then she incurs no other costs. If she does not fully pay claiming that she lacks the necessary funds, then this claim needs to be verified. If the state verification cost is borne by the entrepreneur, the optimal contract is a standard debt contract: The entrepreneur fully pays the fixed claim when her company earns sufficient funds. Otherwise she prefers to pay the lower state verification cost and impose some loss on the investors. This is basically the same as taking a FLP.

    In a related model of Gale and Hellwig (1985), both, the entrepreneur and investors, are risk

    neutral. However, the entrepreneur can only bear limited losses in order to stay solvent. Again, a standard debt contract turns out to be optimal implying a FLP of the entrepreneur. In the previous two papers information asymmetries are resolved through state verification. The more recent literature distinguishes between information-sensitive and -insensitive securities. Information-insensitive securities are subject to little information asymmetries, in contrast to information-sensitive securities. Boot and Thakor (1993) argue that a risky cash

    flow should be split into a senior and a subordinated security. The senior security is information-insensitive and can be sold to uninformed investors while the subordinated security is information-sensitive and should be sold to informed investors. This allows the seller of the cash flow to raise the sales revenue. Riddiough (1997) extends this reasoning by

    showing that loan bundling allows for pool diversification which softens information

     6

    asymmetries. Moreover, the holder of the junior security should control changes in the loan

    2 portfolio because she primarily bears the consequences.

    DeMarzo and Duffie (1999) analyse the security-design assuming a tradeoff between the retention cost of holding cash flows and the liquidity cost of selling information-sensitive securities. They also prove that a standard debt contract is optimal and that an issuer with very profitable investment opportunities retains little default risk in a securitisation transaction. In a recent paper DeMarzo (2005) shows that pooling of assets has an information destruction effect since it prohibits the seller to sell asset cash flows separately and, thereby, optimize asset specific sales. But pooling also has a beneficial diversification effect. Tranching then allows to create more and also less information-sensitive claims and to sell the more liquid information-insensitive claims. This model is generalized to a dynamic model of intermediation. Summarizing these papers, they demonstrate the optimality of a FLP and argue that the senior information-insensitive tranches should be sold to investors. This is in strong contrast to synthetic transactions in which these tranches are not sold and give rise to a TLP of the originator.

     Plantin (2003) shows that sophisticated institutions with high distribution costs buy and sell the junior tranches leaving senior tranches to retail institutions with low distribution costs. David (1997) asks how many tranches should be issued. Tranches are sold to individual and institutional investors. The latter buy tranches to hedge their endowment risk. Hence tranches

    3should be differentiated so as to allow the different groups of investors an effective hedging.

    There are only a few empirical studies related to securitizations. Childs, Ott and Riddiough

    (1996) investigate the pricing of Commercial Mortage-Backed securities and find the correlation structure of the asset pool and the tranching to be important determinants of the launch spreads of the tranches. Higgins and Mason (2004) find that credit card banks provide

    implicit recourse to asset-backed securities to protect their reputation. Cebenoyan and Strahan

    (2004) find that banks securitizing loans hold less capital than other banks and have more risky assets relative to total assets. Downing and Wallace (2005) analyse securitizations of

    commercial mortgage backed securities and find that FLPs are higher than what might be expected looking at the actual performance of mortgages. Downing, Jaffee and Wallace (2006)

    find that participation certificates sold to special purpose vehicles are on average valued less than those not sold. Franke and Krahnen (2006) find that securitization tends to raise the

    2 Gorton and Pennacchi (1995) consider a bank which optimizes the fraction of a single loan to be

    sold and the optimal guarantee against default of the loan. This setup contrasts with a FLP of the

    bank.

    3 Glaeser and Kallal (1997) show that more information may increase information asymmetries.

    Hence limiting information disclosure may improve liquidity of asset-backed securities in the

    secondary market.

     7

bank’s stock market beta indicating more systematic risk taking. Cuchra and Jenkinson (2005)

    analyse the number of tranches in securitizations and find that the number increases with sophistication of investors, with information asymmetry and with the volume of the

    4. Finally, Cuchra (2005) analyses the launch spreads of tranches in securitizations transaction

    and finds that ratings are very important determinants besides of general capital market conditions. He also finds that larger tranches command a lower spread indicating a liquidity premium.

    3. The Originator’s Optimization and Hypotheses

    In this section, we present a simple model for the originator?s optimization problem and derive hypotheses about her optimal choice which then are tested on a set of European transactions. When structuring a securitization transaction, the originator maximizes her net benefit. Her gross benefit in a CLO-transaction may be summarized by the decline in the costs of required equity capital and other regulations and possibly the decline of funding costs, due to the decline in the default risks borne by the originator. The decline in default risks enables the originator to take other new risks. Then the value of these new activities contributes to the gross benefit. The costs of securitization transactions include the setup and management costs, the credit spreads paid to investors, the costs of credit enhancements and reputation costs. The latter costs are incurred if investors suffer from default losses and attribute them to bad management of the originator. Investors would then charge higher spreads in future transactions.

    In a CBO-transaction, the originator also maximizes her net benefit. However, often she purchases the asset pool and securitizes it simultaneously, retaining part of the risks through a FLP. Apart from these risks, the net benefit in such a transaction is an arbitrage profit. This explains why these transactions are often called arbitrage transactions.

    Transferring default risks through a securitization transaction is always subject to problems of information asymmetries between the originator and investors. The originator knows more about the quality of the loans underlying the asset pool in a transaction because she has close contact to the obligors. Moreover, she decides about her effort of monitoring the obligors and enforcing her loan claims. This effort is not observable by investors adding to the information asymmetry. Therefore credit spreads include a penalty for adverse selection and moral hazard problems. To model these information asymmetries, we distinguish between the published and the true asset pool quality. The rating agencies publish information on the asset pool quality. We assume that rating agencies do their best to publish unbiased information. Investors believe this so that they consider the published information as the best predictor of

    4 There are also various empirical studies about implied correlations of tranches in CDO?-transactions.

     8

    the true asset pool quality. But they know that the true quality differs from the published quality by a noise term ε. Hence

    true asset pool quality = published asset pool quality + ε.

    We define the standard deviation of the noise term, σ(ε), as quality uncertainty and assume

    that it is inversely related to the asset pool quality. Asset pool quality is measured in different dimensions. One measure of the average quality of the loans is the weighted average default probability (WADP) of the loans. WADP indicates the expected default losses of the asset pool. The second measure of asset pool quality is a measure of asset pool diversification. The intra- and interindustry-diversification of the loan portfolio can be summarized into a diversity score (DS) as it is done in Moody?s Diversity Score. This score can be interpreted as the diversification-equivalent number of equally sized loans whose defaults are uncorrelated. A third characteristic of the asset pool quality is the weighted average loss given default of the loans. Loss given default is measured by (1-recovery rate). The recovery rate of a loan is the fraction of its par value denoting the present value of all future payments on this defaulted loan discounted to the date of default. Initially the par value of the loan approximately equals its market value so that the loss given default applies equally to the par and the market value. Since we cannot get reliable data on the weighted average loss given default for most CDO-

    5transaction, we assume that this characteristic is the same across all CDO-transactions.

    Moreover, to simplify modelling we assume that the loss given default is non-random. Hence we characterize asset pool quality by the two determinants WADP and DS.

    Returning to information asymmetry and asset pool quality, we assume that quality uncertainty, σ(ε), increases with WADP. The intuition for this is that errors in estimating the true WADP are likely to be proportional to the estimated WADP. We also assume that σ(ε) is

    inversely related to DS. As pointed out by DeMarzo (2005) and others, a high DS reduces

    information asymmetries because the idiosyncratic risks of the assets tend to be diversified

    6away. The lower DS, the stronger is idiosyncratic default risk relative to systematic default risk. The effects of idiosyncratic risks are almost by definition harder to analyse and to predict than those of systematic risk, because idiosyncratic risks are much more diverse and less well understoood. Hence we believe that there are good reasons to assume an inverse relation between asset pool quality and quality uncertainty. Higher quality uncertainty creates more potential for adverse selection and moral hazard, because these activities are more difficult to

    5 Only for a few Spanish transactions we have some data which we then use in our empirical study. 6 De Marzo (2005) argues that stronger diversification makes securitization of asset pools more

    attractive relative to liquidating assets separately because diversification reduces information

    asymmetries.

     9

    discover when the quality risk is stronger. For example, moral hazard of the originator in monitoring loan performance which adds to σ(ε), is harder to discover for loans of low quality

    because these loans are often subject to more idiosyncratic risk factors than high quality loans. Therefore, higher uncertainty about the asset pool quality should reinforce problems of information asymmetries.

    In the following, the two asset pool characteristics WADP and DS are always understood as the published characteristics. Investors assume that these characteristics are unbiased estimates of the true characteristics, but are aware of the quality uncertainty-dependent potential for adverse selection and moral hazard. In the next subsection, we analyse, first, the originator?s choice of the underlying asset pool in terms of the two asset pool characteristics WADP and DS. Second, we present a simple model and analyse the originator?s effort on monitoring the obligors of the underlying assets. Third, we analyse the choice of the size of the FLP. This allows us to derive testable hypotheses.

3.1 Choice of the Asset Pool

    The originator selects the assets to be included in the underlying collateral pool. Since this paper only considers CDO-transactions, we restrict ourselves to corporate loans and bonds. In the case of a loan transaction, the originating bank transfers part of the default risk of a subportfolio of its loans. The choice of this subportfolio relates to the number and the quality distribution of the loans, their maturity structure and its diversification within and across industries. The first question is whether originators choose asset pools with homogeneous quality. Quality is said to be homogeneous if a low WADP is associated with a high DS and vice versa.

    If investors charge higher credits spreads for portfolios with stronger information asymmetries , then it pays for the originator to put together a well diversified asset portfolio. In a CLO-transaction this is easy for a bank with a large loan portfolio. Therefore we conjecture that the loan portfolio in a CLO-transaction will show a high DS. The situation is different for CBO-transactions. In a CBO-transaction the originator has to buy the bonds for the asset pool. This is often costly since the bond market is rather illiquid. Therefore we hypothesize that loan portfolios are better diversified.

    Hypothesis 1: The diversity score of the asset pool is higher in CLO- than in CBO-

    transactions.

    Whether information asymmetries are stronger for CLO- than for CBO-transactions is not clear. Loans are often given to small or medium sized firms whose identity is not revealed to

     10

Report this document

For any questions or suggestions please email
cust-service@docsford.com