Credit FAQ: Standard & Poor's Explains Updates To Global Corporate Cash Flow And Synthetic CDOs Criteria

Transcription

1 September 17, 2009 Credit FAQ: Standard & Poor's Explains Updates To Global Corporate Cash Flow And Synthetic CDOs Criteria Senior Director, U.S. Structured Credit New Issuance: Robert J Radziul, New York (1) ; robert_radziul@standardandpoors.com Global Criteria Officer - Structured Credit: Henry C Albulescu, New York (1) ; henry_albulescu@standardandpoors.com Chief Credit Officer - Structured Finance Ratings: Thomas G Gillis, New York (1) ; tom_gillis@standardandpoors.com Analytical Manager - U.S. Structured Credit New Issuance: Winston Chang, New York (1) ; winston_chang@standardandpoors.com Analytical Manager - U.S. Structured Credit Surveilance: Stephen Anderberg, New York (1) ; stephen_anderberg@standardandpoors.com Lead Analytical Manager - European Structured Credit: Lapo Guadagnuolo, London (44) ; lapo_guadagnuolo@standardandpoors.com Analytical Manager - European Structured Credit Surveillance: Gordon Wright, London (44) ; gordon_wright@standardandpoors.com Analytical Manager - Australia And New Zealand Structured Credit: Mei Lee Da Silva, Melbourne (61) ; mei_dasilva@standardandpoors.com Analytical Manager - Asia-Pacific Structured Credit: Toshihiro Matsuo, Tokyo (81) ; toshihiro_matsuo@standardandpoors.com Primary Credit Analyst - Asia-Pacific Structured Credit: Gloria Lu, CFA, Hong Kong (852) ; gloria_lu@standardandpoors.com Analytical Managers - Quantitative Analytics Research: Cristina Polizu, New York (1) ; cristina_polizu@standardandpoors.com Bob Watson, New York (1) ; bob_watson@standardandpoors.com Media Contact: 1 Standard & Poor's. All rights reserved. No reprint or dissemination without S&P's permission. See Terms of Use/Disclaimer on the last page

2 Adam M Tempkin, New York (1) ; adam_tempkin@standardandpoors.com Table Of Contents Modifications From The Request For Comment (RFC) Proposal RFC Feedback Applicability Of The Updated Criteria Supplemental Tests CDO Evaluator Default Rates Correlation Recovery Rates Other Topics Related Research Standard & Poor s RatingsDirect September 17, Standard & Poor's. All rights reserved. No reprint or dissemination without S&P's permission. See Terms of Use/Disclaimer on the last page

3 Credit FAQ: Standard & Poor's Explains Updates To Global Corporate Cash Flow And Synthetic CDOs Criteria Standard & Poor's Ratings Services has updated its methodologies and assumptions for rating cash flow and synthetic collateralized debt obligations (CDOs) backed by corporate debt (loans and bonds). Here we provide answers to some frequently asked questions to assist market participants to better understand the changes we have made. This FAQ report doesn't replace the updated criteria (hereafter "Criteria Article"), which we encourage readers to review in its entirety (see "Update To Global Methodologies And Assumptions For Corporate Cash Flow And Synthetic CDOs," published Sept. 17, 2009). Modifications From The Request For Comment (RFC) Proposal 1. Based on the corporate CDO RFC, what modifications did you make to the proposed "outside-the model" tests? To provide more clarity on these "supplemental tests," we have programmed them into CDO Evaluator. The tests are separate and distinct from the default simulation that CDO Evaluator performs. Based on the market feedback we received, we have made the following changes: "Largest obligor default test." We factor the credit quality of the underlying obligors into the "largest obligor default test" by tiering the number of obligor defaults in the analysis based on the ratings on the CDO tranches and obligors. Also, we assume a 5% recovery rate for all obligors that are treated as defaulted under this test (see Criteria Article paragraphs 21 and and Appendix A). "Largest industry default test." We modified the "primary largest industry default test" to include a recovery assumption of 17% rather than 0% as initially proposed. We also introduced an "alternative largest industry default test," which is an adaptation of the largest obligor default test. A CDO would satisfy the industry default test if it passes either of these two tests (see Criteria Article paragraphs 22 and 38-46). 2. Did you make any changes to the calibration of CDO Evaluator proposed in the RFC and to any of CDO Evaluator's input parameters? Yes, we made the following changes: CDO Evaluator recalibration. We adopted the recalibration of CDO Evaluator proposed in the RFC, with some minor modifications to the 'AAA' target assumptions for 'BBB' rated assets. In the final criteria we have also provided more information on the historical data we reviewed when setting the "targeted portfolio default rates," for CDOs (see Criteria Article paragraphs and 89-94). CreditWatch notching. To account for potential downgrade risk inherent in ratings on CreditWatch negative, we continue to treat such ratings as if they were one notch lower, rather than two notches lower as proposed in the RFC. We continue to assume that ratings on CreditWatch positive are one notch higher (see Criteria Article paragraphs 77-79). 3

4 Correlation. We increased our assumptions regarding correlation between corporate obligors and corporate CDO tranches and between corporate CDO tranches and other corporate CDO tranches held as assets in the CDO transaction (see Criteria Article paragraphs 80-82, , and ). 3. Did you make any changes to the asset recovery rates proposed in the RFC? We did not incorporate asset recoveries directly into the default simulation as proposed in the RFC. We did, however, tier recoveries for assets, both in cash flow and synthetic CDOs. As part of the final criteria, we reduced the assets recovery rate assumptions further than proposed in the RFC. They are in line with our rating definitions (see "Understanding Standard & Poor's Rating Definitions," published June 3, 2009) and revised expectations for a stressed credit cycle, where higher defaults and a lack of liquidity would likely increase the number of businesses that liquidate rather than restructure. This puts a stress on recoveries, especially for those assets lower in the capital structure (see Criteria Article paragraphs and Appendix F). 4. What other notable changes did you adopt after the RFC? We adopted the following changes: Break-even default rates. We modified our break-even cash flow results analysis in an effort to capture more of the tail of the results, by applying a lower percentile to the break-even default rate results (ranked from high to low) that are excluded (see Criteria Article paragraphs ). Issuance cap. We capped the rated CDO note issuance amount to the economic value retained in the transaction (see Criteria Article paragraphs ). RFC Feedback 5. What feedback did you receive from the RFC on the outside-the model/supplemental tests? We received the following feedback on the two supplemental tests: "Largest obligor default test." Some market participants questioned the largest obligor default test proposed in the RFC because it did not take into account the credit quality of specific assets and that assuming zero recoveries was too onerous. Based partly on these comments, we have included in the test the credit quality of the underlying assets and have implemented a 5% recovery rate for obligors treated as defaulted under this test (see Criteria Article paragraphs 21 and and Appendix A). "Largest industry default test." Most of the comments about the largest industry default test argued that the test may be inappropriate because an entire industry would never default and recoveries would be higher than zero. Although defaults of all companies in a given industry would be extremely unlikely, a CDO typically holds a subset of companies from any given industry that may face higher stresses than other companies in the same industry. Additionally, participants questioned the assumption of zero recovery with the argument that recoveries across a sector should be higher than any individual company. We now assume a 17% recovery assumption (see Criteria Article paragraphs 22 and 38-46). 6. What feedback did you receive on the proposed recalibration of CDO Evaluator? The recalibration of CDO Evaluator generated the most comments. Some respondents supported the proposal while many expressed the view that the results might be too onerous. Some of the comments expressed support for the criteria proposal for new transactions but urged against applying it retroactively to outstanding transactions. No direct empirical default data or alternative stress scenarios were suggested for consideration. Some agreed that using historical data and projecting a stress based on a view of the future economic environment should be sufficient. The Standard & Poor s RatingsDirect September 17,

5 updated criteria adopt the proposed recalibration of CDO Evaluator with some minor modifications to the 'AAA' target assumptions for 'BBB' rated assets (see Criteria Article paragraphs and 89-94). Regarding CreditWatch negative notching, we received mixed comments on the proposal to treat obligor ratings on CreditWatch negative as if they were two notches lower, with most suggesting that the proposed approach would be too onerous and expressing concerns regarding the asymmetrical notching of CreditWatch negative and CreditWatch positive. On further review of the matter, we decided to keep the CreditWatch negative adjustment for corporate assets to one notch (e.g., A/Watch Neg to 'A-'), in line with the adjustment we make for ratings on CreditWatch positive (see Criteria Article paragraphs 77-79). 7. What was the market feedback on the proposed reduction in asset recoveries? Feedback on the proposal to reduce recoveries was mixed with some strong opinions on both sides. On further review, we reduced the expected recovery levels for defaulted corporate debt further than was indicated in the RFC. This is based on empirical data from the current credit cycle and our view of the likely recovery outcomes under the different rating stresses (see Criteria Article paragraphs and Appendix F). 8. What was some of the other market feedback that you received? We received additional feedback on the following proposals in the RFC: Default timing patterns. Market feedback on the proposal to push back the starting time of defaults for CDO tranches rated in the 'A' to 'B' range was positive. We have adopted this proposal (see Criteria Article paragraphs ). Minimum upfront management fee. Market participants provided mixed feedback on the minimum upfront management fee of 50 bps for collateral managers with less than $2.5 billion total assets under management. Some participants supported the proposal and others objected that the level of fees should not concern the rating agencies. The updated criteria include the minimum 50 bps fee because, in our opinion, the collateral manager can have an important role in the current and future performance of the transaction. Modeling the minimum fee will enhance tranche rating stability if increased management fees were needed to secure a replacement manager (see Criteria Article paragraphs ). Stressed portfolio disclosure. On our proposal to provide additional information based on stressed portfolio composition, the comments were positive. We will provide this information in the future. Applicability Of The Updated Criteria 9. Which transactions are subject to the updated methodologies and assumptions? We apply the updated methodologies and assumptions to well-diversified cash flow CDOs and synthetic CDOs primarily backed by corporate credit risk. They also apply to transactions backed by corporate obligors consisting of a mix of cash and synthetic instruments, CDOs of corporate CDOs, and CDOs of hybrid trust preferred securities. The criteria changes augment (and supersede certain elements of) our previous criteria for these transactions. To the extent that individual transactions exhibit unique characteristics in their portfolios or structures (such as small-basket trades, "Nth-to-default", and leveraged super senior with spread transactions), we may apply additional or customized analytical scenarios (see Criteria Article paragraphs 3-6, 141, and ). 5

6 10. Will both existing and new CDO transactions be subject to the updated criteria? Yes, we apply the updated methodologies and assumptions to both new and existing CDOs primarily backed by corporate credit risk (see Criteria Article paragraphs 5, 19, and 150). Supplemental Tests 11. What are the new "supplemental tests"? The "supplemental tests" are additional quantitative and qualitative elements in our analysis and are separate and distinct from the Monte Carlo default simulations we run in Standard & Poor's CDO Evaluator model. We believe that adding these tests to our use of the simulation model enhances our overall analysis. They are intended to address both event risk and model risk that may be present in rated transactions. The two supplemental tests are the "largest obligor default test" and the "largest industry default test" (see Criteria Article paragraphs 20-46). 12. What is the purpose of the largest obligor default test? This test aims to assess potential losses that might not be fully captured by the CDO Evaluator default simulation. Specifically, it considers whether a CDO tranche has sufficient credit enhancement (not counting excess spread), in our opinion, to withstand specified combinations of underlying obligor defaults based on the ratings on the underlying obligors. The ratings on the obligors and on the specific CDO tranche drive this test, thus taking into consideration the credit quality of specific obligors in the pool. Since recent experience in the financial sector demonstrates that there is the potential for very small recoveries under stressful conditions, we assume a flat 5% recovery for obligors that are treated as defaulted under this test (see Criteria Article paragraphs 20-21, 23-27, and 34-37). 13. How does a CDO tranche satisfy the largest obligor default test? We compare an adjusted par value of the assets representative of the obligors in the pool against the CDO tranche balances rated at and above the given rating at which we apply this test. (The adjusted par value represents the par balance of the performing asset pool, plus principal proceeds, minus the highest of the losses from the largest obligor default test, plus expected recoveries on defaulted obligors held in the asset pool.) We determine the test to have passed if the adjusted par value of the assets is greater than the principal balance of the liabilities rated at or above the given rating level. For synthetic CDOs we consider whether the attachment point is set sufficiently high to cover the highest losses from the obligor test without breaching the rated tranche (see Criteria Article paragraphs and Appendix A, and "Application Of Supplemental Tests For Rating Global Corporate Cash Flow And Synthetic CDOs", published Sept. 17, 2009). 14. How do you value the defaulted obligors in the largest obligor default test? For this test, we assume defaulted obligors are those the CDO manager classifies as defaulted along with any obligors rated below 'CCC-'. We assume the value of the defaulted obligors to be the lower of the market or recovery value for cash flow CDOs. For synthetic CDOs, unless the transaction documents require fixed recoveries, we assume the recovery values outlined in our updated criteria until the ISDA protocol or another valuation method determines the recoveries (see Criteria Article paragraphs and 33). 15. What is the purpose of the largest industry default test? This test consists of two parts: The "primary largest industry default test" ("primary industry test") and the "alternative largest industry default test" ("alternative industry test"). The primary industry test assesses whether a CDO tranche rated between 'AAA' and 'AA-' has sufficient credit enhancement (not counting excess spread), in our Standard & Poor s RatingsDirect September 17,

7 opinion, to withstand a default of the largest industry held by the CDO transaction, assuming a flat 17% recovery. We determine the largest industry based on the largest aggregate notional amount of obligors in the portfolio based on CDO Evaluator industry codes. If the primary industry test fails, we run the alternative industry test (see Criteria Article paragraphs 20, 22-24, and 38-46). 16. Is a CDO tranche rating capped at 'A+' if the tranche fails to satisfy the primary industry test? No, under our criteria a tranche can still be rated between 'AAA' and 'AA-' even if it fails the primary industry test if it satisfies the alternative industry test. The tranche would satisfy this alternative test if it has sufficient credit enhancement (not counting excess spread) to survive the highest level of losses associated with the default of the largest of each of the combinations of underlying obligors within that largest industry, assuming 5% recoveries. This alternative industry test is intended to capture gradations of obligor credit quality while applying somewhat higher default intensity than the largest obligor default test (see Criteria Article paragraphs 43-45). 17. Why does the largest obligor default test assume only a 5% recovery when the primary industry test assumes a 17% recovery? The primary industry test applies a higher recovery assumption than the largest obligor default test because recoveries across a whole industry generally imply an averaging effect. The industry-wide recoveries are likely to be higher than the lowest recovery within the group. The alternative industry test assumes a 5% recovery since we apply this second test only to a sub-sector of the entire industry, similar to the largest obligor default test (see Criteria Article paragraphs 40, 42, and 45). 18. What recovery rates would you assume in the supplemental tests for synthetic CDOs that have fixed recoveries in their transaction documents? We would assume the transaction document's fixed recovery values in the supplemental tests regardless of the market value (see Criteria Article paragraph 31). 19. What would happen if a CDO tranche doesn't satisfy the new supplemental tests at the proposed rating level? Under our updated approach, in most instances if a CDO tranche doesn't pass all of the tests associated with a particular rating level (including the supplemental tests), we would, all other things being equal, likely assign the rating at which that CDO tranche passes all of the tests (see Criteria Article paragraphs 23-24). 20. Are there any instances when you would waive or adjust your supplemental tests? In exceptional circumstances, a rating committee might consider waiving or adjusting the supplemental tests after taking into account the specifics of the transaction's asset portfolio, the tranche balance, and the transaction's structural features. Possible exceptions include: The largest obligor default test doesn't apply to a tranche if all of the following conditions are satisfied: (i) The subject tranche is the most senior outstanding tranche in the transaction, (ii) the tranche immediately below it in the capital structure passes the largest obligor default test, (iii) the tranche below it is rated no more than one full rating category below the subject tranche, and (iv) we expect the subject tranche to be retired within six months based on scheduled distributions (see Criteria Article paragraph 37). There is an alternative industry test that can be satisfied if the primary industry test fails (see Criteria Article paragraphs 22 and 43-45). The largest industry default test doesn't apply to the CDOs of hybrid trust preferred securities. Even though these securities use corporate asset default rates for the underlying assets, they may be viewed as a single industry. In 7

8 such transactions, we address the industry risk as explained in "Global Methodology For Rating Trust Preferred/Hybrid Securities Revised," Nov. 21, 2008 (see also Criteria Article paragraph 46). CDO Evaluator Default Rates 21. What changes have you made to the CDO Evaluator framework? We recalibrated CDO Evaluator to generate corporate obligor default rates under different economic stress scenarios that we believe reflect extreme economic stress events. The obligor defaults are the "targeted portfolio default rates" that we believe CDO tranches should pass to be assigned a particular rating. In our view, any 'AAA' rated corporate CDO tranche should in ordinary circumstances be able to withstand such a targeted extreme degree of stress without defaulting. By setting targeted rates, we focus our analysis on the CDO Evaluator model's outputs and less on the model's inputs, which are typically based on historical default data (see Criteria Article paragraphs 47-94). 22. What is the relationship between the targeted portfolio default rates and model inputs? We first established a targeted assets portfolio default rate for each rating level. These rates represent the minimum portfolio default threshold that each CDO tranche should withstand at each rating category, in our opinion. Then, we adjusted the CDO Evaluator inputs (asset default rates, correlation among assets, etc.) so that the CDO Evaluator portfolio default outputs for different asset portfolios are consistent with the targeted portfolio default rates. We did this even at the expense of making some of the model inputs seem less "realistic" relative to the historical data (see Criteria Article paragraphs and 72-76). 23. Why did you focus on targeted portfolio default rates rather than relying on historical data to support model inputs? We believe these updated criteria provide a more intuitive understanding of what our rating is intended to reflect: Each CDO tranche at a given rating is generally intended under ordinary circumstances to "withstand" certain stressed default scenarios without defaulting. Also, we do not ascribe "default probabilities" to each rating category. Rather, our credit ratings express a relative ranking of creditworthiness and may encompass not only relative likelihood of default but also payment priorities, recoveries, and credit stability (see Criteria Article paragraph 50). 24. How did you establish the default targets in CDO Evaluator? A committee of senior analysts developed these targeted portfolio default rates using historical corporate default data and a review of past economic cycles, as well as their judgment, to determine the expected default rates that could materialize under what we view as a high stress level associated with a particular rating scenario (see Criteria Article paragraphs 47-71). 25. Why did you add additional cash flow stresses to lower-rated tranches? "Back-loaded" default stresses which add stresses later on in the transaction's tenor should address scenarios where the default rate increases significantly after a period of low defaults. The current environment may be an example of such a default environment because corporate default levels have been low for several years, but we now project default levels to increase significantly in 2009 and Lower rated tranches issued in and which are amortizing now face such back-loaded stress. We now stress all tranches rated 'AAA' through 'B-' by applying the same default patterns and starting times of those patterns that we previously ran for 'AAA' through 'AA-' tranches (see Criteria Article paragraphs ). Standard & Poor s RatingsDirect September 17,

9 Correlation 26. What changes have you made to the correlation assumptions in CDO Evaluator? We now use correlation as an adjustment parameter to arrive at our targeted portfolio default rates. Overall, we have increased correlation. We have changed the inter-sector correlation to from and changed the intra-sector correlation to 0.20 from Also, we apply a correlation of 0.05 between obligors from different industries in different geographic regions, of 0.10 between corporate obligors and corporate CDO tranches, and of 0.70 between corporate CDO tranches. The updated criteria retain the existing higher correlation levels for certain industries (see Criteria Article Appendix D and paragraphs 80-82, , and ). 27. How important are the updated correlation assumptions to CDO ratings? We used the correlation changes to recalibrate our default simulation model to the targeted portfolio default rates. This means that the "quality" of the input parameters becomes a secondary consideration. By targeting results and adding additional stress tests, we believe we are taking important steps toward buttressing our analysis against the inherent limitations of simulation models and dependence on the quality of the model inputs. We haven't made any changes to the CDO Evaluator industry classifications (sector codes) or other correlation factors (see Criteria Article paragraphs 47-48, 80-82, , and ). Recovery Rates 28. What changes have you made to your recovery assumptions? We link recovery rates with default rates to reflect their observed inverse relationship (when one is high, the other tends to be low). Under this approach, which applies to both cash flow and synthetic CDOs, each asset in a transaction has a different recovery rate based on the different corporate asset types (e.g., loans/bonds, seniority, and security), if a recovery rating exists, and the country grouping to which the asset belongs. The tiered recovery values for the assets are based on the rating on the CDO tranche. For example, if the same asset pool supports both 'AAA' and 'BBB' tranches, our recovery assumptions are lower when rating the 'AAA' tranche relative to when rating the 'BBB' tranche. For synthetic CDOs, we use the "senior unsecured bonds" asset type as our base case recovery assumption and we apply one additional deduction or "haircut" for "old restructuring." These recovery assumptions replace the recoveries given in the article "Updated Global Recovery Rates For Use In Cash Flow CDOs," published July 23, 2007 (see Criteria Article paragraphs ). 29. How did you determine the country recovery groupings? We grouped different countries based on our analyses of their insolvency legal frameworks. We believe this framework is a good indication of the different rights creditors have to secure their claims and realize a recovery. Countries in Group 1 have legal frameworks that give the senior lenders more control and likely higher senior recoveries. Countries in Group 3 have a legal framework that gives a lower priority to senior claims relative to the other creditors (see "Update: Jurisdiction-Specific Adjustments To Recovery And Issue Ratings", published June 20, 2008, for a complete discussion on different insolvency regimes and see Criteria Article paragraphs ). 30. On which tranche rating do you base your recovery rate assumption for existing CDO transactions? We tier asset recovery rate assumptions by CDO tranche rating level, and assume a given asset has a different recovery rate under a 'AAA' scenario, 'AA' scenario, etc. In our analysis, we look to the issuer's target rating on the 9

10 CDO tranche to determine which recovery rate to assume, not the original or current rating. For example, in our analysis of a CDO tranche rated 'BBB' that we anticipated would be lowered into the 'BB' range, we would use the 'BB' recovery assumption. We believe this maintains consistency across transactions in our surveillance analysis, with 'BB' rated tranches from different transactions using the same recovery rate. However, we also believe that when a CDO trustee assigns recovery rates to defaulted obligors in a transaction's overcollateralization ratio test, the recovery rate should be based on the original tranche rating. Otherwise, if a tranche is downgraded the overcollateralization ratio test could be satisfied sooner due to the newly increased recovery rates. As a consequence, less excess cash would be available to the senior noteholders. Other Topics 31. What is credit stability and how do you use it in your CDO analysis? We have incorporated credit stability as an explicit factor in our rating framework. When assigning and monitoring ratings, we consider whether we believe an issuer or security has a high likelihood of experiencing unusually large adverse changes in credit quality under what we consider to be conditions of moderate stress (for example, recessions of moderate severity, such as the U.S. recessions of 1960 and the early 1990s and the European recession of the early 1990s, or appropriate sector-specific stress scenarios). If such is the case, we may assign the CDO tranche a lower rating than we would have otherwise due to lower credit stability. We have defined the maximum projected deterioration under moderate stress conditions that we would associate with each rating level for time horizons of one year and three years. (For more information on credit stability see "Understanding Standard & Poor's Rating Definitions," published June. 3, 2009," and Standard & Poor's To Explicitly Recognize Credit Stability As A Rating Factor," published Oct. 15, 2008.) We tested various asset portfolios by changing their ratings using a rating transition matrix assuming what we consider to be a moderate stress scenario, ran the transition pools through CDO Evaluator, and reviewed the resulting changes in tranche ratings. In our view, this testing of the new analytical framework resulted in tranche ratings that performed within our credit stability guidelines (see Criteria Article paragraphs ). 32. What do you mean by "rating sensitivity to modeling parameters"? The updated criteria article includes additional measures beyond the analysis of credit stability to determine a transaction's sensitivity to various portfolio and structural risks. Rating committees consider each of these measures in addition to other factors that may be relevant for a specific CDO transaction. If a committee finds an increased level of sensitivity in any of these factors for a given transaction, it may give additional weighting to the relevant factor to address this sensitivity. The proposed sensitivity measures include: Considering different levels of correlation factors, in addition to standard levels; Applying a 90%-110% range of the portfolio's weighted-average recovery levels; Biasing defaults toward the largest obligors, the obligors with the widest spreads, and the obligors with the lowest recoveries; and Considering break-even default rates (BDRs) that are lower than the minimum BDRs currently selected (using percentiles) and analyzing "failed" paths for further analysis. (See Criteria Article paragraphs and ) Standard & Poor s RatingsDirect September 17,

11 33. What is the "stressed portfolio" approach? Under the stressed portfolio approach, the criteria deem the portfolio comprises the minimum number of obligors, concentrated in the minimum number of industries permitted in the transaction documents. We also assume under this approach that the largest obligors are all in the same industry and have the lowest ratings allowed by the transaction's eligibility criteria. Further, we assign a rating assuming the portfolio has the minimum weighted-average spread and coupon allowed, the longest weighted-average life, and the lowest projected recoveries under the eligibility and reinvestment criteria (see Criteria Article paragraphs ). 34. Are the "effective date" RAC and CDO Monitor or SROC no longer necessary under the updated criteria? The application of these factors may vary on a case-by-case basis. For our rating analysis, we continue to look to the effective date rating agency confirmation (RAC) and Standard & Poor's CDO Monitor or synthetic rated overcollateralization (SROC) measure to assess the CDO transaction whenever the CDO manager elects to maintain the portfolio's original credit quality under a "stable quality" approach. Alternatively, if the CDO manager elects to manage the transaction within the eligibility criteria given in the governing documents under the stressed portfolio approach, we expect the manager to confirm in writing on the effective date that the trades and portfolio ramp-up meet the asset eligibility, quality, and reinvestment guidelines specified in the applicable transaction documents. We would not need an effective date RAC and a CDO Monitor or SROC for our rating analysis under a stressed portfolio approach (see Criteria Article paragraphs ). 35. Will you issue new CDO Monitor models to reflect the updated criteria? It is not our intention to issue new models. Trustees and collateral managers for the affected transactions should continue to use their current version of CDO Monitor during and after the review period. This is consistent with our current practice of using the current CDO Monitor even after a tranche has been downgraded. CDO Monitor is a model that measures the effect on portfolio credit quality of changes in a given CDO transaction's collateral ratings, maturity profile, and diversification. Trustees and collateral managers use the model to determine whether a given transaction is in compliance with the transaction's CDO Monitor test. For actively managed cash flow and hybrid CDOs, the transaction documents typically detail the use of CDO Monitor on measurement dates during the reinvestment period. If a transaction is failing the test, the documents typically allow subsequent trades only to the extent they maintain or improve the test results. Each transaction would keep its existing CDO Monitor in place to maintain the link between a transaction's test results (that is, whether the transaction is currently passing or failing the CDO Monitor test) and the collateral performance of the transaction since origination. 36. What is the concept behind limiting the rated issuance to the economic value for new transactions? We believe that the recent market dislocation and liquidity squeeze have made it more difficult to differentiate between cheap but fundamentally attractive assets and assets that are stressed. We have some concerns about assuming "par credit" for deeply discounted assets, re-characterizing principal proceeds as interest proceeds, distributing principal proceeds to equity investors, and structures involving significantly more debt than the amount of money used to purchase assets. Going forward, rating committees may consider the sources and uses of funds to better understand the economic benefit to all investors where we have concerns about potential ratings arbitrage or managers that may be attempting to exploit "loopholes" (see Criteria Article paragraphs ). 11

12 37. Are all CDO transactions expected to pay 50 basis points of primary management fees? No, we do not determine how much a collateral manager gets paid. We do, however, assume in our rating analysis a primary management fee of the greater of 50 basis points and the actual primary fee per the CDO transaction documents if the collateral manager has less than $2.5 billion total assets (CDOs plus other assets) under management. We believe this will allow for more rating stability in those situations where investors decide to permit the existing or replacement manager to renegotiate the primary management fee, possibly without negatively affecting ratings on the outstanding tranches (see Criteria Article paragraphs ). 38. What are the potential effects of the criteria changes on outstanding ratings? We anticipate that these criteria changes will likely have a significant negative effect on the current outstanding ratings on many corporate CDO transactions (see "S&P Updates Global Corporate CDO Criteria," published today, for additional information). With the release of these updated criteria, we will be placing on CreditWatch negative almost all corporate CDO tranches (see Criteria Article paragraphs 16-18). 39. When will the CDO Evaluator model be available? We have released CDO Evaluator v5.0 along with the updated criteria. Investors and market participants are invited to download CDO Evaluator from CDOInterface.com. Related Research Update To Global Methodologies And Assumptions For Corporate Cash Flow And Synthetic CDOs, Sept. 17, 2009 Application Of Supplemental Tests For Rating Global Corporate Cash Flow And Synthetic CDOs, Sept. 17, 2009 S&P Updates Global Corporate CDO Criteria, Sept. 17, 2009 Understanding Standard & Poor's Rating Definitions, June. 3, 2009 Request For Comment: Update To Global Methodologies And Assumptions For Corporate Cash Flow CDO And Synthetic CDO Ratings, March 18, 2009 Global Methodology For Rating Trust Preferred/Hybrid Securities Revised, Nov. 21, 2008 Standard & Poor's To Explicitly Recognize Credit Stability As A Rating Factor, Oct. 15, 2008 Update: Jurisdiction-Specific Adjustments To Recovery And Issue Ratings, June 20, 2008 Updated Global Recovery Rates For Use In Cash Flow CDOs, July 23, Related articles are available on RatingsDirect. Criteria, presales, servicer evaluations, and ratings information can also be found on Standard & Poor's Web site at Standard & Poor s RatingsDirect September 17,

13 Copyright 2009 by Standard & Poors Financial Services LLC (S&P), a subsidiary of The McGraw-Hill Companies, Inc. All rights reserved. No part of this information may be reproduced or distributed in any form or by any means, or stored in a database or retrieval system, without the prior written permission of S&P. S&P, its affiliates, and/or their third-party providers have exclusive proprietary rights in the information, including ratings, credit-related analyses and data, provided herein. This information shall not be used for any unlawful or unauthorized purposes. Neither S&P, nor its affiliates, nor their third-party providers guarantee the accuracy, completeness, timeliness or availability of any information. S&P, its affiliates or their third-party providers and their directors, officers, shareholders, employees or agents are not responsible for any errors or omissions, regardless of the cause, or for the results obtained from the use of such information. S&P, ITS AFFILIATES AND THEIR THIRD-PARTY PROVIDERS DISCLAIM ANY AND ALL EXPRESS OR IMPLIED WARRANTIES, INCLUDING, BUT NOT LIMITED TO, ANY WARRANTIES OF MERCHANTABILITY OR FITNESS FOR A PARTICULAR PURPOSE OR USE. In no event shall S&P, its affiliates or their third-party providers and their directors, officers, shareholders, employees or agents be liable to any party for any direct, indirect, incidental, exemplary, compensatory, punitive, special or consequential damages, costs, expenses, legal fees, or losses (including, without limitation, lost income or lost profits and opportunity costs) in connection with any use of the information contained herein even if advised of the possibility of such damages. The ratings and credit-related analyses of S&P and its affiliates and the observations contained herein are statements of opinion as of the date they are expressed and not statements of fact or recommendations to purchase, hold, or sell any securities or make any investment decisions. S&P assumes no obligation to update any information following publication. Users of the information contained herein should not rely on any of it in making any investment decision. S&P's opinions and analyses do not address the suitability of any security. S&P does not act as a fiduciary or an investment advisor. While S&P has obtained information from sources it believes to be reliable, S&P does not perform an audit and undertakes no duty of due diligence or independent verification of any information it receives. S&P keeps certain activities of its business units separate from each other in order to preserve the independence and objectivity of each of these activities. As a result, certain business units of S&P may have information that is not available to other S&P business units. S&P has established policies and procedures to maintain the confidentiality of certain non-public information received in connection with each analytical process. S&P's Ratings Services business may receive compensation for its ratings and credit-related analyses, normally from issuers or underwriters of securities or from obligors. S&P reserves the right to disseminate its opinions and analyses. S&P's public ratings and analyses are made available on its Web sites, (free of charge) and www. ratingsdirect.com (subscription), and may be distributed through other means, including via S&P publications and third-party redistributors. Additional information about our ratings fees is available at Any Passwords/user IDs issued by S&P to users are single user-dedicated and may ONLY be used by the individual to whom they have been assigned. No sharing of passwords/user IDs and no simultaneous access via the same password/user ID is permitted. To reprint, translate, or use the data or information other than as provided herein, contact Client Services, 55 Water Street, New York, NY 10041; (1) or by to: research_request@standardandpoors.com. Copyright by Standard & Poors Financial Services LLC, a subsidiary of The McGraw-Hill Companies, Inc. All Rights Reserved