OVERVIEW AND SCOPE

1. This article is a Guidance Document. It is related to our criteria article "Global Methodology And Assumptions For CLOs And Corporate CDOs" (corporate CDO criteria), published June 21, 2019. It is intended to be read and applied in conjunction with these criteria and with other criteria referenced in this article. For further explanation about guidance documents, please see the description at the end of this document.

Part I of this Guidance Document provides additional information and guidance related to our credit and cash flow stresses. In addition to the corporate CDO criteria, the guidance in this part is also related to our criteria article, "Global Framework For Payment Structure And Cash Flow Analysis Of Structured Finance Securities," Dec. 22, 2020

Part II of this Guidance Document is intended to provide transparency in relation to our analysis of various provisions in corporate CDO transaction documents. In addition to the corporate CDO criteria, the guidance in this part is also related to the following criteria articles:

• "Global Framework For Payment Structure And Cash Flow Analysis Of Structured Finance Securities," Dec. 22, 2020
• "Counterparty Risk Framework: Methodology And Assumptions" (counterparty criteria), March 8, 2019;
• "Structured Finance: Asset Isolation And Special-Purpose Entity Methodology," March 29, 2017.

GUIDANCE

Part I: Credit And Cash Flow Assumptions

2. When we analyze the credit and cash flow risks in a collateralized debt obligation (CDO) transaction, we take into consideration the provisions in the transaction's documents governing how the deal's risk profile is likely to evolve over time through reinvestment and what limitations or protective mechanisms there are. This may influence what assumptions we make.

3. Accordingly, in applying the criteria, we may, in analyzing the credit risk and cash flow profile of a transaction, apply a "stable quality" approach or a "stressed portfolio" approach, depending on our assessment of these provisions (see the section titled Additional Rating Considerations, Stable Quality Versus Stressed Portfolio Approach).

4. We apply the "stable quality" approach to cash flow and hybrid CDOs where the manager commits to using S&P Global Ratings' CDO Monitor model as part of the reinvestment conditions to monitor the quality of the portfolio (the CDO Monitor test). For synthetic CDOs, we apply the "stable quality" approach where the synthetic rated overcollateralization (SROC) test is used (see "CDO Spotlight: What Is A Synthetic CDO?" published April 30, 2010).

5. Where we rate according to a "stressed portfolio" approach, we determine rating inputs and maturities based on the transaction's covenants rather than looking solely at the portfolio's actual characteristics. For such a transaction, we assume that the portfolio comprises the minimum number of obligors concentrated in the minimum number of industries permitted in the documents. In addition, we assume that the largest obligors are all in the same industry and have the lowest ratings allowed under the transaction documents' eligibility criteria. Finally, we assume that the portfolio has the minimum weighted-average spread and coupon allowed, and that it has the longest weighted-average life and lowest projected recoveries allowed under the transaction documents' eligibility criteria and reinvestment guidelines.

CREDIT RISK ANALYSIS

A. Determining inputs for use in the CDO Evaluator (see paragraph 10 in the corporate CDO criteria)

Determining the rating input

6. According to our criteria, the corporate issuers' creditworthiness is a parameter of our credit analysis. When we use the CDO Evaluator to conduct our credit analysis, we therefore use a rating input for each issuer, which is based on information available to S&P Global Ratings. The following summarizes different sources of information and the order of priority in which we use them to determine the rating input for each asset, in application of the provisions of our methodology for determining ratings-based inputs (see Related criteria section). For this purpose, in the context of CLOs, "substantial rating drivers" are typically the largest exposures in transactions where the supplemental tests (and hence a limited number of exposures) effectively drive our rating analysis. In a given portfolio, the rating input of each asset may be determined using a different method as per below, depending on the type of information available to us:

• (i) If there is an S&P Global Ratings long-term credit rating on the issuer--or on an obligor in the same organizational hierarchy, as appropriate--then that rating is the S&P Global Ratings rating input.
• (ii) If a credit estimate from S&P Global Ratings is available, then the credit estimate is the S&P Global Ratings rating input.
• (iii) If no issuer credit rating or credit estimate is available, but any of the issuer's obligations are rated by S&P Global Ratings, then the S&P Global Ratings rating input is determined by notching up or down from the issue rating as follows: (A) If the rated issue is senior unsecured, the rating input is the S&P Global Ratings issue rating on the unsecured obligation; (B) If the rated issue is senior secured, the rating input is one notch below the S&P Global Ratings issue rating on the senior secured obligation; and (C) If the rated issue is subordinated, the rating input is one notch above the S&P Global Ratings issue rating on the subordinated obligation.
• (iv) Except for substantial rating drivers, if a mapping has been provided by S&P Global Ratings for the collateral, the corresponding S&P Global Ratings rating input is determined pursuant to such mapping (see "Mapping A Third Party's Internal Credit Scoring System To Standard & Poor's Global Rating Scale ," published May 8, 2014).
• (v) Except for substantial rating drivers, if there is a public, unqualified, and global scale rating on the issuer from another credit rating agency (CRA) that we may use in our analysis of primary rating drivers according to our methodology for determining ratings-based inputs, then the corresponding S&P Global Ratings rating input is that CRA rating's equivalent, as informed by relevant regulatory mappings. If there is more than one such CRA rating available, we use the lowest equivalent rating input. If the CRA rating is an issue rating, we may infer a rating input considering public information from the relevant CRA regarding their methodology for determining issue ratings. The principal balance of the collateral relying on rating inputs from other CRAs may not exceed 15%.
• (vi) If there is a rating from another CRA, as described under (v) above, but using this input would result in the portion of the principal balance of collateral with rating inputs derived in this way exceeding 15%, or if there is a rating from another CRA that does not meet the conditions described in our methodology for determining ratings-based inputs for use in primary rating drivers, we may rely on sufficient information being provided by the collateral manager within a maximum of 90 days of the asset's inclusion in the portfolio to produce a credit estimate. Pending the determination of such credit estimate, the rating input would be determined by the collateral manager in its reasonable credit judgment and would thereafter be our credit estimate, once available.
• (vii) If (1) neither the issuer nor any of its affiliates is subject to insolvency, bankruptcy, or similar proceedings, and (2) all the issuer's obligations are current and the collateral manager believes they will remain current, then the corresponding S&P Global Ratings rating input for such an obligation is 'CCC-'.
• (viii) With respect to collateral obligations whose rating input cannot be determined using any of the steps described in subparagraphs (i) through (vii) above, then the corresponding S&P Global Ratings rating input is 'CC'.

7. For debtor-in-possession (DIP) financings, the issue point-in-time rating may be used as the S&P Global Ratings rating input for a maximum of 12 months from its initial assignment. However, we may further limit the use of the point-in-time rating if we believe that the credit quality of the DIP loan has deteriorated since its assignment. In order to make this assessment, we may request the collateral manager to provide information related to the DIP loan, such as amortization modifications, extensions of maturity, reductions of the principal amount owed, or nonpayment of timely interest or principal due. We also request that the collateral manager provide us with any other information that, in his or her reasonable business judgment, may have a material adverse impact on the credit quality of the DIP asset.

8. For the purpose of determining the S&P Global Ratings rating input:

• If the rating assigned by S&P Global Ratings or another CRA to an obligor or its obligations is on CreditWatch positive, such rating input will be treated as being one notch above the assigned rating.
• If the applicable rating assigned by S&P Global Ratings or another CRA to an obligor or its obligations is on CreditWatch negative, such rating input will be treated as being one notch below the assigned rating, subject to a floor of 'CCC-'.

9. We typically expect this same approach to apply in all instances where an S&P Global Ratings rating is being utilized within the transaction documents.

In order to perform our surveillance analysis, we rely on the consistent and timely availability of credit opinions for determining the rating input in CDO Evaluator. Therefore, when S&P Global Ratings' rating input is determined by using another CRA's rating as per paragraph 6(v) above, we would look for the manager, trustee, or other collateral administrator to consistently and timely provide S&P Global Ratings with the CRA's ratings. We would also look for the transaction's documentation to reflect, at inception, which CRA ratings the collateral manager chooses to include, as well as the reporting requirements and other relevant definitions that describe the determination of S&P Global Ratings' rating input. If, in our surveillance analysis, we determine that the information provided by the manager, trustee, or collateral administrator is not available, inconsistent, or not timely, we will derive our rating input as per paragraphs 6(vii) and (viii) above.

B. Determining asset maturities used in the CDO Evaluator

10. According to our criteria, the tenor of the exposure to a corporate asset is a parameter of our credit analysis. When we use the CDO Evaluator to conduct our credit analysis, we therefore input a maturity date for each asset.

11. Where we base our analysis on the "stable quality" approach, we will generally use the final maturity of each loan in the portfolio for the purpose of our credit analysis. We may adjust some asset maturity inputs if the resulting pool's weighted-average maturity is less than the length of the reinvestment period.

12. We would then use this portfolio's weighted-average maturity, adjusted as per the paragraph above as appropriate, for those aspects of the criteria that refer to the portfolio's weighted-average maturity, for example to determine the default patterns and the pool's modeled amortization profile.

13. Where a "stressed portfolio" approach is used, we would determine the maturity date inputs based on the transaction's covenants.

C. Asset recovery assumptions (see paragraphs 36-42 of the corporate CDO criteria)

Recovery rates based on recovery ratings

14. Table 1 presents our assumptions for assets with recovery ratings. In addition to the recovery rating, we may provide a recovery point estimate, which is used to signal whether an asset's expected recovery rate resides in the upper or lower end of the range for a given recovery rating. These recovery point estimates are expressed in increments of 5%. For example, if we indicate that an asset has a '3' recovery rating with a recovery point estimate of 50%, this would indicate that the recovery rate assigned by these criteria will fall at the low end of the '3' recovery rating range. This means that, in this example, assuming a CDO target rating of 'AAA', the recovery rate input is 30%. Absent any such information, we will use the lowest range for that recovery rating in table 1.

Table 1

15. If an asset does not have a recovery rating, then we assess whether it is pari passu or subordinate to other debt that does have a recovery rating. This is necessary because it is possible, for example, that the CDO holds subordinated debt of a company that has senior secured debt with negligible recovery prospects (that is, a recovery rating of '6'). Because the debt with a recovery rating is senior to the instrument that the CDO holds, the recovery prospects for the instrument held by the CDO will very likely be less than the recovery prospects for the senior secured debt with the recovery rating.

16. If the CDO holds senior unsecured debt that does not have a recovery rating, and is subordinate to debt that has a recovery rating, then the recovery of the instrument can be determined using tables 2 and 3 below.

Recovery rates for assets junior to assets with recovery ratings

Table 2

Table 3

Recovery rates based on asset type

17. If a recovery rating is not available for use as described above, we apply table 4 below.

18. Table 4 shows the recovery assumptions for corporate and sovereign assets held in a cash flow CDO, based on the different corporate asset types (loans/bonds), and their seniority, security, and country groupings.

Table 4

19. Specifically, in applying the table above for super-senior revolving loans (or super-priority revolving facility loans), we assume the senior secured loan recovery rates provided that the super senior revolving loan in the CLO is limited to a small percentage of the associated first-lien loan. In this case, we would expect the transaction documentation to limit the sum of the principal balance and unfunded commitments of these loans to a percentage (20% for example) of the sum of the revolving facility amount, plus the principal balance of the loan, plus the principal balance of any other debt that is pari passu with such loan.

20. The country recovery grouping we apply in our analysis is found in table 16 in Appendix B of this document.

21. Generally, CLO transactions have the ability to purchase first-lien last-out loans. First-lien last-out loans are generally viewed as obligations that have the benefit of a first-priority perfected security interest in an obligor's collateral package, but also have the ability to become subordinate in right of payment to the obligor's senior secured loan upon the occurrence of a default. Due to the subordinated nature of this type of loan following a default--and that any recoveries typically are subordinated until the senior secured loan has been fully repaid--we would generally view the recovery prospects for first-lien last-out loans similarly to those of second-lien loans.

22. When we conduct cash flow analysis as part of our surveillance of ratings, we use the current liability rating to estimate the ultimate expected recovery of a defaulted asset. This is in contrast to the way we analyze recovery rates in coverage ratios (see paragraph 90 below).

Recovery for synthetic CDOs

23. Asset recovery rates are drawn independently from a beta distribution using the mean and standard deviation from the synthetic recovery rates table below.

24. For synthetic CDOs, unless the terms of the credit derivative relating to deliverable obligations allow us to be more specific, we generally use the "senior unsecured bonds" asset type as our base-case recovery assumption, and we apply additional haircuts--or deductions--where the restructuring convention in the 1999 International Swaps and Derivatives Association's (ISDA) credit derivative definitions applies (also known as "old restructuring").

Table 5

D. Cash flow modeling assumptions

25. The following provides insight into the analysis that we employ in the cash flow modeling of CDO transactions. It expands on the methodology described in the corporate CDO criteria and global cash flow criteria. For the purpose of our cash flow analysis of corporate CDO transactions, we typically use our Cash Flow Evaluator model.

26. CDO transaction structures and collateral eligibility can vary significantly from transaction to transaction. We modify the general assumptions that follow to fit the unique circumstances of each transaction. While comprehensive, this guidance does not attempt to cover all the cash flow modeling stresses that might be applied to any particular transaction.

Determining the maturity and amortization profiles in the Cash Flow Evaluator (see paragraph 45 in the corporate CDO criteria)

27.During the reinvestment period.   For transactions that are still within their reinvestment period, given the active management of the portfolios, the initial portfolio maturity profile could change over this period and differ significantly from that at the end of the reinvestment period. Based on our observations of the historical amortization profile, we have therefore derived standardized amortization curves that we use in our cash flow analysis for portfolios with a weighted-average maturity ranging from four to seven years. We will typically apply these curves such that the weighted-average maturity using this amortization profile will be as close as possible to the portfolio's weighted-average maturity that was determined to assess credit risk using the CDO Evaluator.

Table 6

28. Where other factors may lead to an atypical amortization profile, we may construct specific amortization profiles reflecting those factors, depending on the asset pool's maturity, the length of the reinvestment period, and the maximum covenanted weighted-average life, for example.

29.After the reinvestment period.  For transactions that are past their reinvestment period, we will generally assume the actual portfolio maturity profile. We do so because after the reinvestment period, the transaction can become relatively static and collateral managers are not typically permitted to reinvest the scheduled principal proceeds. However, to the extent that the transaction documents allow the manager some flexibility in reinvesting certain proceeds after the end of the reinvestment period, we may also consider other amortization profiles, including those typically used in reinvestment periods, in our cash flow analysis.

30. The cash flow modeling for a given transaction may show that there are insufficient proceeds to pay full interest on nondeferrable tranches. If we believe these interest shortfalls are due solely to the modeled portfolio amortization profile, we may make minor adjustments to it. This is based on observations that collateral managers typically forecast and manage cash flows by adjusting portfolio maturities, holding back on reinvestments, and selling assets to avoid such shortfalls. Historically, we understand that managers have not invested 100% of their available cash and have maintained small amounts of cash on hand.

Default timing and patterns (see paragraph 45 of the corporate CDO criteria and paragraphs 9-10 of the global cash flow criteria)

31. We determine default timing and patterns considering the pool's and the transaction's characteristics, in particular, the portfolio's credit quality and weighted-average maturity. We generally model annual default amounts spread out in each note payment period. We additionally assume that assets do not pay interest to the CDO in the period in which they default.

32. For typical CLO pools of leveraged loans issued by speculative-grade issuers with weighted-average maturities ranging from four to seven years, we will typically run four default patterns starting in year one (see table 7). We believe that these patterns appropriately stress the amount of excess spread that would be available to the equity component of combination notes. However, we may consider a limited amount of excess spread in our analysis of payment flows to combination notes involving an equity component because the rating of that combination note is typically highly sensitive to the assumptions on excess spread, and therefore could otherwise experience significant rating movements in the very short term. Our analysis in this regard, would depend on our forward-looking outlook of the market over a maximum of one year.

Table 7

33. For shorter or longer tenors, we would adjust these patterns to better match these pools' maturity profiles. For example, for three-year weighted-average maturity portfolios, we would typically run the following patterns:

Table 8

34. For transactions with pro rata payment features, we may apply additional default patterns, for example, to test the ability of the transaction to withstand back-loaded defaults.

Interest rate patterns (see paragraph 45 of the corporate CDO criteria and paragraphs 5, 16, and 18-24 of the global cash flow criteria)

35. To assess whether a transaction will be able to perform in varying interest rate environments, we generally apply five interest rate scenarios to each default pattern. These scenarios (excluding the forward curve) are derived using Cox-Ingersoll-Ross methodology that simulates interest rate curves at various confidence levels depending on the rating scenarios in order to project future interest rate movements. The five interest rate scenarios are as follows:

• Forward curve, if available;
• Rising interest rates (up curve);
• Falling interest rates (down curve);
• Rising then falling interest rates (up/down curve); and
• Falling then rising interest rates (down/up curve).

Interest income on eligible investments (see paragraph 16 of the global cash flow criteria)

36. Proceeds received from assets in the form of scheduled principal and interest payments and recovery proceeds are generally held in eligible investments before being reinvested in substitute collateral or being used to pay liabilities on a payment date.

37. In our cash flow analysis, we assume that scheduled principal and interest proceeds are held in eligible investments for one-half of the collection period before being reinvested in substitute collateral. Additionally, since we also assume that recoveries are received at the end of a payment period, we do not model any interest earned on recovery proceeds in the period in which they are received.

38. We assume that interest is earned on the regular payments received from the eligible investments at a rate equal to the index referenced minus 100 basis points with a floor of 0%. In instances where transactions may use proceeds to pay interest on eligible investments as a result of negative interest rates, we may consider applying additional stresses, absent any mitigating factors.

Payment timing mismatch (see paragraph 45 of the corporate CDO criteria and paragraph 16 of the global cash flow criteria)

39. It is common for cash flow CDO transactions to include a bucket for assets that pay less frequently than the liabilities. In many instances, transactions use an interest reserve mechanism or enter into a basis swap to address this mismatch. In the absence of a mechanism that we believe mitigates this liquidity risk, we may model the mismatch by assuming the maximum concentration limitations of the assets that pay less frequently than the liabilities.

Foreign exchange risk analysis (see paragraphs 48-50 of the corporate CDO criteria)

40. Where structures only partially mitigate the foreign-exchange risk, for example through the use of a natural hedge, and/or of derivatives (such as swaps with predetermined notional options), we run additional cash flow stresses to capture the foreign-exchange risk in the rating analysis.

41. A natural foreign-exchange hedge exists when both the assets and liabilities denominated in each currency make up the same proportion of a given pool. For instance, the collateral pool may have 70% euro-denominated and 30% U.S. dollar-denominated assets matched to 70% euro-denominated and 30% U.S. dollar-denominated liabilities. However, this natural hedge often does not immunize the CDO against foreign-exchange risk. This hedge remains perfectly balanced so long as defaults to the assets occur pro rata across the currency denominations. If defaults do not occur in proportion (the more likely scenario), the resulting imbalance would throw the natural hedge askew. The balance of the natural hedge could also be upset by principal payments on the assets or diversion of payment proceeds to pay down liabilities in a sequential pay structure triggered by the breach of a coverage test.

42. Residual foreign-exchange risk may also exist with the use of certain derivatives. For example, if an issuer enters into a foreign-exchange swap with a predetermined notional balance, the transaction is likewise susceptible to hedging imbalances.

43. To bias defaults, we use the following formula, where X represents the currency bucket that defaults are biased toward (expressed as a percentage of the portfolio), and Y the magnitude of the default bias: FX Default Bias=(Y*X)/(Y-1+X).

44. In applying the default biasing formula, we typically use a Y value expressed as an integer ranging from 2 (most stressful) to 4 (least stressful) to reflect our overall assessment of the potential sensitivity of a transaction to currency mismatches based on the parameters set out in the criteria.

45. When reviewing the coverage tests, we would, for example, consider the exchange rate at which foreign currency denominated assets are carried. In this regard, we typically believe that coverage tests in which assets are carried at the then-relevant spot rate of exchange are more protective of noteholders than those where the assets are carried at a given predetermined rate of exchange. While the latter may increase the stability of the coverage ratios, we also consider that the ratios become less predictive of the true value of the collateral at any given point in time, increasing the potential risk to the noteholders.

46. When reviewing reinvestment guidelines, we consider how specific provisions relating to the reinvestment of proceeds from assets may be used and any other feature of management agreements that we believe have a bearing on the level of risk the noteholders may be exposed to as a direct result of the currency mismatch in the transaction. An example of a provision we have seen that we view as increasing the risk is the manager's ability to reinvest payment/sale proceeds from assets in one currency into assets in a different currency (all else being equal), as these trades have the potential to erode par as the result of foreign-exchange risk.

47. We typically apply a Y value of 4 where the following conditions are met:

• The magnitude of the currency exposure is small in our view;
• Coverage ratios carry currency assets at the then-relevant spot rate of exchange at any point in time; and
• Reinvestment conditions, which in our view, limit the manager's ability to increase the transaction's exposure to foreign-exchange risk.

48. When derivatives are used to partially mitigate foreign-exchange risk, we may also perform additional analysis to assess their effectiveness in mitigating foreign-exchange risk over time. For example, we may want to consider whether the benefit from such derivatives in hedging the transaction may fluctuate over time.

49. We may consider an exposure minimal and not incorporate it into our cash flow analysis if, for example, foreign-exchange exposure arises from the documents, allowing the manager a time lag before entering into perfect asset swaps in relation to foreign-currency assets it has purchased. In this case, we would expect the time that the assets remain unhedged to be short (typically no longer than six months) and the proportion of unhedged assets at any given point in time to be limited (typically 2%-3%). In addition, we expect these assets' adjusted par in the coverage test to reflect the embedded foreign-exchange risk and would consider this value in light of our view of foreign-exchange risk over the unhedged period.

50. In addition to the hedging of periodic payments, our analysis depends on whether the foreign-exchange strategy remains in place to cover the recoveries realized on the defaulted securities. Automatic termination of a foreign-exchange swap upon default of an asset exposes asset recoveries to foreign-exchange risk and potential termination costs. We typically adjust the recovery rate assigned when the swap is required to terminate before the time we believe recoveries will be received. The magnitude of this adjustment is determined according to factors such as the length of time the defaulted asset is exposed to foreign-exchange risk and the particular currencies involved.

51. Foreign-exchange risk also arises when an asset is sold, but the asset-specific foreign-exchange swap is not automatically retired, or, conversely, the foreign-exchange swap terminates before the asset matures. In the first instance, the collateral manager is likely to include the economic effect of the swap in making its sell decision, and, in the latter, the manager may sell the unhedged asset to eliminate foreign-exchange concerns. In both cases, noncredit-based considerations are factored into the decision process, and we consider adjusting the recovery rate assigned.

Pay-in-kind (PIK) assets (see paragraph 45 of the corporate CDO criteria and paragraph 16 of the global cash flow criteria)

52. When more than 5% of the assets in a portfolio by par balance have the ability to pay in kind, we apply a PIK stress to assess the ability of the CDO transaction to make payments on the notes despite the liquidity stress, and/or assess the adequacy of any mitigation schemes, such as liquidity facilities. We determine the PIK stress after taking into account the transaction structure and targeted portfolio profile. We typically do this only for the cash flow scenario yielding the lowest break-even default rate (BDR) to test that the notes would be able to be paid in the rating scenario considered. However, we would assess the minimum BDR without running this stress. For example, a transaction that allows for the purchase of PIK assets up to 7.5% of the portfolio by par balance would be subject to a PIK stress on 2.5%--the amount in excess of 5% and not the entire 7.5% permitted under the documents.

Corporate mezzanine loans (see paragraph 45 of the corporate CDO criteria and paragraphs 16 and 18-24 of the global cash flow criteria)

53. Corporate mezzanine loans are common to many European leveraged loan CDO transactions. These loans have a junior secured position and typically include two components to their interest payments--a current-pay coupon and a PIK coupon. The latter coupon is structured in the loan documents to pay in kind from day one and accrues to principal; in effect, it behaves like a zero coupon bond.

54. We give credit to the accrued portion of the PIK coupon component in the cash flow modeling, subject to the following conditions:

• For the purpose of the coverage tests, we expect accrued PIK interest to be included in the overcollateralization test, provided that the accrued interest is consistently treated as principal proceeds. We do not expect any credit to be given in the interest coverage test because no interest is received in cash during the payment periods.
• We will look to the asset eligibility guidelines and transaction covenants for a minimum mezzanine loan bucket and a minimum PIK interest rate for the mezzanine loans to incorporate accrued PIK interest in our cash flow analysis.
• For the purposes of default and recovery, the defaulted balance is calculated as the product of the asset default rate and the par balance, inclusive of the accrued PIK interest. The recovery balance is calculated as the product of the recovery rate and the par balance, excluding the accrued PIK balance.

E. Supplemental tests (see paragraphs 53-66 of the corporate CDO criteria)

55. We use our CDO Evaluator model to apply the following standard supplemental tests, as described below. According to the criteria, adjustments may be made to these standard tests, such as specified combinations of defaults of the largest obligors and industries as an alternative supplemental test if we believe that better addresses the transaction's specific risk profile. For example, this may be the case if the portfolios are either not well-diversified or have started amortizing after the reinvestment period, and show higher obligor or industry concentration.

Largest obligor test

56. In applying this test, we generally assume a flat recovery rate for all assets of 5% to address event risk. In exceptional cases, for example, where the entire collateral pool structurally can only comprise assets with the highest recovery ratings, we may use a higher assumption. For sovereign assets, the recovery rate used to calculate the largest obligor default test is 25%.

57. For example, we would expect a 'AAA' rated tranche to have sufficient credit enhancement to survive the highest level of losses associated with the defaults of each of the following combinations of underlying obligors, assuming 5% recovery (for sovereign assets, the recovery rate used for the purpose of this test is 25%):

• The two largest obligors rated between 'AAA' and 'CCC-';
• The three largest obligors rated between 'AA+' and 'CCC-';
• The four largest obligors rated between 'A+' and 'CCC-';
• The six largest obligors rated between 'BBB+' and 'CCC-';
• The eight largest obligors rated between 'BB+' and 'CCC-';
• The 10 largest obligors rated between 'B+' and 'CCC-'; and
• The 12 largest obligors rated between 'CCC+' and 'CCC-'.

58. For transactions that do not employ excess spread, such as synthetic CDOs, we consider whether the attachment point is set sufficiently high to withstand the highest losses from the obligor test without breaching the rated tranche's loss attachment point.

59. In cases where we apply this test by running our cash flow modeling according to the criteria, if the transaction allocates principal pro rata, we would apply the default rate derived from the application of the largest obligor or industry tests at different times during the life of the transaction on a prospective basis.

60. For this test, we would treat all obligors rated below 'CCC-' and still included in the CDO asset pool to have defaulted. Also, in applying this test, the value we assume for defaulted assets already held by the CDO is the lower of our recovery assumption or the current market value. For defaulted synthetic reference obligors, the value we assume is the respective recovery value shown in table 5 until the actual recoveries are determined through the International Swaps and Derivatives Association protocol or the applicable valuation mechanism detailed in the transaction documents. If the transaction documents specify fixed recoveries, we use the fixed recovery amounts.

Largest industry test

61. In the primary largest industry test, we would expect corporate CDO tranches rated 'AAA' or 'AA' to be able to withstand the default of all obligors in the largest single industry in the asset pool with a 17% recovery rate. For this test, we use the same industry classifications as in the CDO Evaluator. For example, assume a transaction has a 12% concentration in the largest industry. Under the test, a tranche rated 'AAA', 'AA+', 'AA', or 'AA-' in such a transaction should have sufficient credit enhancement to survive the default of 9.96% (12% industry concentration [100%-17% recovery]) of the asset pool. This is applicable even if the CDO Evaluator simulation model indicates that a lower level of credit enhancement would be sufficient.

62. The 17% assumption is the same recovery rate that we assign to senior secured debt from Group C countries (see the "Asset Recovery Assumptions" section earlier in this article).

63. We may still assign a rating of 'AAA' or 'AA' to a tranche even though it fails the primary largest industry test if it passes the alternative largest industry default test. The flat recovery assumption for this test is 5%.

F. Qualitative factors and additional testing for nondiversified or nonstandard portfolios (see paragraphs 15-16 in the corporate CDO criteria)

64. The standard modeling assumptions and stresses described in our criteria are applicable to the majority of CDO transactions that are typically well-diversified portfolios across obligors, industries, and asset characteristics. In addition to our standard assumptions, we will also consider other qualitative and quantitative factors and may assess additional scenarios for portfolios that may be atypical or not well-diversified. This could include transactions where portfolios show a high concentration of obligors, industries, or certain types of assets, or where the portfolios are lumpy, with a large variance of spreads and recoveries.

65.Changes to correlation assumptions.  We may modify some of our modeling assumptions or apply stresses for portfolios that could show heightened sensitivity to some of our modeling parameters. For example, for portfolios that are highly concentrated in one or a few industries, in addition to running the largest industry test, we may also run additional correlation scenarios to test lower and/or higher correlation assumptions than those we typically assume. Table 9 gives an example of the overrides we may make to our correlation assumptions.

Table 9

66. The above scenarios are for industries that display the 0.200 intraindustry and 0.075 interindustry correlations. As part of this analysis, we also make adjustments to the industry correlation override tables for correlations both higher and lower than the criteria assumptions.

67. In order to adjust the correlation for the higher correlation scenario, if the original correlation is less than 0.10, we would increase it by 0.025, and if it is greater than or equal to 0.10, we would increase it by 0.05. If the original correlation is less than or equal to 0.99, we would cap the adjusted correlation at 0.99. If the original correlation is greater than 0.99, we would set the adjusted correlation to the same value as the original correlation.

68. In order to adjust the correlation for the lower correlation scenario, if the original correlation is less than 0.10, we would decrease it by 0.025, and if it is greater than or equal to 0.10, we would decrease it by 0.05. If the adjusted correlation is less than zero, we would floor the correlation at zero.

69.Adjustments to recovery rates.  Empirical evidence suggests that the recovery rates for corporate assets are influenced by the level of defaults in the economy and the lending standard employed before entering the economic/default cycle. We have also observed considerable variation in recoveries within a given origination or default vintage. To the extent we see such variation, we may assess additional scenarios with 10% positive and negative adjustments to recoveries relative to a transaction's weighted-average recovery.

70.Default biasing:   Most CDO transactions are modeled based on the general pool characteristics, with pro rata defaults applied across all assets, as asset composition in CDO pools tends to be fairly uniform around the mean. For portfolios that are lumpy or barbelled, or have a concentration exposure to certain assets, we may consider default bias scenarios for:

• The largest assets in the pool;
• The assets in the pool with the highest spread;
• The fixed-rate buckets of assets; and
• The assets in the pool with the lowest base-case recoveries.

71. Where we bias default, we will typically apply defaults to the largest assets, assets with the highest spread, assets that pay a fixed rate of interest or assets that have the lowest base-case recoveries, as appropriate, in our cash flow analysis.

Part II: Transaction Document Analysis

A. Reinvestment provisions (see criteria paragraphs 67-69 of the corporate CDO criteria)

72. Most cash flow CDO transactions are actively managed. CDO transaction documents establish the parameters within which a collateral manager can modify the collateral portfolio through trading and reinvestment. Collateral managers generally seek to maximize trading and reinvestment flexibility. Although this increased flexibility may help collateral managers balance the needs of debt and equity investors, S&P Global Ratings believes that certain collateral characteristics are key to mitigating the risks to the transaction's ability to pay the rated debt. Generally, the overarching premise of a managed CDO is to preserve the aggregate collateral par amount and credit quality of the assets when trading, while balancing yield, tenor, and prospects for recovery. In reviewing all reinvestment provisions, we generally focus our analysis on the ability to substitute assets within the parameters described in the transaction documents and the extent to which credit enhancement may be eroded. The following are examples of how we view reinvestment provisions in CDO transactions we rate based on the "stable quality approach," which use S&P Global Ratings' CDO Monitor model as a condition to reinvestment. We refer to this as the CDO Monitor test in the rest of this document.

Table 10

Table 11

Principal collateral maintenance

73. Par maintenance:

• During the reinvestment period, generally, proceeds from discretionary and credit-improved sales are expected to be reinvested such that the new asset par amount after the reinvestment is maintained.
• After the reinvestment period, if the collateral manager chooses to reinvest rather than pay down the notes, we generally expect proceeds from credit-improved sales and unscheduled amortizations to be reinvested such that the new asset par amount after reinvestment is maintained or improved.
• Similarly, we view the maintenance of the overcollateralization (OC) test numerator as generally equivalent to the above conditions.

74. Par for sales proceeds:

• During the reinvestment period, generally, reinvestment of sale proceeds received from defaulted obligations and equity securities only need to have a par value equal to the related sales or recovery proceeds, as applicable.
• During and after the reinvestment period, we generally expect proceeds from the sale of credit risk obligations to be reinvested such that their par value equals their sales proceeds.

75. CDO documentation may also allow for the principal collateral amount to not be maintained if the manager has built sufficient excess par in the transaction, typically such that the aggregate performing collateral balance plus recovery on defaults is greater or equal to the reinvestment target par of the portfolio after reinvestment. Here, we refer to the "reinvestment target par" typically as the initial target par amount after accounting for any additional issuance, and any reduction from principal payments made on the notes.

76. After the reinvestment period, we generally expect that proceeds from discretionary sales, defaulted obligations, equity securities, and scheduled principal proceeds to be used to pay down the notes.

Coverage tests, CDO Monitor test, or SDR test

77. For all proceeds reinvested except for proceeds from the sale of credit risk, equity, and defaulted obligations, we expect the CDO Monitor test to be satisfied, maintained, or improved.

78. After the reinvestment period, the CDO Monitor no longer applies and we would look for the new obligation to have (a) the same or higher rating and (b) the same or shorter maturity than the one sold. As an alternative to (a) and (b) above, we would consider an obligation to maintain or improve the portfolio's scenario default rate (SDR). Similarly, we view maintaining or improving the weighted average maturity of the portfolio after giving effect to the purchase as generally equivalent to the "same or shorter maturity" provision.

79. Generally, other conditions for reinvestment mandate the assessment of the coverage tests. The coverage tests usually include OC and interest coverage (IC) triggers. During the reinvestment period, we would expect the coverage tests to be satisfied, or if not satisfied, to be at least maintained or improved. Typically, after the reinvestment period, we would expect the coverage tests to be satisfied as the transaction should be deleveraging. Where we observe indenture language that introduces scenarios where the tests may not apply, we may choose not to consider those tests in our cash flow modeling.

Trading plans

80. Some transactions allow for the assessment of trading parameters based on a series of trades (rather than on an asset-by-asset basis), which are typically referred to as trading plans.

81. Trading plans allow the collateral manager to sell and purchase a group of assets that may not fully meet the transaction's reinvestment guidelines individually but are expected to satisfy them after evaluating the bundling of several trades of eligible assets together that have offsetting characteristics. Generally, trading plans a) are limited to a small percentage of the collateral pool, typically 5%; b) do not extend past any determination date; c) have a time limit for completion (typically 10 days in U.S. transactions and 15 days in European transactions); and d) are limited to only one active trading plan at one time.

B. Coverage tests: (see paragraph 67 in corporate CDO criteria)

82. In reviewing provisions governing coverage tests, we review the calculation of the numerator of the test and the way certain types of assets are treated (see the section below titled Treatment of certain types of assets). We additionally consider whether all relevant liabilities are included in each coverage test's denominator, particularly including the event of certain note cancellations (see the section below title Note cancellation without payment).

Treatment of certain types of assets

83. Generally, a cash flow CDO structure includes OC tests that correspond to certain classes of notes. Typically, the OC test ratio is determined by dividing the adjusted principal balance of the assets (including any cash amounts representing principal proceeds) in the CDO portfolio by the sum of the aggregate principal balances of the CDO notes of the relevant class (including any deferred interest where applicable), and all other classes of notes that rank senior or pari passu. Generally, when the numerator of the OC test includes principal proceeds, these amounts are included so long as there is no duplication with proceeds in other accounts already included in the calculation, and such amounts may not be reclassified as interest proceeds.

84. Typically, the adjusted principal balance of the assets included in the numerator of the OC tests equals their par amount. However, in certain instances, especially where the assets are considered riskier, the carrying value in the OC test is reduced by haircuts. Since each of these assets are valued at less than their par value for purposes of calculating the OC tests, the presence of such loans in a portfolio will lower the numerator of the OC tests, thereby increasing the probability of triggering an OC test failure and potentially deleveraging the senior notes.

85. We typically expect certain adjustments to the principal balance of specific asset types in the calculation of the OC test numerator (see table 12). We discuss some of these assets in greater detail in the following paragraphs.

Table 12

86. When an asset included in the calculation of the OC test is eligible for a haircut, we typically expect: (a) assets forming the excess to be those that would result in the highest haircut and (b) the highest haircut to apply in instances where the asset can be eligible for two or more haircuts (resulting in the lowest carrying value).

Defaulted obligations

87. In accordance with S&P Global Ratings' corporate default studies, we consider defaulted obligations to be those a) where there has been a default in the payment of principal and/or interest on the issuer's obligation or obligations senior/pari passu to such defaulted obligation; b) where the issuer has become subject to a bankruptcy or insolvency proceeding (as applicable); or c) where the issuer has an issuer credit rating of 'CC', 'D', or 'SD', or had such a rating prior to withdrawal. If any of these conditions is met, we expect an obligation to be treated as defaulted.

88. However, in our analysis, we may consider certain defaulted obligations as performing if they meet specific conditions. The exceptions typically include debtor-in-possession (DIP) loans, current pay obligations, and distressed exchange current pay obligations. They also include obligors that have emerged from bankruptcy that are carried as 'CCC-' assets in the coverage tests for a limited period of time (typically not more than 12 months) pending the assignment of a rating input to the new entity.

89. Other than such exceptions, we expect that all defaulted obligations will generally be carried at the lower of a) their assigned recovery rate multiplied by their defaulted principal balance or b) their current market value for purposes of the OC tests. Where transactions treat assets that pay in kind for a defined time period as defaulted obligations, the recovery rate or market value rate is applied to the PIK asset's original par principal balance, not its principal plus accrued interest balance.

90. Our recovery rate assumptions are tiered, based on the rating scenario considered for each CDO tranche. Generally, the recovery rates used in the coverage tests are those applicable for the original rating of the CDO tranche. In such cases, transaction documents specify that one set of fixed recovery rates would apply for the life of the transaction. This is typically done to avoid a sudden increase of the OC levels due to downgraded liabilities of the CDO, thus possibly causing a failed OC test to cure.

91. Equity securities received as part of a workout are given no value in the OC tests.

92. We typically look for the method used to assess market value to be standardized, consistent, and independent from the collateral manager, and to be a reflection of the market opinion on the value of these obligations. Generally, the meaning of "market value" is derived by applying the following options:

• A price provided by an independent pricing service; or
• The average of three bids from independent broker/dealers; or
• If three bids are not available, the lower of two bids from independent broker/dealers; or
• If two bids are not available, a single independent bid from a broker/dealer.

93. If a value cannot be obtained by the collateral manager as described above, the value may be determined by the collateral manager in a manner consistent with its standards.

94. We expect a similar definition of market value to consistently apply--where necessary--to other asset types such as obligations that are deferring, distressed, current pay, or long dated.

95. Note that while the approach described above is generally typical for CLOs of broadly syndicated loans, we will evaluate on a case-by-case basis variations tailored for more narrowly constructed pools such as CLOs of mid-market loans.

Debtor-in-possession (DIP) loans

96. Debtor-in-possession (DIP) loans are loans made to bankrupt entities. Generally, in our analysis, we consider DIP loans that are current on their interest and principal payments as performing obligations due to their priority status under the U.S. bankruptcy code. DIP loans with S&P Global Ratings of 'CCC-' or above typically qualify for par treatment in the OC ratio regardless of market value.

97. We acknowledge that there may be times when a DIP loan is issued, but has not yet been assigned an issue rating by S&P Global Ratings. In such instances, these loans are generally carried at a rating of 'CCC-' for all purposes in the transaction documents. However, given the specific attributes of DIP obligations, the transaction documents may allow for the use of a slightly higher rating for a limited period of time (usually 'B-' for up to 90 days), so long as the manager reasonably expects that an S&P Global Ratings rating will be issued within such period and that it will be at least equal to the interim rating. The temporary rating should cease to be used upon the earlier of the assignment of a rating by S&P Global Ratings and the expiration of such period.

Corporate rescue loans

98. Certain transactions allow for the purchase of loans made to companies in a restructuring or insolvency process outside of the U.S. ("corporate rescue loans").

99. These loans are entitled to recover proceeds of enforcement of security shared with the other senior secured obligations of the borrower in priority to all such other senior secured obligations, either through the grant of security or otherwise (such as pursuant to insolvency legislation or other law).

100. We expect the rating input for such assets to be determined by application of paragraphs 6 and 8 of this guidance document.

101. The purchase of corporate rescue loans is typically limited to a small percentage of the collateral pool, typically 5%.

102. In our analysis, we would treat corporate rescue loans whose rating input is 'CC', 'D', or 'SD' as defaulted obligations.

103. We expect the recovery rate for such assets to be determined by application of paragraphs 14-22 of this guidance.

Workout related assets

104. Some CLO transactions allow for the ability to exchange, receive, purchase, or exercise a right to purchase an asset from a distressed obligor whose position is already held within the CLO's portfolio if the manager reasonably expects that doing so will result in better overall recovery. We generally understand that the ability to manage distressed credits is seen as a fundamental role of the collateral manager and note that, in some cases, eligibility requirements have hindered their ability to maximize recoveries by not allowing the transaction to participate in the acquisition of these new assets.

105. That said, we believe that the related provisions of the transaction documents should appropriately align with the interests of all noteholders. As such, we give particular scrutiny to the attributes of any such workout related assets, including, but not limited to, their proposed deviation from the collateral obligation term or eligibility criteria.

106. We also examine all mitigating factors when reviewing workout related assets, including, for example, the proceeds that may be used to attain such assets, their carrying value in the transaction once acquired, and how any proceeds generated would be treated. The following sections focus on some of the mitigating factors we examine in our analysis.

107. If a workout related asset is a debt obligation that is senior or pari passu to the collateral obligation already held by the issuer and it is expected that most of the collateral obligation/eligibility criteria requirements will be met, then it may be appropriate for such asset to receive a carrying value greater than zero within the transaction. Typical exceptions to the collateral obligation or eligibility criteria include, for example, the asset being a defaulted, credit risk, or long-dated obligation. In any event, our analysis would focus on each exception presented and its possible implications, and we generally expect such assets to receive defaulted treatment in the transaction.

108. If, on the other hand, the asset deviates further from the collateral obligation or eligibility criteria definition, we typically expect it to receive no credit.

109. Often the deployment of additional funds are needed to acquire workout related assets. If diverted interest proceeds or external contributions are utilized, we typically expect that any proceeds received from the workout related asset should be recovered as principal proceeds in an amount at least equal to the asset's OC carrying value. In addition, where interest proceeds are utilized, we generally expect there to be sufficient interest coverage to meet interest payments on all rated notes on the immediately upcoming payment date.

110. If the use of principal proceeds to acquire workout related assets is permitted, we generally expect that the performing aggregate collateral balance plus recoveries will remain above the reinvestment target par after giving effect to such purchase. Similar to the above, we typically expect that any proceeds received from the workout related asset should be recovered as principal proceeds in an amount at least equal to the asset's OC carrying value.

111. If the maintenance of the reinvestment target par condition is not included, workout related assets that are senior or pari passu to the collateral obligation already held--and that have limited deviation from the collateral obligation definition (as described in paragraph 107 above)--may be purchased with principal. This is provided that:

• The transaction's OC tests (including, in certain instances, the interest diversion test) are passing after the purchase, and
• All proceeds received from such asset would be treated as principal proceeds until the full par amount of the original asset plus the greatest of (a) the principal proceeds used for purchase and (b) the OC carrying value of the new workout related asset are recovered (notwithstanding any other mitigating factors).

112. In instances where just an asset exchange is occurring (with an expectation of better recovery)--such exchanges are typically limited to a small percentage of the portfolio, but may not be restricted to the same obligor as the currently held asset. In addition, they are typically limited to defaulted and credit risk assets, and often circumvent the typical investment criteria. As such, if a performing asset is being exchanged for another performing asset, we generally expect the new asset to have the same or higher S&P Global Credit Rating and the same or shorter maturity. Furthermore, if additional payments (other than customary transfer costs) are required in connection with any such exchanges, we typically expect them to be limited to excess interest or external contributions.

Current pay obligations

113. We also recognize an exemption from defaulted obligation treatment for current-pay obligations, which meet the definition of defaulted obligations but have the following characteristics that qualify them for performing obligation treatment:

• For obligors that are not subject to a bankruptcy proceeding, the obligation must be current on all payments that are contractually due according to the underlying documents, including interest and principal payments (note that under "S&P Global Ratings Definitions," we do not recognize an obligation as current when payments are more than 30 days past due even if the contractual grace period is longer);
• In the reasonable business judgment of the collateral manager, the obligor will continue to remain current on its obligations;
• For obligors that are subject to a bankruptcy proceeding, the bankruptcy court must have issued an order authorizing payments, and the obligation must be current on all such authorized payments; and
• The market value of an obligation should be at least 80% of par, regardless of the asset's rating. We view 80% as the demarcation line for the market's perception of whether an asset is distressed. If the market value of a current-pay obligation falls below 80%, or if no independent mark is available for a current pay asset, the obligation ceases to qualify as a current-pay obligation and reverts to defaulted obligation treatment.

114. The main impact of classifying an asset as current-pay rather than defaulted is the carrying value used to calculate the OC tests. Defaulted securities are carried at the lesser of market value or S&P Global Ratings' recovery value, whereas current-pay obligations may be carried at par. For the purposes of the CDO Evaluator and CDO Monitor, we assume a performing rating input of the higher of its issue rating or 'CCC'. We expect any current-pay obligation with a market value determined by reference to S&P Global Ratings' recovery rate to receive recovery value credit in the calculation of the OC tests.

115. We generally expect current-pay obligations to be limited to a maximum of 10% of the collateral balance. We will treat any current-pay obligations in excess of 10% of the portfolio as defaulted obligations in our analysis. We incorporate current-pay assets in excess of the maximum bucket into our credit analysis using CDO Evaluator (see example in table 13 below).

'CCC' rated obligations

116. Because we view obligations rated in the 'CCC' category as having a high risk of further credit deterioration, most transactions include OC test haircuts to the carrying value of 'CCC' rated assets above a predefined threshold. While S&P Global Ratings' CDO Evaluator generally takes into consideration the likelihood of defaults of 'CCC' rated assets based on their historical performance, we believe these assets are more exposed to event risk. Because the historical performance of our rated universe for this rating category may be less predictive, we expect the transaction documents to limit exposure to 'CCC' category obligations.

117. Typically, we expect the 'CCC' bucket to include all obligations rated in the 'CCC' category regardless of their market value, including current-pay obligations (which are assumed to be rated in the 'CCC' category), discount obligations, and DIP loans rated in the 'CCC' category. If the manager applies the CDO Monitor test when purchasing new assets, the CDO Monitor would take into consideration the credit quality of the assets. However, we generally expect transactions to include haircuts in the coverage tests for 'CCC' rated assets when the exposure to 'CCC' rated assets exceeds 7.5% of the portfolio amount. For transactions where the haircut in the coverage test starts at a higher exposure, we may take into consideration mitigating factors such as modeling the excess exposure in CDO Evaluator at closing.

118. The examples in table 13 below summarize the methods by which we may model excess exposure to current pay and 'CCC' obligations in CDO Evaluator. In accordance with paragraph 115 of this guidance document, we expect any current pay securities in excess of 10% to be treated as defaulted obligations in the coverage tests. We consider the obligations included in the 10% current pay concentration limit are 'CCC' rated assets in our analysis and include them in the calculation of the 'CCC' excess described in paragraph 117.

Table 13

Distressed exchange obligations

119. We may treat debt obligations of issuers that have launched distressed exchange offers as current-pay securities rather than defaulted securities, subject to certain conditions. To start, the CDO must already hold the asset, and the asset must be current on all principal and interest payments that are due and payable according to the underlying documents. Additionally, if there is an exchange, the offer must be a debt-for-debt exchange, or if it is a cash buyback offer, the debt to be repurchased must be retired. Lastly, the debt issue held by the CDO must have an equal or higher seniority ranking in the capital structure than the issue subject to the distressed exchange or buyback.

120. If all of these conditions are met, the 80% of par market value test may not, as part of our analysis, apply to distressed exchanges on the date of the offer unless the CDO holds debt subordinate to the debt subject to the distressed exchange. Also, if the CDO holds any other current-pay security not subject to a distressed exchange, the 80% of par market value test would still apply to our analysis.

121. For purposes of our analysis, we assume the old debt instrument tendered by the CDO will be carried at the current par value allowed by the CDO documents until the tender or buyback offer is finalized, as there is no certainty the distressed exchange transaction will be completed.

122. Once the tender period is finalized, even before the final ratings are assigned, we assume the old debt instrument will be carried at the new lower par value that the new instrument document promises will be paid for the issue obtained in the exchange. If the offer is for a buyback, we assume the instrument will be held at the net monetary value that will be received from the buyback once the tender offer is completed.

Discount obligations

123. Most transactions permit some portion of the collateral pool to be purchased at significant discounts from par. Generally, we view assets purchased below the lower of (a) 80% of par and (b) the price (or close to the price) of a highly recognized loan or bond index (as applicable) as discounted obligations, and expect that these assets will be carried at their purchase price. If a transaction permits the purchase of assets at unusually deep discounts from par, we would likely adjust our modeling of the transaction to reflect our view of additional risk.

Long-dated obligations

124. The inclusion of corporate assets that mature on a date beyond the earliest legal final maturity date of the liabilities may require the CDO transaction to sell these assets before their underlying maturity. This exposes the transaction to noncredit-related losses.

125. This concern is addressed primarily by limiting the concentration of assets in the long-dated bucket to a small amount, such as 5% of the portfolio balance. When the allowance for this bucket exceeds 5%, we expect the par credit in the OC test numerator for each long-dated asset in excess of 5% to be reduced. The haircut may vary depending on the number of years by which the maturity of the obligations exceed the transaction's legal final maturity date. We generally expect to see an OC haircut of at least 10% for each year that the underlying asset maturity exceeds the earliest legal final maturity date of the rated notes. Alternatively, we have also seen documentation applying an OC test carrying value of the lesser of current market value and a haircut of 30% of par, which may be adjusted depending on the difference between the underlying maturity date and the transaction's legal final maturity date.

126. To the extent that exposure to long-dated securities exceeds 5%, and the transaction documents do not apply adequate haircuts to the OC test numerator value, we may apply additional adjustments when we analyze the transaction, for example by modeling the potential par loss incurred for the forced sale of the asset under less-than-ideal market conditions.

127. Regardless of the presence of OC test numerator haircuts for long-dated securities, when the transaction exhibits a significant exposure to long-dated assets, for instance in excess of 20% of the performing collateral, we may apply additional stresses such as those indicated in our market value criteria (see "Methodology And Assumptions For Market Value Securities," published Sept. 17, 2013), especially as the number of years remaining until the transaction's legal final maturity decreases.

Maturity amendments

128. We expect that transaction documents would preclude a collateral manager from extending the maturity of an asset such that it becomes a long-dated, unless the action is taken to avoid imminent default or minimize material loss due to materially adverse financial condition.

Zero-coupon obligations

129. We expect zero-coupon obligations to be carried at their accreted value for OC test purposes.

C. Bivariate risk (see paragraph 10 of the corporate CDO criteria)

130. Under our CDO criteria, we analyze the credit risk of a portfolio of assets based on an obligor's credit quality. However, this analysis does not capture situations where the credit risk arises both from the default of the loan or bond's issuer and a third party. The section below highlights mitigants we consider to this additional source of credit risk.

131. Transaction structures may include "baskets" for assets with bivariate credit risk, which generally include loan participations, securities lending agreements, and other agreements that may expose the transaction to counterparty risk. The basket limitations are a mitigating factor to the credit risks introduced to the CDO by their respective counterparties. In a securities lending agreement, for instance, the issuer, as the lender, maybe exposed to the credit risk of the borrower if the borrower defaults and is unable to return the securities borrowed.

132. Bivariate-risk asset baskets are usually limited to 20% of the aggregate pool balance; however, we expect the baskets to be further limited based on the rating of the counterparty, as shown in the table below:

Table 14

133. In analyzing bivariate risk buckets, we typically reference long-term ratings on the counterparty. Certain counterparties may have only short-term ratings or reference only short-term counterparty ratings in their documentation. In such cases, we would infer a long-term rating from the documented short-term rating. This is the lowest long-term rating that maps to the relevant short-term rating, according to our criteria for linking long- and short-term ratings (see "General Criteria: Methodology For Linking Long-Term and Short-Term Ratings," April 7, 2017).

Participations

134. In a typical loan participation, a participation buyer owns a beneficial interest in the loan, and the participation seller owns the legal interest in the loan and, as such, maintains the servicing responsibilities and relationship with the borrower. A participation agreement dictates the terms of the participation transfer and the participation buyer's and seller's rights with respect to the loan and its proceeds.

135. Because the participation seller maintains legal title in the participated loan, it is the lender of record, receives loan payments, pursues collections against the borrower, and performs other loan-servicing obligations on behalf of the participation buyer.

136. Consequently, if the participation seller files for bankruptcy, the participation seller's bankruptcy case may delay or otherwise disrupt the participation seller's ability to service the loan and forward loan proceeds to the participation buyer. Therefore, the participation seller's creditworthiness may generally be a risk to the rated notes. Furthermore, the failure to elevate the participation to an assignment prolongs this risk to the rated notes.

137. When a participation seller is not in our view a bankruptcy-remote entity, we consider whether the transaction's documentation includes incentives to elevate the participation interest into an assignment in order to conclude that its credit risk to the rated notes is mitigated. Incentives that we typically observe and view as effective include, but are not limited to, haircuts to the value of the participated collateral generally consistent with the asset carried at recovery value if the necessary consents for the participated loans are not obtained within a reasonable time period, typically by the transaction's effective date.

138. If we determine that a participation seller is a bankruptcy-remote entity, we generally do not consider whether its credit risk is mitigated because we consider that entity's bankruptcy risk to be sufficiently remote.

D. Defining interest and principal proceeds (see paragraph 67 in corporate CDO criteria)

139. Generally, transaction documents include distinct definitions for interest and principal proceeds that are received from the underlying collateral and include provisions governing how each income stream is applied during the CDO's life. We review these provisions giving particular scrutiny where principal proceeds can be reclassified as interest proceeds and passed down the payment waterfall to equity investors as discussed in the paragraphs below. Such provisions have the effect of increasing the immediate return to equity investors while reducing the credit support available to offset future losses.

140. We generally expect recoveries on defaulted securities to be treated as principal proceeds until the defaulted securities' original par amount is recovered. Accordingly, we expect proceeds from equity securities held by the issuer and any sub-SPEs to be treated as principal proceeds until the original defaulted securities' full par amounts are recovered in cash.

141. We also expect the entire accreted value of zero-coupon obligations to be paid as principal proceeds when the asset is repaid.

142. A more restrictive definition of what constitutes interest and principal proceeds can provide comfort to debtholders, as anything that is classified as principal will either be used to purchase additional collateral or amortize the secured notes.

143. Where principal proceeds may be recharacterized as interest proceeds, we will review the provisions that allow for par leakage and the presence of mitigating factors in transaction documents. The paragraphs below highlight the types of recharacterization provisions we commonly see. We analyze other recharacterization provisions from a similar perspective.

144.Trading gains:   Some documents allow for recharacterization of principal proceeds into interest proceeds when a trading gain has been realized. We expect this to be possible only when the aggregate performing collateral balance (including recoveries) remains above the reinvestment target par balance.

145.Excess in the ramp-up and/or principal accounts and partial refinancing:   At times, the transaction may reach the effective date target par amount or partly refinance and have cash remaining in the principal account or ramp-up account. Some transaction documents permit the issuer to reclassify such proceeds as interest proceeds and allow the excess par to be released to equity. We expect that this reclassification a) will be restricted to a small percentage of the portfolio; and b) will not result in the performing aggregate collateral balance (including recoveries) falling below the effective date target par amount after giving effect to the distribution. When the issuer provides notice or requests a rating agency confirmation with respect to a partial refinancing, we review the impact of the amendment and related cash flow release to equity to determine the impact on the outstanding ratings. If the transaction documents allow for cash flow release to equity ahead of the rated notes, we may only give credit up to the target par amount in our surveillance analysis, thereby limiting potential upgrades due to the increase in par.

146. At times, however, the leakage of the excess ramp-up amount may occur on other payment dates following the effective date. Generally, we expect the recharacterization and leakage will be limited to the first or second payment date. We believe that as the time horizon permitting recharacterization increases, the degree of sequence risk to the noteholders increases because the transaction's cushion against defaults or negative credit migration may be eroded. To that end, we may make qualitative adjustments when we see unusually long time horizons during which such leakage can occur.

147.Amortizing reinvestment par amount:   Some transactions implement an amortizing reinvestment target par amount purportedly to account for the built-in cushion against losses that exists at the closing date. This is a predetermined reduction in the value of the transaction's target par amount unrelated to whether principal payments are actually made on the notes or not. This is typically used to allow for the recharacterization of principal proceeds as interest proceeds, when the collateral par exceeds these predetermined amounts. We believe that referencing an amortizing reinvestment target par amount would affect the collateral quality tests, the reinvestment criteria, and potentially increase payment of proceeds from trading gains to equity investors instead of distributing principal to the notes according to the payment waterfall. Therefore, we believe particular provisions such as these present an enhanced level of risk to the rated notes, and we generally expect they would not apply, unless other mitigating factors are in place to limit this risk.

148.Exercising warrants:   A warrant is an option that entitles the holder to buy the underlying stock of the issuing company at the exercise price until the expiry date. Generally, we see CDOs receiving such options through the workout of a distressed obligation. We typically see warrants exercised only with interest proceeds. However, some transactions permit principal proceeds to be used. When principal proceeds may be used to exercise a warrant, we expect the performing aggregate collateral balance (including recoveries) to remain above the reinvestment target par after giving effect to the exercise of the warrant. We also expect sale proceeds from equity securities acquired via warrants to be considered as principal proceeds.

E. Treatment of certain types of liquidity risk: assets paying less frequently than liabilities

149. This risk arises when the current frequency with which payments are received on assets is less than the frequency with which payments must be made on the rated notes. We outline below typical mitigating factors to address such mismatch.

150. We consider this concern to be addressed if the amount of such assets that the issuer can purchase is limited to a small percentage of the collateral pool, typically 10% at any time.

151. Alternatively, where interest smoothing mechanisms are incorporated and triggered when the current amount of such assets exceeds a predetermined threshold, we expect this threshold to be limited--for example, 10% of the collateral pool.

152. Where transaction documents include interest smoothing mechanisms where thresholds are set at higher levels, we typically expect additional conditions to be satisfied, such as interest coverage covenants combined with (1) maximum limits on less frequently paying assets and/or (2) par coverage covenants.

F. Note redemption, amendments, refinancing, and repricing (see paragraph 74 of the corporate CDO criteria)

Indenture amendments and related consents 

153. Transaction terms in a CDO indenture can typically be amended at any time by entering into a supplemental indenture, subject to certain conditions. Some amendments do not require noteholder consent if they do not materially or adversely affect any noteholder. These may include implementing name changes, clarifying language, conforming to changes in law, or modifying terms to conform with rating agency methodologies.

154. Other types of amendments may require some form of consent from each noteholder who may be materially affected by the proposed changes. These may include amending the stated maturity date, interest rate, or principal amount of the notes, the payment priority, or certain definitions that affect noteholder consent. Depending on the transaction, the process to execute a supplemental indenture that requires noteholder consent varies. Generally, trustees and collateral managers track down the noteholders and obtain their consent to make a change; if they are unable to reach all noteholders deemed to be materially affected, they may be unable to implement the proposed change. However, in some cases the transaction documents include "deemed consent" or "negative consent" provisions, which allow the trustee and collateral managers to assume noteholder consent if they send out a notice of change and don't receive a formal objection within a predetermined period.

155. S&P Global Ratings expects to receive notice of all amendments prior to and upon their execution. We will review the amendments and determine whether they have an impact on the rating of the notes. If we determine the amendment gives us cause to take a rating action on publicly rated notes, we will publish the rating action and our rationale. As S&P Global Ratings is not a party to the transaction, we generally do not comment on the requirement or process of obtaining noteholders' consent. However, note that:

• We view some changes, such as entering into a hedge agreement, as having a greater likelihood of affecting the ratings than others; and
• We have seen some document provisions giving the collateral manager the ability to make certain changes without an amendment. We would likely review such changes on a case-by-case basis.

Redemption provisions  

156. The transaction documents of most CDO transactions contain provisions allowing for early redemption, refinancing, or repricing of the rated liabilities. The underlying principle of our rating methodology is to address whether the holders of the rated notes receive timely and ultimate payment of interest and principal on the rated notes.

157. We look for redemption, refinancing, and repricing provisions to be consistent with our view of payment to the rated notes. Our analysis typically takes into account how the rights of the holders of rated notes to receive payment in full may be affected by such provisions.

Optional redemption of rated notes 

158. Most CDO transaction documents contain provisions that permit the holders of the equity or the subordinated notes to call the notes in whole after a certain date (usually after the non-call period as defined in the documents). The redemption price includes all interest accrued to the redemption date and full payment of the outstanding principal amount. We generally view this type of redemption provision as ratings-neutral since the holders of the rated notes would be repaid in full.

159. To the extent that optional redemption provisions permit holders of the rated notes to receive less than the full principal amount, or physical delivery of the collateral, we expect 100% consent of the noteholders to be required.

Refinancing of the rated notes 

160. Many CDO transaction documents include provisions for the refinancing of the rated notes, either in whole (where all the rated notes are redeemed) or in part (where only certain classes are fully redeemed).

161. We view refinancing of all the rated notes as analogous to an optional redemption because the holders of the existing rated notes would be paid full principal and accrued interest from the proceeds of the newly issued refinancing notes. We typically expect to be notified in advance of such refinancing in whole.

162. We typically expect certain conditions to be met in connection with a refinancing in part. Examples of such conditions include, but are not limited to:

• Notice to S&P Global Ratings is provided in advance of such refinancing;
• Refinancing proceeds along with other available funds are sufficient to pay the full outstanding principal amount and accrued interest on the existing notes;
• The principal amount of the refinancing notes is equal to the principal amount outstanding of the existing notes on a class-by-class basis;
• The maturity date of the refinancing notes is no earlier than that of the existing notes;
• The interest rate of the refinancing notes does not exceed the interest rate of the existing notes; and
• The refinancing notes are paid at the same priority level as the existing notes.

163. Variations on the concept of refinancing that do not refinance all of the rated notes may have a ratings impact. We typically evaluate, on a case-by-case basis, the impact on the creditworthiness of the nonrefinanced notes. To the extent this potential impact cannot be fully evaluated at closing, we would likely highlight any material risk we believe such provisions may present.

Repricing of the rated notes 

164. Some CDO transaction documents include provisions for the repricing of the rated notes. We view repricing as potentially having a ratings impact due to the change in cash flow requirements repricing would entail. For repricings, we typically would review the transaction documents to determine if a rating action is warranted.

165. In our analysis, we typically pay special attention to the treatment of noteholders that do not consent to a repricing. Generally, we expect nonconsenting noteholders to be redeemed in full. To the extent nonconsenting holders' notes are not redeemed in full, or where consent is deemed from nonresponsive noteholders, we will evaluate such provisions on a case-by-case basis. We would likely highlight any material risk we believe such provisions may present.

G. Additional issuances and note cancellations

Additional note issuance (see Structured Finance: Asset Isolation And Special-Purpose Entity Methodology," published March 29, 2017) 

166. Most CDO transaction documents allow for the issuance of additional notes subordinate in right of payment to the rated notes. Generally, we view such issuance as ratings-neutral.

167. To the extent that an additional issuance of notes is not subordinate to the rated notes, we evaluate the impact on the rated notes. At a minimum, we expect to receive notice of such additional issuance. (Note that our issue ratings are CUSIP/ISIN-specific.) We generally expect to see the following conditions met before additional notes are issued:

• Identical terms to the existing class of notes;
• Same or lower spread or interest rate as the existing class of notes;
• Pro rata issuance unless the additional notes are junior in right of payment;
• Same maturity;
• Issuance may take place only during reinvestment period;
• Proceeds from additional issuance will be applied to the purchase of additional collateral; and
• Maintain or improve OC ratios after giving effect to the additional issuance.

Note cancellation without payment (see paragraph 67 of the corporate CDO criteria) 

168. Occasionally, CDO transaction documents permit cancellation of debt without payment. Cancellation of debt below the most senior tranche affects the payment structure and cash flow mechanics, as well as the level of credit support, in CDO transactions that contain coverage tests (OC tests and/or interest coverage (IC) tests) in their payment waterfall. Generally, the cancellation of subordinate debt without payment according to the payment waterfall makes the coverage tests below the most senior test less sensitive to failing due to defaults or decline in the credit quality of the collateral portfolio. When the subordinate coverage tests are less likely to fail, the most senior notes in transactions with sequential payment priority are less likely to receive principal payments resulting from diversion of interest and/or principal proceeds to cure the test, even though the quality and par amount of the asset pool has not changed.

169. If the transaction documents permit cancellation without payment, our criteria look for the cancelled debt to be counted as outstanding for purposes of calculating the coverage tests and reinvestment target balance. This treatment applies to all debt junior to the most senior notes.

170. If presented with transaction documents that do not prohibit note cancellation without payment and do not count the cancelled debt as outstanding in calculation of the coverage tests, we may apply stresses in our rating analysis that do not give credit to the junior coverage tests.

H. Note events of default (see "Global Framework For Payment Structure And Cash Flow Analysis Of Structured Finance Securities," Dec. 22, 2020)

The use of rating-based haircuts in event of default overcollateralization  

171. According to our criteria, we typically seek to assess whether the likelihood of an event of default (EOD) occurring is commensurate with a given rating scenario. Certain CDO transactions include the breach of coverage tests as an EOD under the terms of the notes. When this happens, we look for specific elements described below, which enables us to assess whether the likelihood the event occurs is commensurate with a rating scenario.

172. A rating-based adjustment of the value of an asset in a CDO--generally to less than its par value--may occur if the asset is downgraded to a specific, predetermined rating level or if certain rating concentrations in the transaction begin to exceed preset amounts. Revaluing assets at less than par causes the OC ratio to drop, which in turn may cause the transaction to fail its EOD OC ratio test.

173. A breach of an EOD OC ratio test often gives the controlling noteholders in a CDO the right to accelerate repayment of the notes. The acceleration of a transaction typically halts interest and principal payments to all but the most senior notes until they are paid in full. Essentially, the transaction turns into a true sequential waterfall payment structure, in which the notes in the transaction are repaid interest and principal in order of priority.

174. A breach of an EOD OC ratio test also typically gives the controlling class of noteholders additional rights, including the sole right to liquidate all of the CDO's collateral. Generally, when transactions are liquidated at the direction of the controlling class, they face market value risk that significantly increases the likelihood that all but the most senior noteholders will suffer losses.

175. Given the potential impact of EOD OC ratio-driven defaults, we will generally rate CDO transactions that have EOD OC ratio tests subject to the following conditions:

• S&P Global Ratings expects the EOD OC ratio test haircuts to be limited to those for defaulted and equity securities. For this purpose, we generally expect defaulted securities to include securities or obligors with a rating of 'CC,' 'D', or 'SD' only, or to be otherwise consistent with our definition of a defaulted obligation. We also expect defaulted securities in the EOD OC ratio test to be carried at their then-current market value.
• We will assess, in our cash flow analysis, whether the likelihood that this EOD OC test is triggered is commensurate with the rating scenario considered for each class of notes. In this case, the controlling class may decide to accelerate note repayment and/or liquidate the assets. We typically see the EOD OC ratio threshold set at 102.5% or lower. To the extent the threshold is set higher (i.e., it is more sensitive to failure), we would likely evaluate the potential impact on the rated notes absent other mitigating factors.
• We review any additional haircuts to EOD OC ratio tests that may be included in future transactions to determine whether the haircut is commensurate with the assigned rating.
• We apply a higher level of scrutiny where we rate a junior 'AAA' class and the senior 'AAA' class has the right to liquidate the transaction based on the EOD OC test failure.

I. Variable funding notes (see the section in the counterparty criteria title Nonderivative Counterparties)

176. Variable-funding notes (VFNs) are issued in certain CDO transactions to counter so-called negative carry arising when the issuer invests some of its note proceeds in revolving or "delayed-draw" loans. These VFNs are typically issued as the most senior class of notes in a sequential pay senior/subordinated capital structure.

177. The risk with such structures arises if the VFN investor is unable to fund the VFN when the borrower in the loan requests a draw.

178. This inability to fund could lead borrowers to make a claim against the SPE or potentially petition the SPE into bankruptcy. As such, the exposure to the VFN investor constitutes counterparty risk.

179. The remoteness of this actually happening lessens lender liability, as a series of events need to occur:

• First, the VFN investor must be unable to fund.
• Then, the collateral manager must be unable to access cash from other sources (for example, cash in the transaction or cash from the sale of assets).
• Next, the borrower must be unable to find alternative sources of funding for his borrowing needs.
• Finally, the borrower must consider that commencing an action against, or petitioning, the SPE is in his best interest despite the uncertainty and significant time delay involved.

180. However, the consequences to the CDO vehicle of the VFN failing to pay could be very material to the rating, in case borrowers petition the SPE into bankruptcy.

181. Therefore, consistent with our counterparty criteria, we treat this nonderivative exposure as limited. Hence the minimum eligible rating on the VFN investor to support a 'AAA' rating on the VFN would be 'A'--or 'A-1' where it has only a short-term rating. Upon the loss of that minimum eligible rating, we expect the transaction documents to include a commitment by the VFN provider to, within 90 days, replace itself, provide a guarantee from an entity carrying the minimum eligible rating, or fund its entire outstanding commitment amount.

182. As with institutional investors, when the holder of the VFN is an ABCP conduit or other structured finance entity with limited sources of liquidity (a "structured conduit investor"), our criteria look to the same minimum eligible rating and downgrade remedies.

183. However, unlike with institutional investors, a key consideration is the ability of the structured conduit investor, as VFN holder, to fund draws upon request, which generally depends on its ability to successfully sell additional commercial paper or otherwise access its sources of liquidity.

184. Structured conduit investors will frequently be supported by an external liquidity facility. However, these facilities are typically only designed for the benefit of the rated investors in the structured conduit investor, not for general creditors of the structured conduit investor, as the CDO issuer would be. In addition, typically, all of the structured credit investor's assets are pledged for the benefit of its rated investors. Accordingly, the structured conduit investor may not have a direct source of liquidity to support the funding requirements under the VFN. To purchase a VFN, a structured conduit investor must ensure that a sufficient source of liquidity is available for the benefit of the CDO issuer.

185. We therefore look for a minimum eligible rating of 'A' (or 'A-1') on the liquidity facility supporting an asset-backed commercial paper VFN investor, to allow for 'AAA' rated VFN notes.

J. Subsidiary special-purpose entities (see "Structured Finance: Asset Isolation And Special-Purpose Entity Methodology," published March 29, 2017)

186. CDO documentation and amendment requests strive to permit CDOs to establish subsidiary SPEs (sub-SPEs) to hold equity securities received as part of a workout or distressed exchange of an underlying defaulted or distressed asset.

187. Our analysis of these sub-SPEs is an application of our published criteria for rating bankruptcy-remote SPEs.

188. The criteria seek to limit the impact of sub-SPEs incurring expenses in excess of any cash flows generated by their underlying assets. We believe the CDO's cash flows are the only source of cash to pay these expenses. To limit any potential cash outflows that might take priority over payments to the rated notes of the CDO, the criteria look for all of the sub-SPEs' expenses to be subject to the administrative expense cap typically found in the CDO's payment waterfall. These expenses include the cost of establishing the sub-SPEs and any ongoing expenses and taxes incurred that are not covered by the cash flows from the sub-SPEs' underlying assets.

189. We believe a CDO could face potential liquidity risks and exposure to third-party liability as a result of the ownership of real property obtained through foreclosure. We view exposure to real property risks and to loans secured primarily by real estate as not appropriate for inclusion in CDOs of corporate debt. Because of this, we would look for express document provisions prohibiting CDO issuers, as well as sub-SPEs, from obtaining title to real property or from obtaining a controlling interest in an entity that owns real property.

190. We do not give any credit to equity securities when we perform our cash flow analysis. We also do not include equity securities when calculating the cash flow CDO's coverage tests. Most CDO transaction documents specify that equity securities are carried at zero in the OC tests. We believe that the equity securities held by the sub-SPEs should also be carried at zero both in our cash flow analysis and when calculating the CDO's OC tests.

191. Our criteria look for any cash received from the disposition of the equity securities held by the sub-SPEs to benefit the CDO by flowing through the payment priority. We believe these proceeds should be treated in the same manner as recoveries on defaulted securities. Under our criteria, recoveries on defaulted securities are treated as principal proceeds until the defaulted securities' full original par amounts are recovered. Accordingly, the criteria consider proceeds from the equity securities held by sub-SPEs to be treated as principal proceeds until the original defaulted securities' full par amounts are recovered in cash.

192. The criteria look for any assets held by the sub-SPEs to be liquidated on or before the CDO's legal final maturity date and paid out to the CDO's investors according to the final payment waterfall in the transaction documents.

K. Combination notes (see paragraph 46 and 75 in the corporate CDO criteria)

193. A combination note is a security that is generally structured by combining two or more different tranches issued by a CDO transaction. It could also be structured by combining one or more rated tranches with equity in a CDO transaction (CDO equity).

194. When a combination note is rated at the time of the transaction's origination, the notional balance of combination notes is often equal to the aggregate of the components, but it may also be higher or lower, depending on the allocation of payment proceeds from the components to the combination notes. For combination notes without a stated coupon, we rate to principal only, as indicated by a 'p' subscript. For our rating purposes in such circumstances, the issued amount outstanding of the combination notes will decline after taking into account paydowns on account of interest distributions, equity distributions, and principal distributions on the underlying components.

195. It is also possible that the notional balance for the combination note may be less than the sum of the principal balance remaining on each of the combination note components. This could happen generally because of the following instances:

• Where the stated interest on the combination notes is lower than the stated interest on the rated note components and where the differential in interest received from the underlying component is applied toward the reduction in the balance of the combination notes.
• Where a CDO equity tranche is one of the components of the combination notes and where the appropriate portion of the equity distribution that goes toward the combination note reduces the outstanding par amount of the combination notes.

196. It is therefore possible that investors in combination notes can potentially be paid out their entire balance while the components still remain outstanding. A likely concern for investors in such a scenario is that their ability to benefit from subsequent residual payments from equity may be capped or limited.

197. To mitigate this risk, underlying documents have often required the notes to be written down when any distributions that were relied upon under the rating scenario are made, but may also allow a nominal amount of the combination notes (usually $1) to remain outstanding until the maturity of the transaction. In doing so, any additional distributions that were not expected under the stressed rating scenario can be paid to combination noteholders as residual upside.

198.Optional redemptions when combination notes include equity securities:   Cash flow CDO transactions may have provisions that allow the CDO, after a specified non-call period, to call the transaction as long as the rated notes can be redeemed in full. However, for those combination notes that are a combination of rated notes and equity notes, there is no guarantee they will be redeemed at par because the underlying documents do not require equity securities to be repaid in full for an optional redemption to occur. Often, in our cash flow analysis, the combination note rating analysis may rely upon many years of cash flow from its constituent parts and, therefore, if the transaction is called early this might mean that not all of the initial investment is received. The redemption price of the subordinated notes may not compensate for the loss of future cash flows.

199. S&P Global Ratings will rate combination securities comprising CDO equity, as long as the combination noteholders are redeemed at par plus accrued interest upon an optional redemption. In application of the criteria, a distribution of the underlying components to the holders is consistent with our approach, and we would withdraw the rating on the combination notes upon such distribution. Full repayment of interest and principal may not be a condition for an optional redemption if the combination noteholders vote in favor of an optional redemption as a separate class from the CDO equity holders.

200.Cash flow assumptions when combination notes include equity securities:  When combination notes include components of CDO equity, S&P Global Ratings needs to be able to assess in its cash flow analysis all payments that must be made in the payment waterfall before the equity payments are disbursed. These payments include any fees, expenses, or other disbursements that do not have a stated amount or rate. Examples include uncapped administrative fees and expenses, hedge termination payments, and subordinate, deferred, and incentive management fees.

201. Hedge termination payments are generally difficult to quantify, given that such payment is a function of the interest rate environment, the strike rates, or the notional amount in the contract. Given the uncertainty around such payments, S&P Global Ratings will assign a rating to the combination notes that is the lower of the rating on the derivative counterparties and the rating commensurate with the cash flow results.

202. To address the uncertainty about uncapped administrative expenses, S&P Global Ratings will use multiples of the administrative expenses capped at the top of the waterfall to stress payments available for equity when assessing cash flows under our rating analysis for combination notes. We assess other expenses such as subordinated, deferred, and incentive management fees based on the caps included in the documentation.

203.Refinancing of the underlying combination securities:  Transactions may include the option to refinance one or more classes of notes at a lower coupon than at issuance. Generally, when classes of notes are refinanced or repriced, the debt is paid off in full at its applicable redemption price. If the offer is accepted by the respective noteholders, their redemption proceeds will be invested into the newly issued note.

204. In a combination note, the refinancing of one or more of the underlying components can change the coupon on the combination note. If we relied on portions of the underlying components' coupons in our cash flow analysis, a decline in the coupons could cause a weakening in the combination note's credit quality. Due to the potential impact on the credit quality of the combination note, we expect that the change in coupon would be voted on and accepted by 100% of the combination noteholders (voting separately from the underlying noteholders). Should the combination noteholders decline to lower the coupon they receive on the underlying note, we expect the combination note to be unwound and the noteholders to be delivered the respective components. If S&P Global Ratings still has a rating outstanding on the combination notes at the time of the refinancing and the underlying components contain subordinated notes, we expect that the combination noteholders would also have the voting right to decide how they receive the equivalent of their subordinated component.

205."Principal-protected" combination securities:   These are securities that are usually backed by CDO notes or certificates issued by a CDO and another type of security. These securities are typically federal, state, or local government-issued bonds, notes, or strips. Due to this feature, the rating assigned to these principal-protected combination notes is typically linked to the rating on the issuer of the supporting security. In order to rate principal-protected combination securities, we expect them to include the following conditions:

• The underlying security matures on or before the legal final maturity date of the combination notes;
• The underlying security has a stated principal amount at maturity that is at least equal to the balance of the combination note that it supports;
• The underlying security is denominated in the same currency as the combination notes, unless the currency risk is otherwise mitigated;
• The supporting security is held as collateral in the CDO transaction until its legal final maturity. If sold, the proceeds from the sale or liquidation of the supporting security are sufficient to repay the principal due and accrued interest on the combo notes and are first used to pay the amounts due on the combination notes in full, before being used otherwise; and
• In certain transactions, the combination noteholders may also elect to receive the supporting security in full satisfaction of the balance of the combination note (physical delivery). When physical delivery is elected, we expect that no additional costs to physically acquire the assets should need to be covered by the noteholders.

K. Turbo Notes (see paragraphs 44-46 in the corporate CDO criteria)

206. A turbo note is a class of secured notes (typically junior) that benefits from accelerated principal repayments from excess interest being diverted away from the subordinated notes to repay their principal amounts, making them more reliant on excess interest. The funds typically available for such principal repayments are the interest proceeds remaining after interest payments to the secured notes, fees, expenses, account deposits, and any other such distributions are made.

207. For paydowns using excess interest to receive credit in our cash flow analysis, we would generally expect that any payments senior to the turbo feature follow the limitations found in paragraphs 200-203 above.

208. Typically, transactions allow for excess interest to be utilized for certain permitted uses, such as to purchase workout related assets, asset exchanges, etc. In order to give credit to excess interest in our cash flow analysis, we would generally expect mitigating factors to be in place (e.g., limitations or small point-in-time buckets).

209. Lastly, we generally expect that any turbo payments would not be considered in the transaction's reinvestment target par or other similar calculations.

210. To mitigate volatility within our ratings (especially at the investment-grade level), we would generally limit the amount of credit given to excess interest in our cash flow analysis (typically, one year). Our analysis will be based, in part, on our forward-looking view of the amount of excess interest that we believe may ultimately be available for such turbo note payments.

211. In addition, our rating on any turbo note would generally not exceed our rating of the non-turbo class of notes immediately senior to it.

APPENDIXES

Appendix A: Asset Type Codes For Use In CDO Evaluator

Table 15

Appendix B: CDO Evaluator Country Codes, Regions, And Recovery Groups

Table 16

REVISIONS AND UPDATES

On Dec. 6, 2019, we updated this guidance document to include updates to the jurisdictional ranking assessments and country grouping contained in table 16. We also made editorial changes to paragraph 147.

On Dec. 8, 2020, we updated this guidance document to provide additional clarity and transparency as to how S&P Global Ratings approaches certain features in U.S. and European CLO transaction documents, and also to address features that have become more common in these documents since we initially published the guidance in June 2019. The changes relate to the items listed below:

• The use of S&P Global Ratings' credit ratings for our CDO Monitor Test and elsewhere, including for debtor-in-possession loans;
• Recovery rate assumptions for first-lien last-out loans;
• Reinvestment of scheduled and unscheduled principal proceeds;
• Treatment of defaulted assets and the use of overcollateralization ratio haircut values;
• What constitutes a "discount obligation" for purposes of determining the carrying value for overcollateralization test purposes; and
• Treatment of workout-related assets.

Further details are available in "CLO Criteria Guidance Updated To Provide Additional Clarity On CLO Document Provisions," published Dec. 8, 2020.

On Feb. 28, 2022, we updated this guidance document to provide additional transparency on how S&P Global Ratings approaches turbo notes and the availability of forward interest rate curves. We also updated references to outdated article titles.

On Oct. 14, 2022, we republished this guidance document to reflect updates to the country recovery groupings of New Zealand (to Group A from Group C) and Italy (to Group A from Group B) in table 16.

On July 21, 2023, we republished this guidance document to update table 15 to reflect updated global industry classification standard (GICS) codes that inform our asset type classification (see Appendix A).

On Oct. 3, 2023, we republished this guidance document to update criteria references.

On Dec. 18, 2023, we republished this guidance document to remove references to mid-market evaluation (MME) ratings from paragraphs 6 and 18 in connection with the retirement of the MME criteria.

On July 26, 2024, we republished this guidance document following the publication of our "Methodology For Determining Ratings-Based Inputs," which was the subject of a request for comment and in connection with which we updated this guidance's "Determining the rating input" section. We also amended this section where it pertains to the treatment of CreditWatch to floor rating inputs resulting from an adjustment on account of a CreditWatch with negative implications.

RELATED PUBLICATIONS

Related criteria

• Methodology For Determining Ratings-Based Inputs, July 26, 2024
• Global Framework For Payment Structure And Cash Flow Analysis Of Structured Finance Securities, Dec. 22, 2020
• Global Methodology And Assumptions For CLOs And Corporate CDOs, June 21, 2019
• Counterparty Risk Framework Methodology And Assumptions, March 8, 2019
• Structured Finance: Asset Isolation And Special-Purpose Entity Methodology, March 29, 2017

Related research

• S&P Global Ratings Definitions, updated from time to time
• Criteria Released To Clarify Method For Determining Ratings-Based Inputs, July 26, 2024
• RFC Process Summary: Methodology For Determining Ratings-Based Inputs, July 26, 2024
• Criteria And Guidance: Understanding The Difference, Dec. 15, 2017