S&P Viewpoint

This step is a fundamental part of Standard & Poor's default and recovery analysis, because we must first understand the forces most likely to cause a default before we can estimate a reasonable level of cash flow at default or value a company. This step draws on the company and sector knowledge of Standard & Poor's credit analysts to formulate and quantify the factors most likely to cause a company to default given its unique business risks and the financial risk inherent in the capital structure that we are evaluating in our default and recovery analysis.

At the outset of this process, we deconstruct the borrower's cash flow projections to understand management's general business, industry, and economic expectations. Once we understand management's view, we make appropriate adjustments to key economic, industry, and firm specific factors to simulate a payment default. While we recognize that there are many possible factors--both foreseen and unforeseen--that could lead to a default, we believe that focusing on the key operating factors that would likely meaningfully contribute to a default is sufficient for purposes of determining the majority of default scenarios.

The simulated default scenario is our assessment of the borrower's most likely path to a payment default. The "insolvency proxy" is the point along that path at which we expect the borrower to default. In other words, the insolvency proxy is the point at which funds available plus free cash flow is insufficient to pay fixed charges:

• (Funds available + Free cash flow) / Fixed charges <= 1.0

The terms in this equation are defined as follows:

• Scheduled principal amortization (We generally do not include "bullet" or "ballooning" maturities as fixed charges, as lenders typically would expect such amounts to be refinanced and would presumably be reluctant to force a company into default that can otherwise comfortably service its fixed charges. Consequently, additional business and cash flow deterioration would be necessary to trigger a default.);
• Required cash interest payments (including assumed increases to LIBOR rates on floating-rate debt and to the margin charged on debt obligations that have maintenance financial covenants); and
• Other cash payments the borrower is either contractually or practically obligated to pay that are not already captured as an expense on the borrower's income statement. (Lease payments, for example, are accounted for within free cash flow and, thus, are not considered a fixed charge.)

The insolvency proxy at the point of projected default may be greater than 1.0x in a few special circumstances:

• For "strategic" bankruptcy filings, when a borrower may attempt to take advantage of the insolvency process primarily to obtain relief from legal claims or onerous contracts;
• When a borrower may rationally be expected to retain a greater amount of cash (e.g., to prepare for a complex, protracted restructuring; if it is in a very capital-intensive industry; if it is in a jurisdiction that does not allow for super-priority standing for new credit in a post-petition financing); and
• When a borrower's financial covenants have deteriorated beyond the level at which even the most patient lender could tolerate further amendments or waivers. (Lenders with no financial covenants have effectively surrendered this option and have reduced their ability to influence company behavior.)

Conversely, the insolvency proxy at the point of default could be lower than 1.0x when the borrower's cash flow is expected to deteriorate further due to cyclicality or business model contraction resulting from the competitive and economic conditions assumed in the simulated default scenario. In any event, our analysis will identify the level of cash flow used as the basis for our valuation.

Valuing a company is a critical component in default and recovery analysis because it is important to start with a reasonable valuation (given our simulated default scenario) before we move to the next step of distributing this value to the various lenders. To help us determine the best valuation for a company, we consider a variety of valuation methodologies, including market multiples, discounted cash flow (DCF) modeling, and discrete asset analysis. The market multiples and DCF methods are used to determine a company's enterprise value as a going concern. This is generally the most appropriate approach when our simulated default and recovery analysis indicates that the borrower's reorganization (or the outright sale of the ongoing business or certain segments) is the most likely outcome of an insolvency proceeding. We use discrete asset valuation most often for industries in which this valuation approach is typically used, or when the simulated default scenario indicates that the borrower's liquidation is the most likely outcome of insolvency. In addition, we may use a combination of the discrete and enterprise valuation methods when we believe that a company will reorganize, but that its debt and organizational structure provides certain creditors with priority claims against particular assets or subsidiaries. For example, Standard & Poor's will consider whether a company's decision to securitize or not securitize material assets impacts the value available to distribute to other creditors.

Ideally, we are interested in multiples for similar firms that have reorganized due to circumstances consistent with our simulated default scenario. In practice, however, the existence of such "emergence" multiple comps are rare. As a result, our analysis often turns to "transaction" or "purchase" multiples for comparable firms because these are generally more numerous. With transaction multiples, we try to use forward multiples (purchase price divided by projected EBITDA) rather than trailing multiples (purchase price divided by historical EBITDA). This is because we believe that forward multiples, which incorporate the benefit of perceived cash flow synergies used to justify the purchase price, provide a more appropriate reference point. In addition, "trading" multiples for publicly traded firms can be useful because they allow us to track how multiples have changed over economic and business cycles. This is especially relevant for cyclical industries and for sectors entering a different stage of development or experiencing changing competitive conditions.

A selection of multiples helps match our valuation with the conditions assumed in our simulated default scenario. For example, a firm projected to default in a cyclical trough may warrant a higher multiple than one expected to default at a cyclical midpoint. Two companies in the same industry may merit meaningfully different multiples if one is highly levered and at risk of default from relatively normal competitive stresses while the other is unlikely to default unless there is a large unexpected fundamental deterioration in the cash flow potential of the business model (which could make historical sector multiples irrelevant).

Our multiples analysis may also consider alternative industry specific multiples--such as subscribers, hospital beds, recurring revenue, etc.--where appropriate. Alternatively, such metrics may serve as a check on the soundness of a valuation that relied on an EBITDA multiple, DCF, or discrete asset approach.

After valuing a company, we must then identify and quantify the debt obligations and other material liabilities that would be expected to have a claim against the company following default. Potential claims fall into three broad categories:

• Principal and accrued interest on all debt outstanding at the point of default, whether issued at the operating company, subsidiary, or holding company level;
• Bankruptcy-related claims, such as debtor-in-possession (DIP) financing and administrative expenses for professional fees and other bankruptcy costs;
• Other nondebt claims, such as taxes payable, certain securitization programs, trade payables, deficiency claims on rejected leases, litigation liabilities, and unfunded post-retirement obligations.

Our analysis of these claims and their potential values takes into consideration each borrower's particular facts and circumstances, as well as the expected impact on the claims as a result of our simulated default scenario.

We estimate debt outstanding at the point of default by reducing term loans by scheduled amortization paid prior to our simulated default and by assuming that all committed debt, such as revolving credit facilities and delayed draw term loans, is fully funded. For asset-based lending (ABL) facilities, we will consider whether the borrowing base formula would allow the company to fully draw the facility in a simulated default scenario. For letters of credit, especially those issued under dedicated synthetic letter of credit tranches, we will assess whether these contingent obligations are likely to be drawn following default. Our estimate of debt outstanding at default also includes an estimate of pre-petition interest, which is calculated by adding six months of interest (based on historical data from Standard & Poor's LossStats® database) to our estimated principal amount at default. The inclusion of pre-petition interest makes our recovery analysis more consistent with banks' credit risk capital requirements under the Basel II Framework.

Our analysis focuses on the recovery prospects for the debt instruments in a company's current or pro forma debt structure, and generally does not make estimates for other debt that may be issued prior to a default. We feel that this approach is prudent and more relevant to investors because the amount and composition of any additional debt (secured, unsecured, and/or subordinated) may materially impact lender recovery rates, and it is not possible to know these particulars in advance. Further, incremental debt added to a company's capital structure may materially affect its probability of default, which, in turn, could impact all aspects of our recovery analysis (i.e., the most likely path to default, valuation given default, and loss given default). Consequently, changes to a company's debt structure are treated as events that require a reevaluation of our default and recovery analysis. This is a key aspect of our ongoing surveillance of our default and recovery ratings. We do, however, make a few exceptions to this approach. Such exceptions will be outlined in our recovery reports and fall under three categories:

• Permitted, but uncommitted, incremental debt may be included as part of our default and recovery analysis if this is consistent with our expectations and our underlying corporate credit rating on a given issuer.
• Revolving credit facilities with near-term maturities will generally be treated as if they are rolled over with similar terms.
• Our default and recovery analysis may assume the repayment of near-term debt maturities if the company is expected to retire these obligations and has the liquidity to do so. Similarly, principal prepayments--whether voluntary or part of an excess cash flow sweep provision--may be considered for certain credits when deemed appropriate.

Our analytical treatment and estimates for bankruptcy-related and other nondebt claims in default is generally specific to the laws and customs of the jurisdictions involved in our simulated default scenario. Please refer to Appendix 1 for a review of our approach and methodology for these claims in the U.S.

After determining the best valuation for a firm and estimating the types and amounts of claims in default, we then consider the distribution of that value. This distribution will follow a "waterfall" approach that will reflect the relative seniority of the claimants and will be specific to the laws, customs, and insolvency regime practices for the relevant jurisdictions for a company. For example, the quantification and classification of bankruptcy-related and nondebt claims for insolvencies outside of the U.S. might be very different than the methodology for U.S. Industrials companies discussed in Appendix 1. Further, local laws and customs may warrant deviations from the waterfall distribution we follow in the U.S. Where relevant, we will publish our guidelines and rationale for these differences before rolling out our unsecured recovery ratings in these jurisdictions. In the U.S., our general assumption of the relative priority of claimants is as follows:

• Super-priority claims, such as DIP financing
• Administrative expenses
• Federal and state tax claims
• Senior secured claims
• Junior secured claims
• Senior unsecured debt and nondebt claims
• Subordinated claims
• Preferred stock
• Common stock

However, this priority of claims is subject to two critical caveats:

• The beneficial position of secured creditor claims, whether first-priority or otherwise, is only valid to the extent that the collateral supporting such claims is equal to, or greater than, the amount of the claim. If the collateral value is insufficient to fully cover a secured claim, then the uncovered amount or "deficiency balance" will be pari passu with all other senior unsecured claims.
• Structural issues may alter the priority of certain claims against specific assets or entities in an organization based on the company's legal entity structure and the relevant terms and conditions of the debt instruments.

As a result of these caveats, the recovery prospects for different debt instruments of the same type (whether they be senior secured, senior unsecured, senior subordinated, etc.) might be very different, depending on the structure of the transactions. While the debt type of an instrument may provide some indication as to its relative seniority, it is the legal structure and associated terms and conditions that are the ultimate arbiter of priority. Consequently, a fundamental review of a company's debt and legal entity structure is required to properly evaluate the relative priority of claimants. This requires an understanding of the terms and conditions of the various debt instruments as they pertain to borrower and guarantor relationships, collateral pledges and exclusions, facility amounts, covenants, and debt maturities. In addition, we must understand the breakout of the company's cash flow and assets as it pertains to its legal organizational structure, and consider the effect of key jurisdictional and intercreditor issues.

Key structural issues to explore include identifying:

• Higher priority liens on specific assets by forms of secured debt such as mortgages, industrial revenue bonds, and ABL facilities;
• Non-guarantor subsidiaries (domestic or foreign) that do not guarantee a company's primary debt obligations or provide asset pledges to support the company's secured debt;
• Claims at non-guarantor subsidiaries that will have a higher priority (i.e., a "structurally superior") claim on the value related to such entities;
• Material exclusions to the collateral pledged to secured lenders, including the lack of asset pledges by foreign subsidiaries or the absence of liens on significant domestic assets, including the stock of foreign or domestic non-guarantor subsidiaries (whether due to concessions demanded by and granted to the borrower, poor transaction structuring, regulatory restrictions, or limitations imposed by other debt indentures); and
• Whether a company's foreign subsidiaries are likely to file for bankruptcy in their local jurisdictions as part of the default and restructuring process.

The presence of obligations with higher priority liens on certain assets means that the enterprise value available to other creditors must be reduced to account for the distribution of value to satisfy these creditors first. In most instances, asset-specific secured debt claims (such as those previously listed) are structured to ensure full collateral coverage even in a default scenario. As such, our analysis will typically reduce the enterprise value by the amount of these claims to determine the remaining enterprise value available for other creditors. That said, there may be exceptions that will be considered on a case-by-case basis if the amounts are material. Well-structured secured bank or bond debt that does not have a first-lien on certain assets will get second-priority liens on assets that are significant and may have meaningful excess collateral value. For example, this is often the case when secured debt collateralized by a first lien on all noncurrent assets also takes a second-priority lien on working capital assets that are already pledged to support an asset-based revolving credit facility.

Significant domestic or foreign non-guarantor entities must be identified because these entities have not explicitly promised to repay the debt. Thus, the portion of enterprise value derived from these subsidiaries does not directly support the rated debt. As a result, debt and certain nondebt claims at these subsidiaries have a structurally higher priority claim against the subsidiary value. Accordingly, the portion of the company's enterprise value stemming from these subsidiaries must be estimated and treated separately in the distribution of value to creditors. This requires an understanding of the breakout of a company's cash flow and assets. Because these subsidiaries are still part of the enterprise being evaluated, any equity value that remains after satisfying the structurally superior claims would be available to satisfy other creditors of the entities that own these subsidiaries. Well-structured debt will often include covenants to restrict the amount of structurally superior debt that can be placed at such subsidiaries. Further, well-structured secured debt will take a lien on the stock of such subsidiaries to ensure a priority interest in the equity value available to support other creditors. In practice, the pledge of foreign subsidiary stock owned by U.S. entities is usually limited to 65% of voting stock for tax reasons. The residual value that is not captured by secured lenders through stock pledges would be expected to be available to all senior unsecured creditors on a pro rata basis.

The exclusion of other material assets (other than whole subsidiaries or subsidiary stock) from the collateral pledged to support secured debt must also be incorporated into our analysis. The value of such assets is typically determined using a discrete asset valuation approach, and our estimated value and related assumptions will be disclosed in our recovery report as appropriate. We expect the value of excluded assets would be shared by all senior unsecured creditors on a pro rata basis.

An evaluation of whether foreign subsidiaries would also be likely to file for bankruptcy is also required, because this would likely increase the cost of the bankruptcy process and create potential multijurisdictional issues that could impact lender recovery rates. The involvement of foreign courts in a bankruptcy process presents a myriad of complexities and uncertainties. For these same reasons, however, U.S.-domiciled borrowers that file for bankruptcy seldom also file their foreign subsidiaries without a specific benefit or reason for doing so. Consequently, we generally assume that foreign subsidiaries of U.S. borrowers do not file for bankruptcy unless there is a compelling reason to assume otherwise, such as a large amount of foreign debt that needs to be restructured to enable the company to emerge from bankruptcy. When foreign subsidiaries are expected to file bankruptcy, our analysis will be tailored to incorporate the particulars of the relevant bankruptcy regimes.

Intercreditor issues may affect the distribution of value and result in deviations from "absolute priority" (i.e., maintenance of the relative priority of the claims, subject to structural considerations, so that a class of claims will not receive any distribution until all classes above it are fully satisfied), which is assumed by Standard & Poor's methodology. In practice, however, the distribution of value may vary somewhat from this ideal for a variety of intercreditor reasons, including, in the U.S., "accommodations" and "substantive consolidation."

Accommodations refer to concessions granted by senior creditors to junior claimants in negotiations to gain their cooperation in a timely restructuring. We generally do not explicitly model for accommodations because it is uncertain whether any concessions will be granted, if those granted will ultimately have value (e.g., warrants as a contingent equity claim), or whether the value will be material enough to meaningfully affect our projected recovery rates.

Substantive consolidation represents a potentially more meaningful deviation from the distribution of value according to absolute priority. In a substantive consolidation, the entities of a corporate group may be treated as a single consolidated entity for the purposes of a bankruptcy reorganization. This effectively would eliminate the credit support provided by unsecured guarantees or the pledge of intercompany loans or subsidiary stock, and dilutes the recovery prospects of creditors that relied on these features to the benefit of those that did not. Even the threat of substantive consolidation may result in a negotiated settlement that could affect recovery distribution. While substantive consolidation can meaningfully impact the recovery prospects of certain creditors, it is a discretionary judicial doctrine that is only relevant in certain situations. It is difficult to predict whether any party would seek to ask a bankruptcy court to apply it in a specific case, or the likelihood that party would succeed in persuading the court to do so. As such, our analysis does not evaluate the likelihood of substantive consolidation, though we acknowledge that this risk could affect recoveries in certain cases.

We estimate recovery rates by dividing the portion of enterprise or liquidation value projected to be available to cover the debt to which the recovery rating applies, by the estimated amount of debt (principal and pre-petition interest) and pari passu claims outstanding at default. We then map the recovery rate to our recovery rating chart to determine the issue and recovery ratings. Standard & Poor's accompanies its recovery ratings with written recovery reports, which identify the simulated payment default, valuation assumptions, and other factors on which the recovery ratings are based.

After our initial analysis at debt origination, we then monitor material changes affecting the borrower and its debt and liability structure to determine if the changes might also alter creditor recovery prospects. Because of the dynamic and evolving nature of credit--particularly recovery--our recovery analysis at origination is unlikely to identify all of the actual claims at bankruptcy, or precisely predict the value of the company or the collateral given a default. Therefore, a fundamental component of recovery analysis is periodic and event specific surveillance designed to monitor developing risk exposures that might affect recovery. Any material changes to our default and recovery ratings or analysis will be disclosed in updates to our recovery reports. Factors that could impact our default and recovery analysis or ratings include:

• Acquisitions and divestitures;
• Updated valuation assumptions;
• Shifts in the profit and cash flow contributions of borrower, guarantor, or non-guarantor entities;
• Changes in debt or the exposure to nondebt liabilities;
• Intercreditor dynamics; and
• Changes in bankruptcy law or case histories.

• Three percent for capital structures with one primary class of debt;
• Five percent for two primary classes of debt (first- and second-lien creditors may be adversaries in a bankruptcy proceeding and are treated as separate classes by Standard & Poor's);
• Seven percent for three primary classes of debt; and
• Ten percent for certain complex capital structures.

When using a discrete asset valuation approach, these costs are implicitly accounted for in the orderly liquidation value discounts used to value a company's assets.

After determining whether it is reasonable to include such claims in our default and recovery analysis, we are left with the challenge of sizing the claims and determining how they might impact creditor recovery prospects. Unfortunately, the case history is very limited in this area and does not offer clear guidelines on how to best handle these inherent uncertainties. As such, we tailor our approach on a case-by-case basis to the borrower's specific circumstances to help us reach the best solution. When significant, our approach and assumptions will be outlined in our recovery report so that investors can evaluate our treatment, and consider alternative assumptions if desired, as part of their investment decision. We note that claims in this category would typically be expected to have general unsecured status in a bankruptcy.

Conversely, a future flow type of securitization financing, which securitizes all or a portion of the borrower's future revenue and cash flow (typically related to particular contracts, patents, trademarks, or other intangible assets), would have a claim against our estimated valuation. Such transactions effectively securitize all or a part of the borrower's enterprise value, and the related claims would have priority claim to the value stemming from the securitized assets. This claim would diminish the enterprise value available to other corporate creditors. Such transactions are typically highly individualized, and the amount of the claims and the value of the assets in our simulated default analysis are evaluated on a case-by-case basis.

If a lease is rejected, the company must discontinue using the asset, and the lessor may file a general unsecured claim against the estate. As a result, we must estimate a reasonable lease rejection rate for the firm given the types of assets leased, the industry, and our simulated default scenario. Leases are typically rejected for one of three reasons:

• The lease is priced above market rates;
• The leased asset is generating negative or insufficient returns; or
• The leased asset is highly vulnerable to obsolescence during the term of the lease.

Our evaluation may ballpark the rejection rate by assuming it matches the percentage decline in revenue in our simulated default scenario or, if applicable, by looking at common industry lease rejection rates. If leases are material, we may further evaluate whether our knowledge of a company's portfolio of leased assets is likely to result in a higher or lower level of unattractive leases (and rejections) in a default scenario. For example, if a company's leased assets are unusually old, underutilized, or priced above current market rates, then a higher rejection rate may be warranted. In practice, this level of refinement in our analysis will be most relevant when a company has a substantial amount of lease obligations and a significant risk of near-term default. Uneconomical leases that are amended through renegotiation in bankruptcy are considered to be rejected.

In bankruptcy, the amount of unsecured claims from rejected leases is determined by taking the amount of lost rental income and subtracting the net value available to the lessor by selling or re-leasing the asset in its next best use. However, the deficiency claims of commercial real estate lessors is further restricted to the greater of one year's rent or 15% of the remaining rental payments not to exceed three years' rent. Lessors of assets other than commercial real property do not have their potential deficiency claims capped, but such leases are generally not material and are usually for relatively short-periods of time. With these issues in mind, Standard & Poor's quantifies lease deficiency claims for most companies by multiplying their estimated lease rejection rate by three times their annual rent.

However, there are a few exceptions to our general approach. Deficiency claims for leases of major transportation equipment (e.g., aircraft, railcars, and ships) are estimated on a case-by-case basis, with our assumptions disclosed in our recovery reports. This is necessary because these lease obligations do not have their claims capped, may be longer term, and are typically for substantial amounts. In addition, we use a lower rent multiple for cases in which a company relies primarily on very short-term leases (three years or less). Further, we do not include any deficiency claim for leases held by individual asset-specific subsidiaries that do not have credit support from other entities (by virtue of guarantees or co-lessee relationships) due to the lack of recourse against other entities and the likelihood that these subsidiaries are likely to be worthless if the leases are rejected. This situation was relevant in many of the movie exhibitor bankruptcies in the early 2000 time period.

COLLECTIVE BARGAINING AGREEMENT REJECTION CLAIMS: A borrower that has collective bargaining agreements (CBA), including above-market wages, benefits, or work rules, is likely to seek to reject these contracts in a bankruptcy. In order to reject a CBA, the borrower must establish, and the bankruptcy court must find that the borrower has proposed, modifications to the CBA that are necessary for its successful reorganization. In addition, the court must find that all creditors and affected parties are treated fairly and equitably, that the borrower has bargained fairly with the relevant union, that the union rejected the proposal without good cause, and that equity considerations clearly favor rejection. Proceedings to reject a CBA typically result in a consensual reduction in wages and benefits, and modified work rules under a replacement or modified agreement prior to the bankruptcy court's decision on the motion to reject.

If a CBA were rejected, the affected employees would have unsecured claims for damages that would be limited to one year's compensation plus any unpaid compensation due under the CBA. However, if a CBA were modified through negotiation without rejection, the damages for lost wages and benefits and modified work rules would not be limited to this amount.

PENSION PLAN TERMINATION CLAIMS: The ability to terminate a defined benefit pension plan is provided under the U.S. Employee Retirement Income Security Act (ERISA). Under ERISA, these plans may be terminated voluntarily by the debtor as the plan sponsor, or involuntarily by the Pension Benefit Guaranty Corporation (PBGC) as the agency that insures plan benefits. Typically, any termination during bankruptcy will be a "distress termination," in which the plan assets are, or would be, insufficient to pay benefits under the plan. However, the bankruptcy of the plan sponsor does not automatically result in the termination of its pension plans, and even underfunded plans may not necessarily be terminated. For example, a borrower may elect to maintain underfunded plans, or may not succeed in terminating a plan, if it fails to demonstrate that it would not be able to pay its debts and successfully reorganize unless the plan is terminated.

In a distress termination, the PBGC assumes the liabilities of the pension plan up to the limits prescribed under ERISA and gets an unsecured claim in bankruptcy against the debtor for the unfunded benefits. The calculation of this liability is based on different assumptions than the borrower's reported liability in its financial statements. This, in addition to the difficulty of predicting the funded status of a plan at some point in the future, complicates our ability to accurately assess the value of these claims.

RETIREE BENEFITS MODIFICATION CLAIMS: Non-pension retiree benefits are payments to retirees for medical, surgical, or hospital care benefits, or benefits in the event of sickness, accident, disability, or death. The requirements for modifying these benefits for plans covered under a union contract during bankruptcy are similar to the requirements for the rejection of a CBA, but they may be modified by order of the bankruptcy court without rejecting the plan or program under which the benefits are provided in its entirety. However, these obligations are often amended prior to bankruptcy for companies that are placed at a competitive disadvantage because of these costs. As such, we must consider whether the borrower has modified, or is likely to modify, the benefits prior to bankruptcy.

In the case of benefits provided to employees that were not represented by unions, the borrower may be able to revise the benefits prior to bankruptcy with little or no negotiation with the retirees. For union retirees, benefit modifications prior to bankruptcy likely would occur in the context of concessions in negotiations with the relevant union. In either case, modifications prior to bankruptcy would not result in claims in bankruptcy that could dilute recoveries. If the borrower reduces its retiree benefits liability prior to bankruptcy, further modifications in bankruptcy may result in a smaller unsecured claim than if it had entered the proceeding with a greater liability. If we conclude that the borrower will modify its retiree benefits prior to bankruptcy, our recovery analysis will consider the likely effect of that modification on the borrower's reduced benefit liability in bankruptcy. Conversely, if we conclude that these plans will be modified in bankruptcy, but not before, then the potential liability will be more significant.