In the aftermath of the 2020 pandemic-driven economic downturn and amidst current turbulent market conditions, including concerns about high inflation, rising interest rates, and slowing economic growth, U.S. collateralized loan obligation (CLO) indenture provisions continue to evolve. A number of the recent developments in indenture provisions are geared towards increasing CLO manager flexibility, while others are responses to market and regulatory changes. Based on conversations S&P Global Ratings have had with investors across the CLO capital stack and with the document analysts on our CLO rating team, we're doing a deep dive on these key changes and how they could potentially affect noteholders in two separate groups: first, indenture provisions governing asset classification and trading flexibility, and second, other topics that have varying reasons for introduction.

CLO indentures are typically heavily negotiated documents that reflect the incentives of different participants in a given transaction, in particular, the prospective CLO 'AAA' noteholder, CLO equity holder, and CLO collateral manager. As such, U.S. CLO transaction document provisions have continued to develop over time, and we have periodically reviewed the changes we have observed in our previous articles, "Par Wars: The Investor Strikes Back," published May 2, 2018; "Par Wars: Return Of The Covenants?" published May 17, 2019; "Par Wars: Short Debrief On U.S. CLO Document Trends," published Sept. 29, 2019; and "Par Wars: The Phantom Limits" and "Are There Holes In The Buckets (And Could They Lead To A Delay In Coverage Test Failures)?" both published Feb. 21, 2020, just prior to the pandemic-related downturn.

Indenture Provisions Governing Asset Classification And Trading Flexibility

Expansion of workout loan language

Workout-related assets are typically loans or bonds issued in connection with the workout or restructuring of a distressed company's debt. Under many CLO indentures, these assets likely do not meet the definition of collateral obligation in full but can still be purchased, a construct that has become common in CLO transaction documents since the pandemic. In most CLO indentures, there are two categories of workout-related assets: one, debt obligations that are senior or pari passu to an existing debt obligation in the portfolio and which meet most of the collateral obligation requirements outside of a small number of carve-outs (workout loans); and two, debt obligations that don't necessarily have to be senior or pari passu to an existing debt obligation and/or which further deviate from the collateral obligation definition (restructured loans). A carve-out to any clause of the collateral obligation definition is the allowance of such clause to be inapplicable for a certain type of asset. Typically, workout loans have been carried as defaulted obligations, while restructured loans have been carried at zero.

Since the introduction of language around workout-related assets in CLO documents and their widespread inclusion after the 2020 pandemic-driven economic downturn, the assets covered in these definitions have expanded. When the workout loan concept was first introduced to CLO documents, most transactions would only grant full par credit when the asset fully met the definition of collateral obligation without the benefit of any carve-outs. More recently, there has been a growing trend to attempt to carve-out restructured loans from certain requirements of the collateral obligation definition without also stipulating that they cannot receive full par credit until meeting the definition of collateral obligation in full without the benefit of any carve-outs.

For example, in some recent documents we have reviewed, we have seen clauses that would grant full par credit for restructured loans that are long-dated obligations (mature after the earliest stated maturity of the secured notes), have a S&P Global Ratings' credit rating below 'CCC-', or had a purchase price below a certain percentage of par (typically, 60%). This concept would provide additional flexibility on the credit treatment for restructured loans, particularly given that these loans are typically not required to be senior or pari passu to the existing debt of the distressed obligor held by the CLO.

In the past, carve-out provisions had typically been restricted to certain bankruptcy exchanges where the manager can exchange one asset from a distressed (or defaulted) obligor for another. The manager would generally be limited to acquiring an asset that fully met the definition of collateral obligation other than the fact that it was a defaulted or credit risk obligation, which managers are normally prohibited from purchasing. Newer document terms would allow a manager to receive non-performing assets in a distressed exchange scenario so long as (in their view) it was necessary to realize a higher recovery on the distressed asset. As demonstrated above, this concept has been broadened through the introduction of workout loan language.

New classifications of workout asset types

After the advent of workout asset provisions in transaction documents, they have continued to evolve with the introduction of new concepts/terms, including uptier priming loans and asset drop-downs. In the corporate loan market, uptier priming loans are new obligations that are senior in priority to existing debt of an obligor. They may be purchased with new money, a cashless exchange (or roll-up) of existing debt into the senior position, or a combination. Asset drop-downs are new obligations backed by collateral that was moved away from existing lenders to an "unrestricted" subsidiary. In either case, the existing lenders or non-participating lenders are put at a disadvantage as their position sees its recovery prospects diminished.

Generally, both circumstances occur in connection with a bankruptcy, workout, or restructuring of a corporate loan issuer. However, after our review of numerous transaction documents containing these concepts, we've seen numerous approaches to dealing with these occurrences. At times, we have seen such concepts be nothing more than a further bifurcation of the standard workout terms as well as a further breakout of the transaction's concentration limitations; and at other times, we have seen such concepts being included to create additional workout asset types that have their own separate purchase and recoupment requirements. S&P Global Ratings considers all such concepts akin to workout or restructuring obligations and would look for certain mitigants to be in place on such assets.

Segregation of recovery proceeds from equity securities during refinancings

Recently, there has been a growing trend to attempt to segregate certain specifically identified securities in a transaction's portfolio (typically, equity securities), and classify the proceeds from the sale of these assets as interest proceeds upon the completion of either a refinancing or reset of the CLO transaction. Typically, equity securities are received by a CLO in connection with a defaulted or distressed obligor and are given zero credit throughout the transaction document.

In transactions where this segregation feature is not present, proceeds from the sale of equity securities would generally be classified as principal proceeds until the full par amount of the initially held defaulted obligation has been recovered. However, we have seen transaction documents that allow sale proceeds of some specifically identified equity securities to be paid out to the CLO equity holder after a refinancing or reset of the CLO has taken place.

There is a difference in how this concept is used in a reset of the CLO notes versus a refinancing. In a reset, all the previous secured noteholders are required to be repaid full par plus accrued and unpaid interest, and there is typically a requirement for the transaction to achieve a target par threshold again. After the target par requirement has been met, non-performing assets (including equity) above this threshold may be treated as excess assets. However, in the proposals we've seen under a refinancing scenario, there may not be a requirement to re-achieve a target par threshold (and therefore no excess assets concept).

As long as there is a requirement for the proceeds of such excess assets to stay in the transaction, such proceeds would still flow down either the interest proceeds or principal proceeds waterfalls, according to the terms of the transaction documents. If the deal were failing coverage tests, the proceeds would likely not make it to the equity holders anyway. However, the opposite is also true, if the coverage tests were in compliance, the proceeds would likely flow down the waterfall and find their way to equity holders.

Growing proliferation of exchanges

CLOs typically have reinvestment criteria contained in section 12 of the indenture that are intended to govern which assets a manager can purchase or sell. These requirements put guardrails in place when it comes to investing any cash proceeds both during and after the transaction's reinvestment period. Historically, one typical requirement is that if you are going to reinvest proceeds from a defaulted collateral obligation, the new asset purchased with those proceeds must be a performing (i.e., non-defaulted) asset.

Asset exchange concepts provide carveouts to these requirements. While some asset exchange concepts have been a longstanding part of U.S. CLO documents, we have recently observed reinvestment criteria carve-outs that are relatively new. For several years prior to 2020, most U.S. CLO indentures contained only one type of exchange that provided a carve out to the reinvestment criteria. Then, with the arrival of the pandemic and associated economic turmoil, collateral managers started inserting more exchange concepts into the documents intended to provide more flexibility to manage the portfolio during what they expected to be a potentially extended period of economic stress.

Initially, the most prominent exchange type were bankruptcy exchanges, which are an exchange of a defaulted obligation for another defaulted obligation or credit risk obligation from the same (or affiliated) obligor. These exchanges typically all had similar requirements, including the expectation for a higher recovery amount and that the new obligation received could not be more junior than the existing asset. Now, most transactions have multiple exchange concepts. Examples of more common exchanges today are a defaulted obligation for another defaulted obligation of a different obligor, a credit risk obligation for another credit risk obligation of the same or different obligor, and exchanges that utilize additional proceeds (typically, excess interest proceeds and/or contributions) to affect the exchange. There are typically no requirements for par maintenance, and/or the requirement that the new obligation has the same or earlier maturity (except in the case of a credit risk obligation being exchanged for another credit risk obligation).

The introduction of several exchange concepts is a double-edged sword. On the one hand, such exchanges make it easier for collateral managers to swap out of troubled assets. This allows the collateral managers to demonstrate both their acumen and their value-add proposition in comparison to other collateral managers. If the exchange is successful, the deal could be better off as more proceeds are recovered and less par is lost. However, since there are fewer credit-related requirements to such exchanges, if the exchange is not successful, the deal could become worse off as less proceeds are recovered and more par loss is incurred.

Change in post-reinvestment period trading requirements

Actively managed CLO transactions have provisions governing purchases and sales that typically differ during the CLO's reinvestment period and after. Most transactions require that trades done after the reinvestment period has concluded maintain the credit quality of the portfolio, a requirement that can be met in several different ways. One means to assess post-reinvestment period trades is to require the new obligation to have the same or better S&P Global Ratings' credit rating than the sold obligation. Another would be to compare the output of S&P's CDO Evaluator model for the overall portfolio both before and after the proposed trade and determine that the overall credit quality of the portfolio would not decline if the trade were to take place.

In the past, where the requirement was based directly on the ratings on the assets being sold and purchased, this was compared on an asset-by-asset basis. If the old obligation had an S&P Global Ratings' credit rating of 'B-', then the new obligation would be required to have an S&P Global Ratings' credit rating of 'B-' or higher and so forth. In more recent documents, we have seen proposals to compare the weighted average S&P Global Ratings' credit rating across all of the newly purchased obligations against the weighted average S&P Global Ratings' credit ratings of all the sold obligations. At first glance, this seems straightforward enough, but issues can arise if the transaction documents do not specify how such average should be calculated.

For example, the use of a linear scale to capture the average credit quality (i.e., starting at 1 for 'AAA'-rated assets and cascading linearly down to 19 for 'CCC-'-rated assets) is likely easier for the collateral manager to implement while trading as well as it being considered a somewhat approximation of credit risk. However, when the same or better S&P Global Ratings' credit rating requirement is gauged by a linear scale, the credit risk might not be fully accounted for. Credit risk is not linear especially at lower, non-investment-grade rating categories (i.e., 'BB-' and below). This is evidenced by S&P Global Ratings' Rating Factor scale that is utilized when running the non-model version of CDO Monitor.

The progression of discount obligations

Discount obligations are generally obligations that are purchased at a significant discount to par (historically, 80% or lower), which is typically an indication of some type of distress. Distress in this context could be obligor specific (e.g., deteriorating financials, poor operating results, etc.) or on a more macro-scale, such as a recession, slower economic growth, etc. In either case, the obligation is trading at levels that indicate some type of trouble may be ahead for the issuing obligor. To account for the distress, discount obligations are often held at their purchase price in the transaction's overcollateralization (O/C) ratio calculations.

Since we've been publishing our "Par War" series of articles, the definitions for discount obligations and swapped non-discount obligations (discussed in the following section) have evolved. Some of these changes include percentage-of-par thresholds when determining what is a discount obligation, differing thresholds based on seniority levels, and the introduction of new thresholds involving the prices of indices.

Previously, most transaction documents specified a single threshold for determining whether an asset was a discount obligation, with anything trading at less than 80% of par typically being considered a discount obligation and being carried at its purchase price rather than par until certain conditions were met. Now, we are seeing this percentage fluctuate between 75% and 85%. In addition, instead of having one threshold apply to all assets, there is now often a bifurcation of the threshold by asset seniority. Typically, the threshold associated with more senior assets (i.e., senior secured loans) will be set higher than the threshold associated with less senior or subordinated assets (i.e., second-lien loans, first-lien last-out loans, etc.).

Additionally, some transaction documents also include the ability to utilize the price of certain leveraged loan or bond indices rather than a fixed price threshold. The price of the index is usually multiplied by a certain percentage (for example, 90% multiplied by the Morningstar LSTA Leverage Loan Price Index). If the asset's purchase price is below 90% of the leveraged loan index price, then that asset would be considered a discount obligation and vice versa. This allows assets that are negative outliers during a broader market dislocation to be treated differently from assets that are trading near the market average price. This language became much more common in CLO documents after the early stages of the COVID-19 pandemic, when a substantial portion of the loan market was trading at levels below 80% of par.

When a static threshold like 80% of par is utilized, the manager may feel they are being penalized by having to carry a credit at its purchase price because they bought it at 79% of par (for example) even though the entire market (i.e., Morningstar LSTA Leverage Loan Price Index) is trading at this level, highlighting that 79% of par may not indicate distress for particular credit.

A related set of changes involves how transaction documents "cure" a discount obligation (i.e., allow it to be carried at par once certain conditions have been met). We have seen proposals to utilize the Morningstar LSTA Leverage Loan Price Index multiplied by, for example, 90% for 30 consecutive days to give a once discounted obligation full par credit in the O/C test numerator. If the market is now trading at par (i.e., 100%), and if for 30 consecutive days the price of such asset was greater than 90% of the Morningstar LSTA Leverage Loan Price Index, the asset would now be able to shed its discounted obligation classification and receive full par credit in the O/C numerator.

The evolution of swapped non-discount obligations

Related to the discount purchase obligations discussed above, swapped non-discount obligations are indenture provisions that have evolved to allow managers flexibility to manage assets during periods of economic dislocation. If an asset is purchased at a price above the discount purchase obligation threshold (and therefore granted full par credit in the O/C test calculation), it may later trade at a price below the discount purchase obligation threshold for a variety of reasons. In these cases, the manager may be disincentivized from trading the asset for another asset trading at the same price (or even higher) if the new asset would be treated as a discount purchase obligation and carried in the O/C test at a discount to par. Swapped non-discount obligation provisions allow new assets acquired at prices that would otherwise classify them as discount obligations to be carried at full par instead of the purchase price, as long as the asset is being acquired in connection with the sale of an existing asset in the portfolio that is trading at a discount.

Additionally, there are some guardrails in place on this concept. Typically, there are several requirements to classify an asset as a swapped non-discount obligation, which include but are not limited to: the purchase price of the new asset be higher than that of the sold asset, the purchase price is at or above a minimum threshold, the new asset must have the same or better rating than the disposed asset, and the new asset must be purchased in relatively short order after the sale of the existing asset.

These provisions may allow collateral managers the flexibility to swap out of troubled assets and improve the collateral credit quality without artificial constraints. This concept allows the collateral manager flexibility to avoid loss of par in the scenario when selling an old asset (which, unless a haircut applied, would still be carried at full par), only to purchase a new asset that would now receive a lower value (haircut) in the O/C test calculation, since the new asset would have been purchased at a price below 80% of par.

However, we have also heard investors make the opposite argument, with some CLO senior noteholders simply preferring the O/C tests to fail (and senior notes to pay down) during periods of economic stress when the loan market is trading at a significant discount to par.

Other Topics

Mezzanine CLO note turbo features and delayed draws

Over the past year or so, we've noticed new structural features in mezzanine tranches of some new issue CLO transactions we have rated, including delayed-draw tranches and 'BBB' and/or 'BB' tranches that include a turbo feature. These structural features can reduce the weighted average cost of debt for a CLO and/or increase projected equity returns by repaying costlier tranches first and/or issuing these costlier tranches when CLO market spreads are lower.

The parameters for delayed-draw tranches are built into the CLO indenture from day one. For instance, the spread on the delay-draw tranche is typically variable but capped at a specified amount. If market conditions positively evolve and CLO mezzanine tranche spreads drop, the CLO can issue and price that tranche at a lower cost later and generate additional excess spread. There are some aspects to consider when reviewing delayed-draw tranches, including, among others:

• The timing of the additional CLO note issuance (for example, there is a question if only a one-time issuance is allowed or if additional notes can be issued over months or years);
• How the CLO coverage tests that include that tranche are calculated; and
• How coverage tests junior to a delayed-draw tranche could be affected by the issuance of additional notes under the delayed-draw mechanism.

There may also be counterparty risk considerations to take into account based on who is responsible for funding the additional note issuance.

Another innovation that has appeared are mezzanine CLO tranches that include a turbo feature where some portion of the CLO's excess spread proceeds (interest proceeds) are used to pay down the tranche balance rather than being distributed to equity. Because these tranches often represent a larger proportion of the CLO capital structure than a typical 'BBB' or 'BB' tranche, they are sometimes referred to as "thick" tranches. Unlike typical CLOs whereby any excess spread proceeds available after paying interest on the secured notes and fees are distributed to equity, some or all of the excess spread proceeds will be used to repay the "thick" tranche(s). In addition, these payments are independent of any coverage test (or other test) failures; meaning, if any interest proceeds are left at the bottom of the waterfall, payments will be made to reduce the 'BBB' note balance.

As a result of these payments, the CLO is able to repay costlier tranches first, resulting in the reduction of the overall cost of funding and increase in excess spread. All else equal, this represents an additional credit protection to the CLO notes; however, the initial subordination of these tranches tends to be lower than average 'BBB' and 'BB' tranches. If the deal is managed well and, as a result, a significant portion of the 'BBB' and/or 'BB' tranche is paid down, the subordination for such tranche is quickly increased. Like the delayed-draw tranches, there are some considerations to make when reviewing these tranches:

• The reliance and sensitivity of the tranche's payment on excess spread;
• The coverage tests and their potential volatility;
• The ability of the manager to re-allocate proceeds away from the "thick" 'BBBs' and 'BBs'; and
• In some cases, the presence of a large class of CLO X notes, senior in the capital structure, that would reduce excess spread.

Changes in issuer jurisdiction

In a typical CLO, the issuer will be incorporated in a tax-beneficial jurisdiction. Common jurisdictions we've seen include the Cayman Islands, Jersey, Ireland, and Bermuda. Once these issuers are incorporated in a given jurisdiction, it can be a complicated and expensive process to re-incorporate elsewhere. As a result, it's not common practice, outside of situations where there is a significant-enough governmental, regulatory, or a market-driven reason, for an issuer to change jurisdictions.

In response to the Cayman Islands being added to the European Union's Anti-Money Laundering (AML) "blacklist" in February 2022 (which is a list of countries that the European Union deems to have significant deficiencies regarding their anti-money laundering and counter-terrorism financial policies), some transaction parties have started to incorporate more recent issuer special purpose vehicles (SPEs) in jurisdictions outside of the Cayman Islands.

Due to the inclusion of the Cayman Islands on the EU's AML blacklist, European investors will generally not be able to invest in a CLO whose issuer is domiciled in the Cayman Islands. The question on whether European investors need to divest from existing investments that predated the Cayman Islands' blacklisting is still being debated within the market.

While the inclusion of the Cayman Islands on the EU's AML blacklist is a recent development, most transaction documentation we've observed already included the flexibility to change the issuer's domicile, if necessary, so long as the change is determined to not be disadvantageous in any material respect to the noteholders.

While we've continued to rate transactions with Cayman Islands issuers, we have also observed some CLO issuers trending toward setting up issuers in alternative jurisdictions such as Jersey, Ireland, and Bermuda, to name a few. In addition, we've also seen some CLO issuers that were originally incorporated in Cayman Islands redomicile via continuation, merger, consolidation, or reincorporation (or by other legal means) into other jurisdictions prior to, on, and sometimes, after a transaction's closing date. In rating a transaction, S&P Global Ratings analyzes the bankruptcy remoteness of the issuer SPE as described under our published methodology, which can be applied to any of the aforementioned jurisdictions. S&P Global Ratings also reviews any potential for tax liabilities in the relevant jurisdictions on a case-by-case basis throughout our analysis.

Related Research

• Are There Holes In The Buckets (And Could They Lead To A Delay In Coverage Test Failures)? Feb. 21, 2020
• Par Wars: The Phantom Limits, Feb. 21, 2020
• Par Wars: Short Debrief On U.S. CLO Document Trends, Sept. 29, 2019
• Par Wars: Return Of The Covenants? May 17, 2019
• Par Wars: The Investor Strikes Back, May 2, 2018