Nineteen new European collateralized loan obligation (CLO) transactions priced in second-quarter 2019, adding a €7.83 billion of supply to the €6.89 billion from 16 transactions in first-quarter 2019. Despite both macroeconomic and arbitrage concerns earlier in the year, the CLO market has remained robust, with total issuance to-date standing at €20.1 billion across 48 CLO transactions, up from €17.9 billion from 43 transactions over the same period last year.

Four new managers joined the CLO market in the first half of 2019 with Fair Oaks Capital, Napier Park, MeDirect, and CIFC CLO Management II LLC launching their debut European CLOs. While arbitrage concerns due to tight loan spreads are likely to carry into the second half of the year, there are signs of improving arbitrage conditions with CLO 'AAA' spreads tightening with more established managers.

Fundamental concerns continue to persist as much of the market's focus remains on how the next cycle will pan out. (In our July 11, 2019 webcast, we addressed this and discussed other themes, including liquidity versus solvency in the European leveraged loan market, and covenant-lite loans and recovery rates, as well as looked into European CLO defaults to see what lessons could be gleaned from the last crisis. To access the webcast recording "Under The Microscope: CLO Default Deep Dive And The European Leveraged Loan Market" see the "Related Research" section.) Nevertheless, given continuing benign economic conditions and a low European speculative loan default rate, we expect CLO performance to remain stable and for loan defaults to remain low in the near term. As a result of market participant requests, we have included the list of second-quarter 2019 corporate rating actions and the exposure in European CLOs (see the Appendix section).

In terms of CLO structures and documentation, we note the following observations (or at least proposed language in initial drafts of offering circulars):

• Increasing leverage in CLO structures. In first-quarter 2019, there were only two transactions with a leverage (calculated as total debt divided by equity) above 10x. This increased to 11 transactions in second-quarter 2019, and the trend looks to continue into the next quarter with GoldenTree Loan Management EUR CLO 3 having a leverage ratio of 12.8x.
• Larger bucket of trading plans to 7.5% from 5.0%.
• Proposed carve-out of certain assets from the 'CCC' bucket such as assets that are priced above par.
• Proposed carve-out of certain assets from the definition of defaulted obligations such as assets that have been deferring interest for a limited period of time and a market value above a certain value.
• Proposed changes to reinvestment conditions. For example, during the reinvestment period, the ability for managers to purchase assets at a premium without having to meet the condition of the portfolio being above reinvestment target par. We also have seen some cases where the manager has the flexibility of meeting a lower reinvestment target par condition than how it is typically defined (target par balance less note paydowns plus additional note issuance).
• Ability to make non-senior note cancellation. Although collateral enhancement proceeds rather than principal are used to make these redemptions, if not kept outstanding for the purposes of the coverage test or reinvestment target par, such a feature could erode protections for the senior noteholders.

In this quarterly index publication, we take a look at some of the key metrics behind our ratings on the CLO notes. A month-to-month negative performance of these parameters could lead to a rating pressure on the notes.

Measuring CLO Performance Using Key Metrics

CLO issuance has gained momentum over the past five years, and investors have become more familiar with CLO structures and the associated risks, as well as assessing and suitably measuring credit and cash flow risks.

Credit risk, which includes default risk and an increase in 'CCC' category-rated assets in the portfolio, among others, can be mitigated by better measures on the cash flow side, like increased available credit enhancement, weighted-average spread, and recoveries, for example.

In this article, we display how these individual parameters have evolved over the last few months to broadly gauge the performance of European CLOs.

Changes in corporate CLO methodology and non-model CDO Monitor assumptions incorporated

We recently revised our corporate CLO methodology. The changes to our criteria incorporate 10 years of additional data on corporate ratings performance and global CLO performance while aiming to provide a simplification and increased transparency to our rating approach. (For more information, see "Credit FAQ: Understanding S&P Global Ratings' Updated CLO And Corporate CDO Criteria," published June 26, 2019, and "Global Methodology And Assumptions For CLOs And Corporate CDOs," published June 21, 2019.). At the same time, we updated the six benchmarks that portfolio managers can use to assess and manage their portfolios. (For details on all six benchmarks, see "S&P Global Ratings' Updated Assumptions For CDO Monitor Non-Model Version," published June 21, 2019). In this report, we have hence introduced the overall S&P Global Ratings' weighted average rating factor (SPWARF), which looks at the underlying asset ratings to provide an indication of the overall credit quality of the portfolio (see chart 9)

CLO Performance Remained Stable In Second-Quarter 2019

Overall, CLO performance in the second-quarter of 2019 was similar to the previous three quarters. Most of the metrics we capture that may affect ratings on the notes showed stable performance.

Collateral portfolios of older vintage cohorts are becoming more concentrated as the assets wind down and they approach their final maturities, while newer vintages are benefiting from still being in their reinvestment phases, when collateral managers can actively mitigate default risk through active trading. We attribute these trends more to the stage in a transaction's life cycle than to significant changes in the portfolios at the collateral level.

Credit Metrics

European CLO 2.0 collateral ratings

While CLOs enjoy the senior secured status of leveraged loans in the portfolio, it is important to note that these loans are issued to speculative-grade companies.

Underlying collateral ratings contribute significantly to the ratings on transactions that have closed since January 2013. Below we show the rating distribution of the CLO collateral portfolio for the different vintages in European CLO 2.0 transactions over a one-year period (see charts 1 to 6). Note that we have considered transactions that have been reset or refinanced to be in the original vintage as when it was first issued. The CLO portfolio rating performance across all CLO vintages indicates stable performance.

Chart 1

Chart 2

Chart 3

Chart 4

Chart 5

Chart 6

Exposure to 'CCC' rated assets continues to be well below the allowable limits

'CCC' category-rated assets are an important measure of European CLO performance because an increase in 'CCC'-rated assets can indicate worsening credit quality of the collateral portfolio. The level of 'CCC' assets can also reduce overcollateralization (O/C) test cushions because they may not be carried at their full par value.

The percentage of assets rated in the 'CCC' category ('CCC+', 'CCC', or 'CCC-') has shown a mixed performance for the European CLO cohorts we track. These changes reflect rating migration in the underlying portfolios and may also depend on the pool composition of individual transactions, which is based on the CLO manager's strategy to manage the vehicle.

By vintage, the reported level of 'CCC'-rated assets in European cash flow CLOs, as a percentage of total assets in May 2019, was:

• 2013 vintage CLOs: 1.08% of total assets (up from 0.92% in February 2019);
• 2014 vintage CLOs: 1.27% of total assets (up from 1.09% in February 2019);
• 2015 vintage CLOs: 1.67% of total assets (up from 1.65% in February 2019);
• 2016 vintage CLOs: 0.92% of total assets (down from 1.05% in February 2019);
• 2017 vintage CLOs: 0.88% of total assets (up from 0.86% in February 2019); and
• 2018 vintage CLOs: 0.85% of total assets (unchanged from 0.85% in February 2019).

Chart 7

Individual CLOs exhibited some variances among European CLOs from the same vintages. These CLOs are more likely to breach their thresholds sooner than other types of CLOs. Having said that, in this low interest rate and benign default environment, having a 'CCC'-rated asset may not necessarily be negative, because cost of debt and equity return can be better managed.

Exposure to defaulted assets remain limited

CLOs performed well through the financial crisis and beyond. Defaulted assets were one of the key indicators of CLO performance because a defaulted asset may result in a loss of principal to the CLO and a corresponding decline in credit enhancement.

From February 2019 to May 2019, the percentage of defaulted assets (i.e., assets from obligors rated 'CC', 'C', 'SD' [selective default], or 'D') slightly decreased for the 2013, 2016, 2017, and 2018 vintages, and remained stable for the remaining two cohorts. We note that in the previous quarter, there was an increase in defaulted assets, largely due to the exposure to the U.K. apparel retailer New Look, whose rating was lowered to 'D' in January 2019 (see "New Look Downgraded To 'SD' On Approval Of Debt Restructuring; Debt Ratings Lowered To 'D'," published Jan. 31, 2019).

As of May 2019, the percentage of defaulted assets in each underlying collateral portfolio was:

• 2013 vintage CLOs: 0.25% of total assets (down from 0.42% in February 2019);
• 2014 vintage CLOs: 0.37% of total assets (unchanged from 0.37% in February 2019);
• 2015 vintage CLOs: 0.48% of total assets (unchanged from 0.48% in February 2019);
• 2016 vintage CLOs: 0.02% of total assets (down from 0.09% in February 2019);
• 2017 vintage CLOs: 0.05% of total assets (down from 0.06% in February 2019); and
• 2018 vintage CLOs: 0.04% of total assets (down from 0.06% in February 2019).

These calculations show the proportion of assets that are currently in default, over total assets (not including principal cash).

Chart 8

Benchmarking credit risk in a CLO: S&P Weighted Average Rating Factor

Although CLOs are generally restricted to eligibility criteria that govern what assets can be part of their portfolios and sets their limitations, it is challenging to size a portfolio's default risk during the typical four-year reinvestment period where the collateral manager is allowed to actively trade assets. These trading activities could change the asset portfolio's composition significantly and thus, increase its risk profile and, possibly, the required par subordination.

The SPWARF provides an indication of the overall credit quality of the portfolio and is the individual asset five-year default rate assumed in our corporate collateralized debt obligation (CDO) criteria, weighted by each asset's par balance, and multiplied by 10,000 (see the Related Criteria And Research section).

In second-quarter 2019, the overall SPWARF increased from 2618 to 2626.

Chart 9

Weighted-average life (WAL)

The WAL is the number of years between the current date and the maturity date of assets in the CLO portfolio.

At 5.2, the WAL is decreasing quarter on quarter.

Chart 10

Scenario default rates (SDRs)

Together with the SPWARF and WAL, we use four other benchmarks (the three diversity measures and the default rate dispersion [DRD]), to produce the approximate 'AAA' SDR (i.e., the expected default levels for the portfolio under the 'AAA' stress scenarios).

While the SPWARF only looks at the credit rating on the assets, SDRs (or the expected target default rate) look into all six components when measuring the overall risk profile of a CLO portfolio (SPWARF + DRD + WAL + the three diversity measures).

On average, the current portfolio credit risk ('AAA' SDRs) has improved in comparison with first-quarter 2019, decreasing to 63.85% from 64.64%.

Chart 11

Cash Flow Metrics

Credit enhancement has remained stable

Our analysis of CDO transactions, as in our other structured finance ratings, focuses on how much credit enhancement is needed for a given level of credit risk to achieve a specific rating. Typically, credit enhancement is provided by a combination of overcollateralization/subordination, and cash collateral. In this case, credit enhancement is the percentage of total performing assets plus cash, minus the tranche balance (including senior and pari passu note balance), divided by total performing assets plus cash plus recovery on defaulted assets. The credit enhancement levels across the capital structure remained stable over 2018 and 2019.

Table 1

Weighted-average spread

Spreads vary based on a variety of factors, including the levels of relative liquidity for leveraged loans or the actual and perceived level of credit risk in the leveraged loan market, among others.

Over the past two to three years, leveraged loans have refinanced at a lower cost, leading to increased difficulty in managing the weighted-average spread test in CLOs and in maintaining the weighted-average cost of debt and a healthy return to equity. Consequently, weighted-average spreads are monitored closely in CLOs. If this measure decreases significantly, the risk of a negative rating action on the notes would increase.

On average, the weighted-average spread has remained stable over the past three quarters, which has provided some comfort to CLO managers in managing the weighted-average spread test.

Chart 12

Senior O/C ratios show slight decline

The senior O/C ratio test is a par value test to protect senior noteholders. Declines in the senior O/C ratio test results can indicate decreasing credit quality of the CLO. The O/C ratio is the difference between the O/C test calculated for a particular tranche and the trigger associated with it. Breach of these triggers will mean that senior notes are repaid (until the tests are met again), or if the transaction is in its reinvestment period, the proceeds due on junior notes are either invested in substitute collateral or used to repay the notes.

The senior O/C ratio test cushions worsen from February 2019 to May 2019 for all the cohorts (see chart 13).

The senior O/C ratio test cushions (based on reported information) as of May 2019 were:

• 2013 vintage CLOs: 8.53% (down from 8.76% in February 2019);
• 2014 vintage CLOs: 6.03% (down from 8.81% in February 2019);
• 2015 vintage CLOs: 8.69% (down from 9.08% in February 2019);
• 2016 vintage CLOs: 9.76% (down from 9.83% in February 2019);
• 2017 vintage CLOs: 9.53% (down from 9.59% in February 2019); and
• 2018 vintage CLOs: 9.03% (down from 9.50% in February 2019).

Chart 13

Subordinated O/C ratios are still contracting

The subordinated O/C ratio test is the par value test for the junior notes in the CLO. Failure to satisfy this test will typically cause interest and principal to be redirected to pay down the most-senior class of notes until the test is satisfied.

From February 2019 to May 2019, the subordinated O/C ratios showed mixed performance for all of the CLO 2.0 vintages we track (see chart 14).

As of May 2019, the subordinated O/C ratio test cushions (based on reported information) were:

• 2013 vintage CLOs: 3.47% (down from 3.74% in February 2019);
• 2014 vintage CLOs: 1.73% (down from 3.49% in February 2019);
• 2015 vintage CLOs: 3.71% (down from 3.92% in February 2019);
• 2016 vintage CLOs: 4.39% (down from 4.71% in February 2019);
• 2017 vintage CLOs: 4.28% (down from 4.34% in February 2019); and
• 2018 vintage CLOs: 4.30% (up from 4.28% in February 2019).

Chart 14

Notes

Our European CLO Performance Index Report provides aggregate performance statistics across most of our rated European cash flow CLO transactions backed primarily by corporate loans. We provide this information to help market participants track the overall performance of European cash flow CLO transactions and to benchmark the performance of the transactions they follow against the performance of cohorts of similar transactions.

Our report highlights what we view as a number of key risk areas for the transactions, and which we use as part of our analysis of the credit quality of securitized portfolios and of the transactions' payment structure and cash flow mechanics. These include rating migration within the underlying collateral portfolios, as well as other information relevant to the sector.

We divide the performance information in the CLO indexes into cohorts, each containing data for most of the European CLO transactions we rated and issued in a specific vintage year. We collect the performance information from transaction-level performance data in our CDO surveillance databases.

Information prior to the most recent 12 months is available on CDO Interface, S&P Global Ratings' web-based portal for CDO performance information, at www.cdointerface.com. To generate, view, and download data from the CDO indexes, log onto CDO Interface, and then select the "Indexes" tab.

Appendix

Appendix

Related Criteria

• Global Methodology And Assumptions For CLOs And Corporate CDOs, June 21, 2019

Related Research

• Webcast: Under The Microscope: CLO Default Deep Dive And The European Leveraged Loan Market, July 11, 2019 https://event.on24.com/wcc/r/2039751/C183B9BB4232395E75E1F65054930032
• Second-Quarter 2019 Non-Model CDO Monitor Benchmarks Reveal Relative Credit Quality And Diversity Of CLO Portfolios, Aug. 2, 2019
• Credit FAQ: Understanding S&P Global Ratings' Updated CLO And Corporate CDO Criteria, June 26, 2019
• S&P Global Ratings' Updated Assumptions For CDO Monitor Non-Model Version, June 21, 2019
• 2018 Annual Global Leveraged Loan CLO Default And Rating Transition Study, June 19, 2019
• A Cycle Turn Will Test European CLO 2.0 Defaults, June 7, 2019
• European CLOs: Lack Of Loan Supply Is Causing Further Portfolio Overlap, May 30, 2019
• New Look Downgraded To 'SD' On Approval Of Debt Restructuring; Debt Ratings Lowered To 'D', Jan. 31, 2019
• Glossary Of Cash Flow CLO Performance Index Fields, Jan. 30, 2009