CLO Equity: a Nice Spot?

I waited for a prudent time for writing these lines. I wanted to see the latest transactions of the CLO market closed in 2022 and how the managers and investors were preparing themselves for the first quarter of 2023. I wanted to understand if there was room for yield tightening from the extreme levels reached by the liabilities toward the end of last year (AAA at discount margin at 220 and BBB at 700 for example.) I was also interested to observe if a potential rally in the credit market could have had the strength to attract the investors in the CLO space and to test the appetite of the players who stayed out of the CLO market for at least a couple of quarters or more in some cases. Either way, the CLO liabilities during January and February tightened; managers took the opportunity to come back to the market with “normal” transactions characterized almost by standard reinvestment and non-call periods. No more static deals imposed by market conditions or unusual combinations of one year reinvestment period associated with short non-call tenor that we saw from mid-2022 onwards (in difficult moments the CLO market reaches level of imagination not commonly found in other ABS structures). ?We will watch with the usual interest what 2023 will bring to the table for CLO platforms and investors in the asset class. What kind of vintage we will face remains a difficult task to say at this moment due to the economic uncertainty and some technical features of the loan market and the CLO space itself.

The credit indices tightened, the cash rallied, the CLO liabilities tightened (more down the stack than for the investment grade tranches). The total cost of liabilities is back to inside 300 bps coupon. The B tranches are sometimes available, in some other cases not offered.

I repeat here what I wrote in previous posts: the CLO market is the most studied asset class in the structured world and banks’ research and dedicated analytical platforms provide an enormous amount of data and information for all the actors involved. There is virtually all sort of data analysis for understanding the quality and exposure of every underlying pool of assets, the relevant tests and statistics, the past and present portfolio’s behavior (the future is still unknown!), cross sections examinations for comparing a manager’s performance against defined cohorts and similar peers. It is true how the data are sometimes not uniform or incomplete, but we can derive insight into every portfolio or manager’s style in a way and with the depth impossible to imagine in the 1.0 world (before the GFC).

This being said, I will not enter into any comment on portfolio rating migration, CCCs, OC cushions, manager’s profile. It will be a dangerous exercise, but most likely, a probable repetition of what has been already examined and carefully detailed by others.

Three things remain guaranteed:

1. CLO 2.0 and current vintages will not face (for prudence we should say very unlikely) the stress levels we have seen during the Great Financial Crisis where defaults and CCCs percentages reached double digits levels;
2. There is a lot of variety available across the managers and the vehicles of new production and the old vintages in the secondary markets. The more time passes, the more this trend accentuates. This started with COVID in 2020 for subsequently increasing quarter after quarter. This gives us the luxury of choosing the manager we like; this heterogeneity creates opportunities and offers strategies with a lot of convexity;
3. More than 20 years of financial performance of CLO structures should allow to clarify any doubt or perplexity around the behavior and the resilience of the asset class. The topics today- they come out every 5 years or so, over and over again- are the same as the ones studied in the GFC and the COVID spike. In possession of this dataset, it should be feasible to grasp undoubtedly a host of extremely valuable information for exploiting opportunities and for avoiding inappropriate trades. Following my usual investment approach, I should say regarding the last ones, the trades where the risk is not correctly paid. (In 2022, before summer, AAAs at 120 bps looked interesting, but they appeared to me less compelling in a macro scenario of rising interest rates. I warned about the widening in CLO market in June 2022).

Market participants (CLO managers and investors) agree we will continue to navigate uncharted territory and the direction of the loan market will drive volumes, performance and management style of the portfolios. (It is under debate, and for me it is conundrum, if it is the loan market who drives the CLO business, or it is the other way around).

What I will try to highlight here are the main points of the most important part of CLO capital structure (the AAA note buyers might not like this definition): the equity. ?I remember a piece of research some years ago produced by JP Morgan where the equity was described the “belle of the credit ball”. It is a fantastic definition because it describes accurately the complex, controversial and never properly modelled security of the CLO debt market. Let’s see what this asset class is about. We should avoid underrating the details that make this investment an interesting proposition even in a world where the CLO equity, in many circumstances, comes from origination vehicles or dedicated funds set up by big credit managers that tap the money when it is deemed appropriate for market conditions and favorable investment timing. It is rare today raising the equity through a road show like it was common practice some years ago. This is truer in difficult choppy markets and, of course, in the bearish credit environment that we witnessed over the recent years (I am trying to reason here if a CLO issue is driven by the investors’ choice in a particular market moment or by the asset manager decision… but this is very philosophical and thorny topic, and it will bring us off the road).

Loans and CLOs should overperform in a raising interest rate environment aided by the nature of being floating rate products (we saw…. this is not true). Many participants and analysts like to spend their time in comparing structured products- concretely here, the CLO notes- with the IG and HY spaces; I don’t see the value of these analysis, because they are misleading exemplifications, and the performance numbers of the various segments hide the real substance of what lies behind the numbers and what are the drivers that explain the yields and the securities’ behaviors (cheap and expensive comparisons are futile and ludicrous if the macro world is not duly taken into account. The investors in the asset classes are also different).

But, what about the equity? Equity payment distributions are allocated from excess cash flow, if available, following the payment of scheduled principal and interest payments to rated debt notes that rank higher in the CLO structure’s payment waterfall. Furthermore, the equity is the first loss piece in the capital structure which absorbs losses from defaulted assets in the underlying portfolio and benefits/suffers from capital gains/losses generated by the portfolio manager activity during the life of the CLO vehicle. CLO equity is a 10x levered asset class. These are my thoughts that I summarize as much as I can:

1. One of the main points, the principal in my opinion, is that CLO equity is a cash flow product who benefits from cash distributions. I see CLO equity as a carry trade with mid-teens (or higher) percentage yield. All distributions by managers across different vintages in US and Europe, demonstrate along the years those returns (the two markets are similar to each other in 2.0 world, but they were different in late CLOs 1.0 because the US market came out of the GFC well before Europe). It is true how there is significant dispersion in the managers spectrum: there are marked differences (in the same year we can have more than 10% range) in the dividend yield paid by the best performing platforms and the IRR generated by more defensive portfolio strategies. It is worth noting how the managers who are more “equity friendly” tend to price the rest of the capital structure- the debt- wider than other conservative portfolio managers. It is also true how some CLO platforms change their portfolio management style during the years, affecting the equity cash distributions: the style drift is the result of many factors such as different response to risk factors, opportunities that arise in the market (COVID for example), different ways of addressing specific demands and target yields coming from debt holders and/or the equity investors. The managers’ tiering phenomenon in Europe is less accentuated than in the US marketplace, but there are clear differences for the attentive eyes. The liabilities pricing, in my view, in Europe is still not adequately reflecting those differences.
2. The loans in a CLO portfolio are a negatively convex asset class and the equity is levered 10X against them.?The consequence for the equity holders lies in the fact they have limited possibilities to see their NAV moving up naturally over the life of the vehicles (volatility reasons excluded). This explains why the interest rate component paid to the equity is the main source of IRR if we exclude the capital repayment when the CLO structure is called or reissued. It is widely accepted the returns achieved by the equity can only be challenged by the stock market, but it is evident the different nature of the two asset classes: one is based on capital appreciation while the other finds its justification in the cash flows. There is also another difference: the real and exact IRR of the equity investors in the structured product should be calculated when the CLO structure is unwound, and when it is finally and exactly known how much capital is returned to the equity from the liquidation of the assets in the portfolio. CLOs suffer from “tail problems” and the final IRR could be different from the initial assumptions. Expected and realized IRR can be misaligned. This explains the first component of the CLO equity modelling problem.
3. It should be clear how the liabilities pricing has an impact on the future cash flows of the equity. I’m hearing this topic since 1.0 world. The perfect arbitrage conditions between assets and liabilities never seem attainable, and managers and equity investors are always inclined to complain against unfavorable market conditions. This is only half of the truth. What should be more relevant instead are two elements:1) the long-term value of the manager in creating alpha for the equity and improving the cash flows 2) what is happening on the asset side after the CLO vehicle has printed: volatility in the prices of the loans, portfolio behavior, repricing and repayments in the collateral pool, forward loan calendar volume. Some elements are falling in the manager’s ability to manage the portfolio efficiently, some others are more difficult to control, and they are a function of the market mechanisms and the general trends in the financial markets. The good managers should be capable of exploiting the ample opportunities available time to time, to drive value to the equity investors through trading activities and adding risk or switching to a more conservative approach depending on the market’s view. I define this activity “navigate through the matrices” of the rating agencies by adopting different measures for different times. Only a few managers do this and when a management drift is happening is sometimes more imposed by external circumstances than the consequence of a change in strategy. (An example is the current wide usage of the bond bucket by all managers in Europe, while others, including myself, embraced this approach a long time ago and defended the fixed income strategy together with loans against some reluctant equity investors).
4. The CLO liability side and its weighted average cost of capital (WACC) can be driven by supply/demand mechanisms and financial markets conditions that could prevent meaningful CLO spread tightening even in favorable market circumstances. This happened many times and the most recent example is the widening of the upper part of the capital stack occurred in the summer 2021, followed by complicated months, due to the excess of offer from new deals, resets and refinancings. This brings me to the delicate point of ALM even in the CLO market. Managers can boost equity returns by actively managing their liabilities through the repricing option. I call it the “wild card for arbitrage”. The important decision of correctly setting up the no-call period (and the reinvestment period) when the manager presents and prints the transaction in the market, together with the strategic initiatives of resetting or refinancing the vehicle at the right moment, have deep consequences for the future cash flows received by the equity portion. The side of the market on the CLO repricing took a while to develop and to mature, but not all asset managers have been active in their organizations in hiring relevant people with structuring capabilities and market vision. The equity investors should start to evaluate the credit platforms also under this different light, not only on asset selection. Market volatility occurred at the time of CLO pricing should not represent a drag on equity performance in the medium and long term. From this sentence we derive another strong point: the longer is the life of the reinvestment period, extended with the optionality just mentioned, together with the right combination of WAL and collateral quality tests, the better is for the dividend yield in future equity distributions.( This is the second part of the modelling cash flow problem for equity which is emerging again: how to “price the option” given to the manager and how wisely this will be used).
5. CLO equity tends to trade less than other parts of the capital stack, but I struggle to classify it a low beta asset class, less volatile than the junior mezzanine tranches (this comparison appears sometime in the research). By observing the average NAV volatility movement over the last 12 months, we can easily verify the large swings suffered by the CLO equity that started when the loan market sold off in February 2022 after the Russia’s invasion of Ukraine. The market value of the collateral pool started depreciating and the NAV for the median deal in Europe collapsed to a low of -40% reached in September 2022. The market recovered since then, and in many deals the NAV values are back in positive territory. What I do notice is the wide dispersion of NAV in the market for the 90th percentage deal and the average of the market. This dispersion should drive more “managers tiering” and different pricing in the rest of capital stack than it is actually happening in secondary market and new issues. If this is not the case, it is due to the relatively small investors base for CLO products in Europe.
6. Manager choice for equity performance. I left this point at the end of the list because it helps to draw the final conclusions. A CLO structure has a strong and unique point working in its favor: the existence of the CLO manager who runs the pool of assets. Good managers can keep track of the risks in the market and their portfolios and minimize negative credit migration, accumulation of CCC credits in their specific bucket and, finally, defaults. Active management of the underlying portfolio, along with the ability to identify dislocations and therefore opportunities, is a definitive and distinctive characteristic of all CLO platforms and notably of the top quartile managers. The dispersion in Europe across the different platforms has increased over the years and I am not exaggerating in stating that this trend started some time before COVID 2020. To be more accurate, in my opinion, between the end of 2018 and the first quarter of 2019, when we registered the first signs of volatility in the credit market and challenging environment. Before that period, we had a monotonous bunch of credit platforms performing in similar terms and doing roughly the same things. The managers tested indeed the first downgrades, CCC credits and their volatility from mid-2018. Nowadays, in Europe, we have enough data, numbers, and statistics for studying managers’ performances, par builds and equity distributions and for understanding how and why managers make investment decisions. Despite some obvious pitfalls in analyzing just numbers, it is reassuring to know that most of the CLO portfolio managers are able to outperform the loan index over time and to reduce and contain the default rate of the credits pool. The existing dispersion across the managers on equity distributions is not necessarily a negative point because equity investors have contrasting preferences and views regarding portfolio management and active vs passive trading mentality. The relevant and key factor here is to give emphasis to fulfilling the equity IRR expectations with a given (or chosen) volatility and risk target for the portfolio. Data abundance and granularity should be interpreted sometime with the help of the big picture; this is not done all the time. We look at the ants, but we miss the elephant in the room. There are old and actual examples in the market who support what I am writing here but it is wise just this brief mention.

Linked to the last point are my conclusions: investing in equity of CLO should be seen as a long-term investment strategy. It is, of course, possible to shorten the investment horizon but we risk altering the nature of the asset class. In the circumstances of a reduced holding period, the equity of CLO will resemble an investment in a stock exchange. It will be an opportunistic position looked through the lens of the CLO market.

All the above leads to the end of this post: what are the drivers of equity performance in a CLO structure? I wrote about good managers, arbitrage between WAS and WACC, trading activity for par building, capacity to use and exploit the optionality given to the equity for resetting and refinancing (look at late 2017 vintage and 2018 vintage if you are curious to observe what is going to happen to those structures and their cash flow profiles), thickness of the equity in the cap stack ( the average of 2022 has been 8,78%, all included, even the static deals). On the last one (I left explicitly this theme at this point): there are big differences across the managers’ capital structures in the CLOs vehicles. A 1% difference in a position levered more than 10 times and with an arbitrage of 200 basis points, can really change a manager’s performance and alter any market analysis conducted on the different platforms and their IRRs. Any study on trading gains and/or asset portfolio rotation, for example, loses its importance and value if we forget to look at how much equity has been deployed in the capital stack. (Material variance in the first CLO structures of 2023 already appeared).

With these lines in mind and some caveats in our pocket, we remain with a question not yet answered. Are CLOs in 2023 a commodity product or still an exotic asset class set up by managers to meet the demand of certain investors willing to put money to work in securities with different layers of risk/return profile? The difference is not insignificant (for the equity and for the debt) because it can explain what is happening in the CLO world with its protagonists; armed with a possible CLO definition we are provided with the instructions for reading the events that unfold under our eyes in this part of financial markets.

Note of the editor: Sergio Grasso has been manager of CLO vehicles since 2006 in 1.0 and 2.0 structures. He started to invest in CLO junior mezzanine debt in 2004.

Author: Sergio Grasso , Director at iason ?

Previous Market Views available here ?