Introduction

The junior portion of the CLO capital structure has seen solid performance so far in 2023. Over the course of the past several months, Pretium has highlighted this outperformance and made the argument that CLO mezzanine securities were undervalued relative to HY single-name credit, supporting this latter point by comparing the currently wide CLO BB spreads to the historical spread relationship the two asset classes have traditionally maintained. Given the uncertainty surrounding where the economy and the Federal Reserve may go from here, we have received inquiry from our investor base as to whether or not we view this outperformance as sustainable. In summary, we believe it is. Pretium believes that current CLO mezzanine and equity pricing reflects worst-case assumptions related to forward defaults and recoveries. While Pretium expects loan default rates to rise and recovery rates to remain depressed, we believe that current CLO spread and yield levels more than compensate for the fundamental strain that credit markets have already begun to realize and are likely to continue to experience over the coming quarters (Exhibit 1 below).

CLO BBs and CLO equity provided +8.6% and +7.6%, respectively, in (non-annualized) total returns over the first six months of the year vs. just +4.8% for HY corporate bonds1. Pretium believes this strong YTD performance for CLO instruments is sustainable across a variety of possible forward market outcomes. The market environment in the first half of 2023 reflected only muted spread changes and close to flat point-to-point long-term interest rate moves. Thus, the strong 23H1 CLO results for that period reflect what we see as only slightly above average expected returns for a sector which entered the year offering particularly generous spreads and yields.

BB CLOs began 2023 with average yields of approximately 14%, reflecting SOFR margins at the time of over 900 basis points2. The strong total return for BB CLOs over the first half of 2023 largely reflected interest carry, as opposed to mark-to-market price gains which might have been harder to replicate in the future. By contrast, BB corporate bonds have yields-to-maturity of just 7%, with correspondingly smaller scope for carry returns (Exhibits 2-3 below).

To assess the sustainability of CLO outperformance going forward in greater detail, we review expected returns for CLOs across a range of possible future interest rate and macro-economic scenarios. Our analysis suggests that mezzanine CLOs should outperform generic HY corporate bonds over a short-term (e.g., 1-year) horizon across most scenarios, with an even higher probability of outperformance over a hold-to-maturity horizon. Long-term expected returns for CLO BBs remain elevated under adverse scenarios at current pricing levels due to the fact that the investments benefit from significant structural protections which allow for full principal repayment under all but the most extreme default and recovery rate assumptions (Exhibit 4 below).

While CLO BBs can provide long-horizon downside protection in deep recession scenarios, CLO equity can complement the profile of CLO debt by offering a return distribution with significant upside potential in scenarios involving only mild recession or recovery. Combined, a portfolio which incorporates both CLO equity and CLO debt securities produces a highly attractive distribution of returns across a range of potential future economic outcomes.

Interest Rate Scenarios and Their Impacts on CLO Returns

The scenarios we include in our analysis are:

• 1. Follow Forwards (high probability). Current interest rate curve plays out in line with current forwards, inflation continues to move down from current spot levels but remains above 2% target, unemployment ticks up modestly and corporate earnings stay flat or see a muted pullback: a soft landing scenario in which credit will outperform vs. current pricing and realized defaults are slightly below the market baseline forecast.
• 2. Higher for Longer (high probability). Interest rates stay higher for longer, remaining close to current spot levels, inflation comes down slightly but remains elevated, unemployment ticks up and corporate earnings stay flat or see a muted pullback: a soft landing scenario in which credit will marginally outperform vs. current pricing and defaults are in-line with the market baseline forecast.
• 3. Stubborn Inflation (low probability). The Federal Reserve hikes rates over the next twelve months in response to stubborn inflation and strong corporate earnings, inflation remains unchanged vs.current spot levels and unemployment remains low: a short term bullish market for credit with default rates lower than forecast due to strong corporate earnings.
• 4. Hard Landing (low probability). Material economic weakness emerges in the second half of 2023, inflation trails lower towards target and unemployment becomes elevated: a hard landing scenario for the economy and credit with short term and long term default rates above baseline forecasts and pulled forward.

Further detailed assumptions regarding these scenarios, including the assumed unemployment rate, inflation rate and short and longer term interest rates, are provided in Appendix A.

The calculations of expected CLO and corporate bond returns under the alternative scenarios incorporate both projected coupon income (e.g., with lower income for floating rate CLO instruments in falling rate scenarios) combined with, over the short horizon, expected mark-to-market price changes, and, over the longer term, potential realized credit losses.

Short-Term Outlook

Exhibit 5 below shows the estimated scenario returns over a one-year horizon holding period. In scenarios 1 to 3 the returns for BB CLOs are projected to significantly exceed those for BB corporate bonds. This outperformance is driven by higher coupon income for CLOs. In effect, the projected outperformance of CLOs in these scenarios mirrors the recent historical outperformance of CLOs vs. HY bonds, which has also been largely driven by superior carry for the CLOs.

For scenario 4, a Hard Landing, CLO BBs are projected to deliver lower returns than fixed rate HY bonds. This modeled underperformance is a consequence of an expected decline in CLO floating rate coupons as the Federal Reserve cuts interest rates in response to an economic contraction, combined with an assumption that generic fixed rate bond spreads widen less sharply than CLO BB spreads into the contraction.

The short-term CLO equity performance estimates shown in Exhibit 5 are calculated from the cash-on-cash equity yields expected in these scenarios combined with expected instrument price changes; the price change estimates, in turn are estimated based on the historical relationships between CLO equity and HY bond prices across different market environments. In scenarios 1 to 3, CLO equity is projected to significantly outperform BB corporate bonds as high current carry is offset by only marginal changes in cash flow discount yields. Near-term changes to discount rates remain range-bound as default expectations over the first year are primarily constructed from credits that are already identified as candidates for default as a function of their current trading price. In scenario 4, CLO equity is projected to underperform as the pull-forward of defaults in credits that are currently trading above $90 is likely to result in a downward revision to forward cashflow projections combined with an increase in market yields used to discount the cashflows.

Hold-to-Maturity Outlook

Exhibit 6 shows the expected hold-to-maturity returns for the different asset classes across scenarios. Here, CLO BBs are projected to outperform HY BB bonds across all the scenarios, including the Hard Landing. Historically, only a small percentage of BB CLOs have ever experienced principal losses, including during the global financial crisis period, due largely to the structural protections from which the bonds benefit3. Accordingly, we project no CLO BB losses under the four scenarios considered. Generic HY corporate bonds lack these structural protections and hence, are projected to experience small but non-zero performance drags due to defaults, which are modeled here as ranging from a 0.1% annualized yield degradation in the more benign scenarios to a 0.3% yield impairment in the Hard Landing scenario. Most of the projected BB CLO vs. BB corporate bond outperformance though, comes not from differences in credit loss assumptions but rather, from the substantially wider spreads CLO BBs currently contain, which helps generate much stronger carry income over the lives of the respective bonds.

While CLO debt is seen in Exhibit 6 to provide high prospective yields and significant long-horizon downside protection, CLO equity offers a return distribution with significant upside potential in scenarios involving milder recession or recovery. For example, in the Follow Forwards scenario, CLO equity would be expected to deliver a 17% annualized return. Combined, a portfolio incorporating CLO equity and CLO BB securities can produce a risk-return profile superior to those available in public equity or single-name HY credit markets.

The hold-to-maturity CLO equity performance estimates in Exhibit 6 are based on IRR calculations which project CLO equity cash flows under the alternative scenarios. The modeled CLO equity held-to-maturity return is lowest in the Hard Landing scenario, in which loan default rates are assumed to be relatively elevated and CLO equity cash flows are consequently reduced. Lower interest rates in this scenario also put downward pressure on CLO equity cash flows in this adverse scenario. Even in the Hard Landing scenario however, we project a portfolio of CLO equity securities to ultimately deliver low double-digit IRRs given the currently favorable pricing of these instruments

Summary

Our analyses suggest that the strong total returns delivered by CLO mezzanine debt and equity securities in the first half of 2023 are sustainable going forward given current pricing. An analysis of the projected returns for CLOs vs. fixed rate HY corporate bonds suggests that CLOs should outperform over a 1-year horizon across most economic scenarios, with outperformance even more probable over longer holding periods. Pretium believes the attractive relative value provided by a portfolio of CLO equity and CLO BBs clearly indicates that CLOs should be an integral component for every allocator looking to optimize their corporate credit exposure in this environment.

Appendix A — Economic scenario assumptions

Mezz opportunities

Pretium structured credit and CLO liabilities MD Ian Wolkoff and Director Marty Young argue that mezzanine CLO bonds should be on the radar screens of most asset allocators with long investment horizons

US CLO mezzanine debt offers high expected return rates, with yields maintained under a default scenario more extreme than the GFC credit cycle and with upside potential in a market recovery scenario.

CLO debt offers strong long-run expected returns with bounded downside

In the current market context, macro uncertainty and the potential for a growth downturn loom large while valuations across many sectors still screen as relatively expensive. Given this set-up, asset allocators are searching even more actively than usual for investment opportunities which offer high potential yields but with contained downside protection in bearish scenarios. CLO mezzanine debt is a sector which appears to offer this combination.

The US CLO market has grown to over US$950bn in outstanding balance, with over US$160bn of balance in mezzanine (single-A through single-B) tranches. As a result, a CLO mezzanine debt strategy addresses a market large enough to allow investors to allocate selectively while still targeting high yields with contained risks.

Yields on CLO mezzanine debt securities have moved up sharply over the past two years; double-B CLO bonds purchased in the secondary market offer average lifetime quoted yields of 14.4% as of June 2023 versus 7.7% in mid-2021. The current baseline yields for double-B CLOs compare quite favourably to the yields available from other similarly rated fixed income assets: double-B leveraged loans and double-B corporate bonds, for example, currently yield just 7.6% and 7.2% respectively.

The extra yield earned by double-B CLOs versus double-B corporate bonds does not appear to reflect added default risk as the comparison holds debt ratings constant; indeed, during the 2008 financial crisis episode, CLO bonds had significantly lower default rates than similarly rated corporate bonds. CLO transactions are designed so that the bond classes can withstand elevated default rates on the underlying loan portfolios without taking principal losses; the double-B CLO bonds benefit from multiple structural protections, including overcollateralisation and excess spread that absorb losses before they can be passed to the double-B bonds.

A price-yield analysis of a double-B tranche from a sample 2021 vintage CLO transaction indicates that the double-B bond yield remains in double-digits under a scenario in which 28% of the underlying loan assets default over the next five years with a 60% recovery rate – a default intensity over 1.6 times the five-year cumulative default rate experienced following the global financial crisis. Indeed, the double-B tranche yield remains positive, even if the assumed five year cumulative loan default rate rises to 35%, more than 2x the level seen during 2007-2013. CLO double-B bond yields remain positive in this high default rate scenario, despite the fact that much of the bond principal would be projected to be written down in such an event, as the high projected coupon payments over the lifetime of the bond would compensate for the principal loss.

CLO debt offers upside potential in a market recovery scenario

Per above, double-B CLOs can offer around 14% yields under a moderate recession baseline scenario. The bonds can maintain these double-digit yields in a loss scenario comparable to that experienced during the global financial crisis and can continue to earn positive yields in scenarios significantly more extreme than in 2008. Thus, the long-run downside risk for the majority of bonds in the double-B CLO sector, while not zero, appears bounded.

At the same time, the bonds provide meaningful potential return upside in a scenario in which double-B CLO spreads partially normalise relative to their current wide levels. Double-B CLO spreads are currently trading 250bp above the level that would be predicted, given their historical relationship to high yield corporate bond spreads. If 60% of this pricing anomaly were to be slowly corrected over a two-year period, the double-B CLO horizon yields would be expected to reach around 18%, with 14% of the yield coming from coupon income and 4% from price appreciation.

Summary

Mezzanine CLO yields are currently elevated relative to the yields available on comparably rated corporate bonds. While CLO bonds may experience short-term mark-to-market price volatility, over a longer-term horizon, yields would be expected to remain in double-digits across a wide range of loss scenarios – including scenarios featuring default rates more extreme than those that were realised during the challenging 2007-2013 period.

At the same time, mezzanine CLO bonds offer realistic chances of return upside if bond spreads partially revert back to more historically typical levels. Given this favourable risk/return profile for mezzanine CLO debt, Pretium believes the sector should be on the radar screens of most asset allocators with long investment horizons.

Based on Pretium’s research, data and estimates from publicly available sources. Statements throughout these materials, including those regarding the market, represent the opinions and beliefs of Pretium. There can be no assurance that these will materialise.

This article was published in Structured Credit Investor on 28 June 2023.

Pretium believes single-family rentals are an essential part of the US housing landscape because they provide access to suburban neighborhoods where opportunities and amenities for residents have historically been most abundant.¹ Unfortunately, the housing market has over time struggled to create enough rentals in single-family neighborhoods. Recent research by Harvard’s Joint Center for Housing Studies finds that rental deserts — neighborhoods with the fewest rental options — tend to be disproportionately found in suburban locations.² We believe the primary driver of the shortfall of single-family rental options is that relatively few single-family homes have historically been built as rentals. As shown in Exhibit 1, over the past 10 years only 5% of single-family homes were built as rentals — a much smaller proportion than the 18% of existing single-family homes that serve as rentals. The burden to create single-family rentals thus falls on investors purchasing existing homes; however, as described in Pretium’s October Housing Insights the stock of single-family rentals fell during the pandemic despite increased investor activity.³ Build-to-rent capital flows have increased in recent years⁴ and residential land surveys show increased activity in the build-to-rent sector⁵, but even with this increased activity the percent of homes built for rent in the four quarters ending 3Q22 was just 6%.⁶ Overall, the data suggests that there is considerable scope to increase investment in the build-to-rent sector. Not only would this investment broaden access to single-family rentals; but also, it would help to alleviate the long-term housing supply shortage that has worsened housing affordability. 

Growth in the build-to-rent sector would be particularly beneficial in terms of creating new housing supply because the sector has consistently built smaller, more affordable units compared to homes that are built for sale/ownership. As shown in Exhibit 2, the existing stock of single-family rentals is both older and smaller than the existing stock of owned homes. Importantly, homes built for rent in 2021 remain similar in size to existing singlefamily rentals at just over 1,500 sq. ft. By contrast, homes built for sale have become progressively larger over time to the extent that recently built homes for sale are 17% larger than existing owned homes. In other words, even as worsening supply constraints over time have prompted homebuilders to build larger homes for sale, the build-to-rent sector has maintained its focus on creating the affordable home supply that is in greatest need.