But Cerberus hasn’t exited everything CHL owned. It sold the CHL back book to Bank of America, which printed the Auburn 15 deal before Barcelona to finance these assets.

CHL had resumed new origination in 2021, when all seemed right with the world, and procured warehouses for this lending. As Cerberus looked to exit, it ran a sale process for this recent-vintage origination, likely targeting asset hungry deposit-takers.

Chetwood, which bought the platform, would be a likely target — it’s bought a ton of buy-to-let from LendInvest — but seemingly there was no meeting of minds on price, as Cerberus retained this portfolio, and has since financed the book with Edenbrook Mortgage Funding (named for a country park/village near Fleet).

Cerberus has a bit of a reputation for missing calls, especially on the CHL back book deals under the Towd Point-Auburn brand, but the market punishment meted out here doesn’t seem too egregious.

The standard specialist lender step-up is 1.5x capped at 100bps, as it is in this case, and there’s a turbo after the step-up date, which is also not unusual — LendCo’s Atlas 2024-1, an ongoing shelf rather than an orphan pool, has the same mechanic.

Atlas does not, however, feature a Mortgage Pool Auction mechanism, more regularly seen in NPL/RPL or legacy deals; if the deal isn’t called in four years, the pool gets offered out.

Cerberus is keeping the class X, the class Z and the vertical risk retention. But the Edenbrook deal does take essentially all its money off the table, including the remaining origination pipeline through August (thanks to a prefunding facility), so it’s not bad going for a back book — and potentially a better exit than would be achievable selling a portfolio in whole loan format. Securitisation markets continue to be kind to Cerberus, calls or otherwise.

There’s probably some spread pickup to reflect the deal’s unusual genesis, but you’d have to squint hard to see it. Senior notes came at 87bps, versus 85bps for Atlas at the end of May, and 80bps for the STS-designated Lanebrook in early May. Heavy supply over the last month or so is more than sufficient to explain a 2bps pickup, though the majority has been in euros; sterling buyers may feel under-served.

Preplacing the senior notes, with reverse interest in the loan note and in the bonds, undoubtedly helped, while any mezz with a little juice tends to fly out the door. Small tranche sizes help, but we wonder how much of the “asset-backed private credit” crowd are tempted to do a bit of public mezz while they queue up to gain access to scarce and hard-to-find private opportunities.

Given Cerberus’s recent history with the CHL assets, we assume they’re open to bids on the residuals; Edenbrook isn’t even “Edenbrook Mortgage Funding No. 1”, so it’s a one-and-done walk away type trade.

Chetwood itself has dabbled in securitisation, but only with the retained “Chetwood Funding 2024”, while Fleet Mortgages hasn’t sold into a deal since the Starling takeover. The Hampshire securitisation market will be sleepy for a while.

CLO structures are bad

With my levfin hat on, my team have written lots of content about the Vodafone Spain deal, the biggest LBO in Europe for more than two years, with €3.4bn of distributed debt across bonds, loans, euros and dollars.

The secured debt is rated Ba3 / BB / BBB-, pretty close to the top end of the typical CLO investing universe, so buying this deal would tend to enhance the weighted average rating factor (WARF) metric for most deals. Even better, the price talk looks quite generous, at 400-425bps and 99.5 across euro and dollar loans — that’s cheaper than the B3-rated generics company Stada’s recent A&E.

So this is an obvious win-win on the basis of two standard CLO tests, on which manager style and performance are often plotted. But it is cheap for a reason!

The credit is a turnaround story, an unwanted subsidiary that Vodafone didn’t want to fix itself, and it’s squeezed by a highly competitive market. The buyer, Zegona, put in very little of its own equity (just €300m for the €5bn transaction), and the purchase multiple was lower than almost any other European telco acquisition. If it was levered to fairly standard telco levels (MasOrange, the biggest player in the Spanish market has 4.7x total leverage), it would already be underwater.

It is not necessarily a credit to back up the truck on, despite the compelling rating vs spread calculation.

The tension between the guardrails built in to CLO documentation and the actual job of buying the good loans and not buying the bad ones is well understood — which is part of the reason PMs and equity investors often kick against the restrictions — but what’s the best way to look through this into the real elements of management style?

It seems great to be a manager that’s squarely in the high WAS / low WARF quadrant of the market (Deutsche Bank’s CLO manager barometer for June shows Tikehau, Chenavari, CIFC, Sound Point and Oaktree among others). But as the Vodafone Spain deal underlines, one way to achieve that positioning is to buy “cheap for a reason” transactions.

The Deutsche data also shows the majority of larger managers are not in this quadrant. Most of the 15+ deal managers, the likes of Blackstone, Carlyle, CVC, Investcorp, Barings are in the opposite, high WARF low WAS situation. The portfolios are (by rating) riskier and they’re not even getting paid!

This could be for several reasons. Perhaps big managers with big brands don’t need to try so hard, as their deals get done on brand name? Perhaps it’s impossible to be nimble and cherry-pick deals when you have €10bn of European loan AuM?

Or perhaps these managers have the track record that means they feel happier ignoring crude metrics like WARF/WAS, and concentrating on “you get what you pay for” quality credits?

There are 10 types of people….

…those who understand binary and those who don’t, as the old joke goes. But which investor or investors understand the potentially binary outcome for the E-MAC NL shelf?

This is a legacy Dutch nonconforming RMBS shelf, originally part of the GMAC empire and now sponsored indirectly by Fortress, via CMIS. It’s also the subject of some high value litigation between NatWest, the hedging provider, and CMIS, with NatWest looking to claim more than €150m relating to swap payments.

We discussed this fight last year, and the legal situation hasn’t moved on massively. NatWest filed an updated claim form on 20 June, which removed some of its claims over Subordinated Step-Up Amounts, but didn’t change much else other than updating the figures.

The E-MAC shelf has basically been abandoned for more than a decade, despite the Fortress ownership. Deals were supposed to be redeemed via an investor put option, but the issuers were never granted the servicer advances required to exercise the put — Fortress had no interest in throwing more money into the structures. With low yielding interest-only collateral, the bonds have paid down very slowly, and have flipped back and forth between sequential pay and pro-rata depending collateral performance.

The court fight turns on subordinated swap payments.

Instead of a balance guaranteed swap coming out senior, there was a swap based on estimated amortisation, which was supposed to be periodically rebalanced by (subordinated) Notional Adjustment Payments.

After the deal was supposed to have been redeemed through the puts, there was another mechanic allowing the hedge counterparty to take the swap price up to market rate, with payments also accruing at the bottom of the waterfall.

There’s not enough money coming through the deals (and so much interest deferral) that these payments are unlikely to be made if they’re back-dated to the redemption of the deals, or if the plan is just to wait for the deals to catch everything up. The docs basically never envisaged this! But NatWest and CMIS are fighting over an indemnity allegedly given by CMIS to protect these payments.

If it turns out CMIS is on the hook, this is very bad news for the deals. Without Fortress’s support, the award would bankrupt CMIS at a stroke, and the last thing these need is an absent servicer.

Anyway, there’s been some unusual movement in the transactions — Fitch upgraded them on June, taking the class B and C up from BBB+ and BB+, respectively, to A+. This may have prompted a longtime legacy account (potentially an owner of the bonds since before 2008) to put a bit of the E-MAC mezz out for BWIC, apparently hoping for a price in the 90s.

Bids have been all over the place, as befits a trade with a heavy duty lawsuit hanging over it. Even if you judge that CMIS has, say, a 65% chance of success, how do you trade that? The bonds are either great or they’re gone. The expected value might be positive but you can only roll the dice once.

The covers were, respectively, 75 and 59 for the class B and C; a bargain if NatWest loses, cash in the bin if it wins. The covers could be some way above the actual trade level; if just one investor out there had strong conviction on the outcome of the lawsuit, there was a great opportunity here.

Want to get market wraps, breaking news and in-depth analysis ahead of the competition? Request a trial of the 9fin platform below.