Excess Spread — Mashed Swede, drop it, exchange offer
- Owen Sanderson
If you enjoy our content, see what else we cover on structured credit here.
Mashed Swede
The troubles of Europe’s largest debt collector Intrum have been taking up lots of time at 9fin HQ — last week’s announcement that it had appointed Houlihan Lokey and Milbank to advise on debt structure options sent the bonds off a cliff (the 2027s and 2028s are trading in the mid-50s).
Here’s a quick primer on the situation (ping me if you want a copy) and a follow-up on next steps, but at a high level, we’re talking about not enough assets and too much debt — Intrum has roughly SEK60bn (€5.28bn) of gross debt versus book value assets of SEK35bn, and a servicing business which is worth… something.
Whether you value servicing on a doValue multiple starting with a 3x, or a Prelios multiple at 9x matters quite a lot for assumed recoveries!
Anyway, most of Intrum’s routes to more liquidity involve some sort of asset-backed financing. Bondholders and advisors are poring over the capital structure, and contemplating where and how they might be primed.
Would the short-dated 2024s or 2025s take a deal which pushes out their maturities for better security? Would the 2027s and 2028s trade their current time-subordination for structural subordination? Would a credit opportunities fund offer a flexible secured facility that would take the banks out of the RCF.
One solution would be further deals like the Cerberus transaction. Admittedly this probably fueled the currrent run on the bonds, but it still brings cash in the door. This was written up as an asset sale, and indeed it is, but Intrum keeps 35% and the deal is mostly funded by a private securitisation from Goldman Sachs.
Large portions of the Intrum portfolio are held in Irish section 110 vehicles; selling 51% equity plus raising securitised debt basically gets Intrum access to more off-balance sheet, non-recourse funding.
There’s already about SEK60bn of off-balance sheet debt in Intrum’s existing joint ventures; these are mostly senior tranches of GACS or HAPS transactions, where the actual Intrum tickets are tiny. If we consider deals with Piraeus Bank, Project Phoenix and Project Vega I, II and III… these have gross book value of €4.8bn (Phoenix) and 1.92bn (Vega).
Intrum paid €15.7m for the Vega mezz (nominal value €64.2m) and €11.9m for the Phoenix mezz (€54.6m notional). The junior notes are basically out of the money on day one, and so Intrum paid €300k for the junior notes in Vega (€1.5bn notional) and €100k for Phoenix (€396m notional).
One can distinguish the southern Europe “stock” markets (a big pile of outstanding NPLs, often secured, being worked out over time until they’re gone) from northern European flow markets (relationship-driven ongoing offtake agreements with lenders to handle their bad debt).
These have different characteristics, and the natural buyers and approaches will be different; other debt purchasing firms are better set up for the short-term servicing-intensive unsecured portfolios, while funds suit the lumpy secured books better.
Selling or securitising assets will mean engaging with clear-eyed NPL specialists for deals potentially at big discounts to book value. The advantage of an uptiering transaction, from Intrum’s perspective, is that bondholders may implicitly attach more value to these assets (they might even value them at ERC!). They’ll be happy to get security of some kind (all the bonds are currently unsecured) and won’t dig too hard into the portfolios.
This is a fast moving situation with a lot of potential securitisation-like solutions; Intrum has a call with bondholders on Monday, so watch 9fin for more updates!
Drop it like it’s hot
My old shop’s Most Innovative Deal of the Year (brought the market forward the most) has now been tested!
The Stocking of Nationwide’s Silverstone 2023-1 RMBS bonds was late last year, and now the Dropping has begun, with the placement of the £600m class 2A (4.8 year) bonds this week.
This was (out of four) the tight UK prime placement of the year, at 48bps — Perma and Holmes at 55bps in five years, Lanark at 50bps in three years — but, given the credit rally since early January elsewhere in the capital markets, I think you’d have to say prime RMBS triple-A is a pretty poor way of participating.
As signalled at the start of the programme, the approach to market was remarkably similar to any other prime RMBS deal — four lead managers attached (BNP Paribas and Lloyds are the shelf dealers which missed out this time), a full bookbuilding with announcements, updates, disclosed books and so forth.
Given the April call date of the placed 2A tranche, it seems likely Nationwide had always been eyeing this approximate window, and the finger-in-the-air 50bps coupon was a remarkably good guess, allowing the bonds to be priced with just a sliver of premium.
So while this is a textbook prime RMBS placement (1.8x done at the second update, a tiny touch of sensitivity pushing through the 50bps barrier), in a sense it doesn’t test either the advantages or the potential downsides of the “stock and drop” approach.
With the market open all year, there’s been no real need to be nimble so far in 2024, and the original placement plan was readily available. There was little doubt that investors would take down the paper, but the only minor wrinkle would have been a big price move that left the bonds pricing at an appreciable premium; all else being equal, senior securitisation investors would rather buy par paper.
Nationwide’s planned £2.9bn purchase of Virgin Money has much to interest those who follow UK financial institutions (the press seem to be screaming for the mutual’s members to get a vote) but from a securitisation perspective, the main point of interest is that it could further shrink the pool of UK master trust issuers.
I can’t remember who first used the “nuclear power station” analogy to describe a master trust (it may well have been during discussions about Stock and Drop!) but it’s a reasonable one — they’re very big, expensive to run, and you don’t decommission them lightly. But realistically, each lender only needs one; I can’t see why a bulked-up Nationwide would keep the lights on for Lanark long term.
But with central bank funding rolling off, prime RMBS costs close to covered bonds, will we ever see Barclays dust off Gracechurch? The UK bank has a new “head of securitisation and principal funding”, and high conviction on the use of securitisation for risk transfer and deconsolidation trades, and it’s had a tantalising prospectus for a Coventry-style master trust sat on the IR site since 2020.
Exchange offer
It’s going to be a big year for CMBS extensions, restructurings, amendments and all manner of can-kicking activity — and Italy is going to be at the heart of it. Some would argue that you can’t really enforce in Italy, or at least it’s incredibly slow and painful, which cuts off the nuclear option; it’s basically about balancing the economics rather than beating up on one side or the other. How much maturity extension, how much margin increase, how much protection against cash leakage?
The two large and potentially stressed situations are Blackstone’s Pietra Nera Una, backed by four outlet centres across three loans, and Blackstone’s Deco 2019-Vivaldi, also backed by fashion retail outlets.
Pietra Nera Una’s loans were granted a one year extension from May 2023 to May 2024, and there’s an exploratory sale process underway via Eastdil. LTV is over 80% on two of the loans, so what’s the backup if a sale can’t be closed at an acceptable level?
The loans in Vivaldi mature in August 2024 and haven’t yet been extended. At 77% and 74% LTV, they’re a little more refinanceable, though the Franciacorta loan has breached its 75% cash trap threshold.
By far the best-looking Italian CMBS when eyeballed from the outside is Erna Srl, a deal backed mostly by Telecom Italia exchanges, plus a grab bag of associated offices and warehouses. The other big tenant is Enel, which also offers a stable anchor. It was originally sponsored by TPG Sixth Street, through the TAO fund which went with Sixth Street in the split.
There’s a bit of a story in the background, in that Telecom Italia (TIM) is selling off its fixed line infrastructure (NetCo) in a massive €19bn disposal to KKR. The deal hasn’t closed, and negotiations have been tortuous, with the Italian government and Vivendi, the largest TIM shareholder, sticking their respective oars in. But it seems logical that telephone exchanges would form part of “fixed line network infrastructure” going into NetCo, and it also seems logical that KKR would be seeking to drive efficiencies — one of the biggest “efficiencies” probably comes from downsizing the property portfolio.
We’re more of a securitisation newsletter than a telcoms technology newsletter, but we’re pretty sure that big exchange buildings aren’t as necessary as they once were to keep a fixed line network humming… so could cutting the estate be on the cards? The leases are long but is there a route to wriggle out?
LTVs on the two largest loans are in the mid-30s, while the smaller loans are even less levered. But that doesn’t mean that a bit more time isn’t useful.
Sixth Street would probably like a more comprehensive refinancing to take leverage right back up again. The assets are infra-like and low returning (around €30m a year, vs a current valuation of €490m) so you need a lot of (cheap) debt to make them exciting again.
But in the short term, it’s sponsor vs bondholders. An initial proposal early this year included a margin bump across the notes and on the loans, plus partial prepayment, surplus cashflows diverted to paying down the notes and so on.
This apparently wasn’t good enough, with the planned meetings at the end of February called off and certain changes and clarifications proposed; the new proposals clarified that the class A notes would get 275bps instead of 250bps, and the class B would get 500bps instead of 425bps. The class C notes will get paid down entirely under the new proposals; ordinarily this would tend to annoy senior bondholders, but LTV is very low already, and I guess it means there’s more spread available in the deal? One wonders if the class C notes were making a lot of trouble for small bond position.
Catalogue of errors
The Very Group (Shop Direct) has been a name regularly coming across the desks of 9fin's distressed debt team, especially since the troubles of the Barclay brothers started to hit the headlines (except in the Daily Telegraph, which they owned). Here’s our latest piece on the bonds plunging on Wednesday after auditor Deloitte resigned.
The family had raised financing at a holding company level, tying the fortunes of a money-losing broadsheet (which happens to be the house rag of Britain's ruling Conservative party) to a football gambling-turned-department store-turned-online shopping-portal-turned unsecured lender, in the shape of The Very Group, and Britain’s sixth favourite online delivery company (Yodel).
The Very Group started life as Littlewoods, a now-defunct department store, which was combined with home shopping service Kay’s Catalogues, but makes an increasing proportion of its profit from financing activities rather than shopping.
The group’s consumer financing is mostly funded through a large private securitisation facility, Securitisation of Catalogue Assets, with senior notes subscribed by banks including NatWest, RBC, Deutsche Bank, Barclays, and HSBC, and mezzanine notes from Insight and CarVal.
The issue for the poor high yield investors buying Very Group's "Senior Secured Notes" (due 2026 and trading in the mid 80s) is that these have conventional HY security packages (share pledges, opco bank accounts, opco guarantees and so forth) but are basically subordinated to the whole securitisation, and dependent on whatever interest flows out of the bottom of the securitisation structure. There’s a substantial retailing business as well, but with a high cost base; without the money dropping out of SOCA, there’s not much coverage left for the bonds.
This subordinated opco bond situation isn’t unique; it applies to NewDay as well (out in ABS this week with a very well oversubscribed deal) and Together (which announced it was switching some property development lending out of the holdco bonds into a dedicated property development securitisation). But these have scale and dominance in their particular areas of activity, and it’s hard to say what Very’s edge is.
Like every specialist lender, it has been struggling with rising rates. High cost unsecured credit doesn't necessarily move with swap rates the way mortgages do, but Sonia increases pass straight through to the bond coupons. The higher the rates initially, the less this matters (so NewDay has coped well) but less money dropping out of the securitisation is straightforwardly bad news.
You can question the performance, but the particular noise about the company stems from the ownership structure. Lloyds, which lent to Barclay family holding company, put the Telegraph into receivership last year. Rescue financing in the form of a convertible from Middle Eastern-owned Redbird IMI may allow them to get it back, though this required pledging stakes in the rest of the Barclays businesses, including Very.
Very itself raised a £125m “funding package” from Carlyle and IMI earlier this year. Since this came with board seats and without affecting the existing bond security package, one assumes it came in a subordinated level.
So you’ve got a real tower of capital here. Class A notes, mezz and junior in the private securitisation, opco secured bonds, something subordinated at Very (no details out there), common equity, holdco lending. Refinancing £575m of 2026s at 14% won’t be easy, but will this matter for the securitisation? Very’s short-dated lending should roll off fast, and there is, in theory, plenty of equity in the securitisation. An originator blowing up can always cause some trouble, but at least securitisation lenders actually are “senior secured”.