But now we’re in the middle (or maybe at the end?) of a good old fashioned banking crisis, what does this mean for bank investment? 

Enra’s Elstree No. 3 cleared the market in last week’s brutal conditions, and the reverse enquiry bank bid for triple-A not only stayed in place but held to its level of 125 bps. It seems unlikely the same account is good at 125 again. 

Not only is that just palpably not where the market is any more, but the turmoil messed up the inputs.

Bank investors don’t just eyeball their deposit funding and try to earn a spread above it; big bank treasuries tend to have sophisticated models for the cost of funds they pass through to front office or investment desks, incorporating term liquidity premium, the bank’s own cost of capital and cost of risk and so on. 

So, TLDR, a crisis that spikes bank CDS and draws the focus to bank liquidity buffers tends to discourage bank investing in securitisation. It doesn’t necessarily mean banks will sell their existing portfolios, as this funding may have been locked at the time of investment, but it might mean a smaller bank bid going forward.

Perhaps issuers can accept levels going up, but volatility in rates and credit spreads makes it hard to pin down just what that level is, especially if you’re trying to lock in an anchor account ahead of a longish execution window.

Medium term, as we wrote last week, a surge of deposits into the big US money center banks could be positive for CLO investing, but it might take a while to get there. My intuition that the regionals weren’t huge in CLOs looks right, at least — The excellent team at Barclays credit research put some numbers around it.

Regional banks as a whole had $12bn in CLOs, split $1.8bn in Held To Maturity (HTM) and $10.58bn in Available For Sale (AFS). The big banks, however, are….bigger. Columns are HTM, AFS and Trading assets.

The strange parts here, to be honest, are the somewhat underpowered presence of Bank of America (the treasury certainly has been seen in the market in the past), and Barclays (also a buyer, though potentially this comes from a non-US entity and isn’t recorded). 

Perhaps Morgan Stanley’s traders are exceptionally good at closing books into year-end, or they’re somehow running their inventory through a different entity — either way it’s generally reckoned a decent CLO shop and we assumed it would have more than $1m on hand to trade.

Anyway, we digress. Conditions in CLOs are not especially attractive right now, with spreads blown out to almost the depths of 2022’s gloom at the back end of last week and this. To borrow Bank of America’s account for last week:

Spreads widened across the capital structure, the CLO credit curve steepened, and its curvature increased: AAA headline spreads reached 225 bps, an increase by 40 bps. BBB bonds widened the most on average, finishing the week 90 bps wider, at 650 bps. Sub-IG spreads widened by around 75 bps, with BB and B headline spreads reaching 1,000 bps and 1,450 bps, respectively.

CLO AAA and B headline spreads are just 25 bps and 50 bps below their 2022 peaks, respectively. We also note that trading ranges for each rating bucket have increased, as expected when the backdrop deteriorates.

Cracks are appearing in loan pricing — Albea Beauty’s A&E transaction is generally reckoned decent from a credit perspective, but the price talk of 95-96 was very much a “no banking crisis” level, and the deal came into land at 94 on Tuesday. That’s disappointing for sponsor PAI Partners, but it doesn’t signal a flood of cheap loans disgorging onto the market, or a meaningful improvement of the CLO arb conditions.

Once again, CLOs were on the wrong side of the market, widening faster than loans when the going gets tough. 

That’s rather less true in the US, though, where contagion worries about various regional banks are still running high. It’s kind of amazing and ironic that Atlas SP Partners (the recently separated Credit Suisse securitized products group) is now already in the business of trying to rescue banks itself, as it’s reported to be in talks with PacWest.

US loans have slid more, while CLO spreads are stronger, opening the way for “print and sprint” deals, as my US colleagues wrote this week (Ares and CVC are in the market). This has never really worked in Europe, though CVC is one of the only successful examples…..European loans remain too illiquid, too sensitive

On the topic of bank investments, we hear Standard Chartered has been selling quite a few bonds this week (some decent BWICs at the senior end of things). You might think this is not a great backdrop to be disposing of assets, particularly as StanChart’s most active buying periods suggest a lot of the portfolio is underwater, and you’d probably be right. 

Two observers linked the sales to the arrival of ex-Deutsche Bank trader John Parker at the firm to run global securitized products trading. Parker doubtless has assets he likes and assets he doesn’t, and may want to stamp his mark on the business. Equally, it could be a pricing exercise, making sure the marks on the StanChart book are sufficiently linked to reality.

Bad = good

The Significant Risk Transfer market is a strange and counter-cyclical beast — tenuously tethered to liquid credit, but often open when other capital markets are closed, and benefitting from problems elsewhere.

Fundamentally it’s a tool for banks to raise regulatory capital, so the driver of issuance is capital pressure and the cost of using equity or AT1 to plump up the relevant ratios. Regulatory shifts like IFRS 9 can help SRT supply by squeezing bank capital. But the driver of pricing tends to be other forms of levered corporate credit risk, such as credit index tranches or CLO equity, since these are closest to the risk profile of the large corporate and SME deals which account for the bulk of the market.

The past couple of weeks have been rough for bank stocks, and rougher still for AT1, as the market digests the implications of the Swiss regulators’ decision to torch Credit Suisse’s entire AT1 stock. The Bank of England and ECB both released statements suggesting that they’d act differently in resolution (and make sure equity was toast as well) and furious arguments have erupted over the rights and wrongs for the Credit Suisse bondholders. Litigation is nailed on; credit claims on the zeroed bonds are trading with non-negligible option value, so clearly some see opportunities here.

But the point for SRT markets is that AT1 markets may be difficult or complex for banks to access in the months ahead, but for banks that want to crank up their ratios, SRT is still there, so there should be more issuance. 

Robert Bradbury at A&M thinks this makes sense; my former colleague Richard Metcalf has written on this, and it was a huge topic at the always-excellent IMN Significant Risk Transfer conference. At least one multi-strat fund we met had shown up on the day hoping this trend would come to pass (and drive some price widening). 

There are some practical difficulties. Every big bank is set up differently, but it’s not necessarily the case that the SRT-issuing part (often credit portfolio management or similar) sits close to the AT1-issuing part (often treasury). For those which have an investment bank arm which self-arranges these deals, securitised products, handling the SRTs, may have a different P&L and management structure from the FIG group that runs capital products like AT1.

Credit Suisse, though, will doubtless be missed in the risk transfer market — it was both extremely active and extremely innovative. It’s done standard large corporate deals, but a plethora of other structures and collateral pools as well, including counterparty credit risk, income-producing real estate, SMEs, sub-IG mid-cap loans, as well as, of course, the infamous “yacht securitisation”. It had three active shelves, US leveraged finance, Asian corporate loans and DACH loans as well, plus some more bespoke trades.

A quick look at Credit Suisse’s Pillar 3 report shows CHF39bn of senior exposures to synthetic securitization with CS as originator or sponsor, plus another CHF1.5bn of higher RW exposures and a few snippets of capital deduction (likely the first loss / expected loss piece of synthetic structures).

If we assume that this is hedging around CHF45bn of notional, and the placed tranches are about 7% of the capital structure, then that’s about CHF3bn of Credit Suisse’s portfolio equity risk placed in the market…coincidentally, about the price that UBS bought the whole bank for on Sunday.

UBS is a much lighter user of the technique — there’s nothing at all in the “UBS as originator or sponsor” line. SRT sceptics might point out Credit Suisse has been just that bit too clever in the past (consider structures like paying bonuses through the Partner Asset Facility structures, or that yacht deal) and this general pattern of over-fondness for financial structuring is part of what messed the bank up. 

But you could flip it around, and consider the fact that Credit Suisse is the only banks to successfully package and sell the risk of its own screwups, in the form of a catastrophe bond referencing operational risk capital (Operational Re). Perhaps the bank would still be with us if it had done, like, 10 yards of this.

So can the SRT fans at Credit Suisse persuade their sceptical new colleagues? Or are they all looking for a new position in the near future?

In the broader SRT market the main fallout from the Credit Suisse failure was a massively increased focus on bank counterparty risk.

Many of the banks issuing SRTs are large G-SIFI type institutions (like Credit Suisse!) which have been making their shelves more and more standardised and efficient over the past few years, and along the way using unsecured credit-linked notes structures as standard. This is cheaper than using collateralised CLNs, and much cheaper than a full scale orphan SPV type setup. 

But it does have a downside! As well as giving exposure to the actual reference portfolio through a derivative or financial guarantee, it’s a senior unsecured obligation of the bank. SRT investors are happy to take subordinated leveraged risk on asset portfolios; they’re not necessarily keen to layer counterparty risk into the mix as well. 

The question of whether CLNs are subject to “bail-in” or resolution process is difficult and inconsistent. If Credit Suisse’s downfall is any guide, you can read the regulations all you like but they can be subject to last minute changes at the whim of the regulator. Bailing in CLNs is probably painful and barely worth it in a bank resolution, but you never know.

Smaller banks haven’t necessarily gone down the unsecured CLN route, and, in general, the precedents are good. Getin Noble Bank in Poland has gone into resolution, and has an outstanding SRT which is apparently unaffected; it’s perfectly possible and indeed desirable to build these deals to withstand bank failure (they generate bank capital, and what else is bank capital for?)

There are various structural bells and whistles that help, such as ratings triggers (actually, nobody thinks these help), CDS spread triggers (somewhat more useful), and, best of all, collateralizing the deal. Historically, according to big bank issuers, investors in the sector would rather get paid more coupon that have these extra security measures, but perhaps that’s changing.

Certainly based on our conference conversations investors want to clean out counterparty risk, and claim they’re willing to pay for it. Funds do want to win deals, and do want to be commercial, but there’s been a real shift; the specialist asset managers in this sector want to be able to say “bank capital but no bank credit risk” to their LPs at the moment. 

The big funds are willing to beat up a little on their counterparties….as one substantial SRT investor told us, “if a big bank says they can’t do it, just insist. They don’t like it, but they will and they can.”

Call of the mild

Last week saw more extension risk stalking the land, with a missed call for Precise Mortgage Funding 2018-2B, an RMBS issued by Charter Court Financial Services. Now, Charter Court itself (now part of OneSavings Bank) should not be blamed for this directly, as it sold the residual notes to a third party in 2019 (Bank of America handled the distribution).

You can, perhaps, blame CCFS just a little — it’s a very obvious illustration of the “nobody knows about the equity” issue we talked about the other week. Pore over the loan data as much as you like (TLDR performance is good) but if you don’t know the incentives and behaviour of the sponsor, you don’t know the value of the bonds. 

Perhaps you bought into the deal thinking that front book challenger bank deals always get called, and then, a year later, found yourself in a hedge fund/sponsor controlled transaction. I mean if that bothers you, sell the position, you’ve had four years. Or better yet, examine the structure for risk mitigants and make sure you get paid (as the PMF bondholders seem to have done). 

PMF 2018-2B has a full turbo after the First Optional Redemption Date (FORD) (interest diverts to pay off principal), so it’s not like the sponsor is sitting there rubbing their hands with glee and enjoying the off-market funding they’ve left in place; all the money is going to pay out bondholders and resids are getting nothing. The senior bonds have also amortised down to just £55m, from the original £338m, so it’s not even like it’s efficient funding, turbo or not.

That suggests that it’s more likely to be a non-call which gets remedied sooner rather than later, and hence causes limited ill-will in the ABS investor community. TwentyFour Asset Management and NewDay missed calls in the post-Covid era but caught up on the next payment date. They’ve not acquired a reputation as worrisome sponsors, quite the reverse.

It’s not clear whether the non-call came as a direct result of the market turmoil last week, or just the outright levels that were on offer. PMF 2018-2B got done on very good terms; the senior notes pay 68 bps and step up to 102 bps, so a refi would have come wider, even before the banks started failing. 

The turbo feature means there’s no real advantage in leaving the existing capital structure in place on a coupon vs coupon basis, but given the improving market backdrop, reasonable people might have expected a better execution of a new deal around the next IPD, and a more efficient capital structure for the next five years of the portfolio’s life. The amortisation of the portfolio might also be relevant; the pool is down to £91m, too small to securitise in a distributed deal, so perhaps the sponsor is waiting to combine it with something.

Cerberus, on the other hand, has taken more of a hit from its failure to call Auburn 12 and Auburn 14 on time. As legacy portfolios these amortise slowly, so the funding is still cheap, and the step-ups are low. The 1.5x / 100 bps cap structure is pretty on market, but that still puts a coupon of 135 bps on some non-conforming seniors, and a very modest 310 bps on the class E, a tranche with the magnificently split A / B / B ratings. 

More to the point, as one investor put it, the Auburn 14 deal was a “sell bonds above market value of collateral and walk away” trade, not a “clip the resids and refi” trade. My rough maths gets me to rated notes sold at 97% of pool purchase price on Auburn 14, and 99% on Auburn 12, and one can quibble over the portfolio price in question. Presumably now the pools are now worth even less, as interest rates have risen. Even if the incipient banking crisis gets back in its box and credit spreads tighten, is it worth calling?

Wake up sheeple

European CMBS isn’t big or ugly enough to get any real doomer headlines going (unlike the US!). The looming maturity wall is more like a looming picket fence. 

But even if the issue isn’t systemic, each asset matters to its owners and its bondholders, and sponsors still need to come up with something. We’ve already seen Frosn 2018 flop into special servicing, Pietra Nera Uno hit with downgrades, waivers and amendments galore, and a few deals refi’d out. But to our knowledge, one element in the A&E toolbox yet to be deployed is the discounted bond tender. Until now!

Sponsor Cale Street Investments (backed by the Kuwait Investment Authority) has launched a tender for bonds in Deco 2019-RAM. As the Deco name implies, this was a Deutsche Bank deal from 2019, and the “RAM” part refers to the fact that the asset is the former Intu Derby shopping centre (Derby County Football Club are “The Rams”).

Shopping centre group Intu has had a rough decade or so. It’s difficult to find investors who have anything other than contempt for bricks and mortar retail, especially in the UK, where business rates and labour costs ravaged the retail sector, and the “CVA” restructuring mechanic meant that landlords got hosed when retailers couldn’t service their debts. Intu was on the brink of a rescue rights issue and last ditch turnaround in February 2020, and, well, the deal didn’t go well given what happened afterwards.

Since then the once-proud empire of Manchester businessman John Whittaker (he owns Peel Group, and a chunk of Peel Ports too) has been dismembered, with bondholders taking control of the various shopping centres (Intu had several separate financing vehicles).

The business plan (bit of Capex, new tenants etc) for SGS is here, and Newcastle’s Metrocentre is here. The Trafford Centre, the prized Manchester asset at the heart of group, was taken over by creditor CPPIB in 2020. 

Derby was a minnow besides these, but it was the only Intu asset to be financed by a “true” CMBS with a fully tranched capital structure governed under securitisation regs rather than a secured corporate bond. The sponsor was originally a 50:50 JV between Cale Street and Intu, but Cale Street bought out 100% after Intu group fell into administration in 2020.

The loan backing the CMBS was heavily amended in 2020 and 2021 to handle the impact of Covid, which closed shops, slashed footfall and prompted a shift to online shopping — there were waivers to various covenants, a cash trap, and some prepayments applied to the bonds.

At the asset level, some of the leases have been switched to turnover-based rather than traditional (good for struggling retailers, less certain for the landlord), and valuations have been trimmed. As of last year, S&P’s model gave a £169m valuation, down from an original appraisal value of £351m, and original S&P valuation of £270m. 

There’s only £122m of securitised debt against it, so we’re still in the realm of sub 75% LTV….if the assumptions are good. 

S&P assumed an 8.5% cap rate in the middle of 2022 (bizarrely, the same cap rate assumption as its February 2021 valuation). If we look at Savills market update for Q3 2022, “super prime centre” shopping centres are yielding 7.75%, with prime centre at 9% and “town centre dominant” at 10.5%.

July 2023 is something of a decision point — the covenant waivers and cash traps fall away, though loan maturity isn’t until 2024. But the point is, Cale Street is now attacking the problem with the tender, hoping to spend £50m buying back the notes, and offering 92.5 for the class A and 84 for the class B via Goldman Sachs.

If it’s completely successful and cleans out the expensive mezz first, this would leave £67m of senior bonds outstanding, a very manageable leverage point for a refi or another round of waivers. So the asset is safe!

The downside of adopting this approach is that it’s wildly expensive. Cale Street bought 50% of an expensive asset in 2019, then doubled down taking over Intu’s stake, and is now doubling down again plugging in another £50m. A sponsor backed by a deep-pocketed Middle Eastern SWF is a fine thing for bondholders, but it’s hard to see how all of this translates into much of a return. I would simply not be expecting a small city shopping centre to bounce back to pre-Covid valuations!

If you prefer your CMBS aged like a fine milk, perhaps Windermere XIV is the deal for you. This was a Lehman Brothers conduit CMBS from 2007, which went through the usual round of post-GFC conflict. There was a long fight in which Lone Star was trying to replace the special servicer with its own Hudson Advisors unit, prompting a long round of self-justification from incumbent special servicer Hatfield Phillips. The deal blew through note maturity in 2018 with the assets in the Fortezza II loan still unsold (Italian offices). Now it’s coming up to its revised, five year-extended note maturity in 2023, and the current loan special servicer Mount Street wants to contact investors to discuss this upcoming deadline. 

We raise this largely because it’s astonishing how long these transactions can drag on when they go wrong — long-suffering noteholders saw their holdings downgraded to CC or below long ago, and may well have written off recoveries from the positions, but still they are dragged to noteholder meetings to grapple with how to finally put this to bed.