Excess Spread is our weekly newsletter, covering trends, deals and more in structured credit and ABS — subscribe to this newsletter here.

JDGAF arbitrage

Celeste has been writing about the ever-controversial topic of SRT leverage lately, picking up some of the UK bank responses to the “Dear CFO” letter in April.

Our views on whether SRTs should be levered through repo financing are quite relaxed — have a listen to our podcast here dubbed “SRT leverage is fun and cool” — broadly on the grounds that the financial system is quite happy to lever all sorts of comparable risk to the hilt.

What’s the implied leverage you can get doing bank equity options? Is it more or less aggressive than the relatively low LTV facilities financing SRT positions?

A regulatory policy saying “regulated banks shouldn’t leverage the risk-bearing parts of other regulated bank capital structures” would be consistent (so strike out equity leverage and AT1 leverage) but it seems unfair to pick on SRT specifically.

But the rights and wrongs of the matter don’t really matter — in the actual world where we live, there’s clearly a lot of scrutiny on the topic, and frankly, for some banks, it’s just not going to be worth having the fights required to keep doing it.

Take Barclays, formerly a big player in the market, but with a SRT repo book that’s now, according to a spokesperson, “de minimis”. To pick some imaginary numbers, if it had £500m or so outstanding, paying 200bps, that’s an ok earner for Barclays SPG (better than buying senior bonds), but that’s going to be £10m a year after funding costs?

So it’s slightly weird that equity analysts are asking about it on the Barclays earning call.

Barclays group did £7.7bn in revenue last quarter, for £2.7bn profit before tax. Even just considering global markets financing, we’re looking at £800m per quarter. The presence or otherwise of an SRT financing business is quite literally a rounding error.

Which is exactly why banks might be willing to pack it in. Institutions that provided it as a broader client service, giving leverage terms on whatever Apollo, Blackstone or Ares might want to show them, don’t need the brain damage associated with SRT financing specifically.

But different kinds of financial institution exist.

Banks like the big UK clearers are very sensitive to the way the regulatory winds are blowing. Top of the house doesn’t like awkward questions from the PRA concerning a business that’s basically tiny, and the same is true for institutions like Deutsche Bank with respect to the ECB.

Some fights with the regulator are worth having, and DB went to the mattress over the leveraged loan underwriting capital add-on. Some aren’t worth it.

But where there’s a business that’s worth doing, and incumbents get pushed out by the regulators, there’s an arb available. Specialist brokers, banks whose primary regulator sits outside the UK and Europe, or even non-banks are likely to play a bigger role in this area, as are leverage structures which attract less scrutiny.

Should banks be allowed to provide fund-level leverage to the biggest multi-asset credit funds out there? Of course they should! Do these funds buy SRT in their broad mix of other credit exposures? Sure! But it’s an area which attracts less regulatory heat, with less mark-to-market and a less obvious loop returning risk to the banking system.

Banks bid better

The story of Finance Ireland’s retreat from mortgage lending is really a story about bank competition.

M&G, which had an extensive forward flow to fund Finance Ireland’s mortgage origination, funded these loans in the capital markets, through Finance Ireland RMBS No. 1 to 7, a well-liked and well-followed shelf giving investors much appreciated diversification into prime Irish home lending.

But overall cost of funds simply couldn’t compete with Irish banks’ deposit funding once interest rates started to rise. Passing the higher costs of securitisation exit through to origination yields simply meant that origination dried up, and Finance Ireland eventually shut its mortgage lending business earlier this year.

With the flow shut off, M&G also sold off the outstanding residual notes in Finance Ireland RMBS No. 4-7. The collateral was well known and attractive, and the usual securitisation suspects looked at the deal, but, we understand, a bank was by far the best bid.

That’s relatively unusual in a securitisation context, because holding residual notes in an existing securitisation attracts punitive capital charges — straight 1:1 capital deduction is usual.

That’s not an unreasonable regulatory position, given the structural leverage attached to a residual note, but it’s generally more attractive for a bank to hold whole loans (because a bank can be even more structurally levered than a securitisation).

But the sale of the mortgage portfolio can be structured around this issue. Finance Ireland RMBS No. 4 was called on Tuesday, and the call notice explains “Funds advised by M&G (collectively, the “Option Holder”) will provide irrevocable notice to the Issuer that it elects to exercise the Call Option and that its nominee will act as Beneficial Title Transferee to purchase the beneficial title to all the Loans and their Related Security comprising the Portfolio, in consideration of the payment of the Optional Purchase Price to the Issuer.”

In plain English, M&G is still holder of the residual notes but has nominated the bank buyer of the portfolio to purchase the loans out of the structure, having presumably secured a fee related to the price of the residuals in the auction process.

There are three more Finance Ireland RMBS deals to call, with No. 5 coming in March 2026 and 6 and 7 in September 2026

In the post-crisis period, foreign banks fled the Irish mortgage market, selling off assets to investment bank principal desks, private equity, hedge funds and more. But now the flow has started to reverse.

Foreign banks building up in Ireland include Bankinter, which is writing loans under the Avant brand, and Austria’s BAWAG, which is in the middle of a European acquisition spree which saw it buy Barclays Consumer Bank Europe (a German credit card lender with about €3.4bn in assets) and Knab (a Dutch digital bank with €12.7bn in mortgages). BAWAG set up a branch in Ireland in March 2023, and has been writing loans under the MoCo brand. It also owns Depfa Bank Ireland, but appears to have opted to passport in and originate through the branch instead.

Most intriguing is Nua Money, which set up in Ireland in 2021 and started lending in 2024. A quick perusal of an Irish mortgage comparison site shows this is the sole non-bank currently in the mix (3.6% for a three year 60% LTV), which naturally raises questions about how it’s funded. Poring over the Irish corporate registry turns up Churchill Funding No. 1 DAC, a facility in which Barclays is class A noteholder.

Barclays isn’t a charity, and one presumes it would give Nua similar leverage terms to any other Irish mortgage lender or owner of mortgages, including M&G. So there’s clearly some magic here allowing Nua to compete; if you know, write in!

Inefficient 18 year olds

The 2008 crisis is a long way in the rearview mirror, and the world has mostly moved on. But some of the RMBS deals are still outstanding, and they don’t seem to be going anywhere, just amortising on to legal final. If a borrower couldn’t refinance at any point in the last 20 years, what would change that in the next five?

But there are (in theory) trades to be done. Market participants have been eyeing up these opportunities for a long time, but the problem is in actually executing anything. Pre-crisis non-conforming deals could often switch back and forth between sequential payment and pro-rata, which would have large effects on the redemption profile of the mezz.

If you could, for example, identify a deal which economic tailwinds would lift back into pro-rata pay, scoop up the mezz which was trading to full extension under a sequential scenario, and book the profits when the switch came, that’d be a nice trade. The trouble, in practice, was that it was hard to put this on in meaningful size. Many of the bonds are tightly held, by accounts which aren’t terribly engaged, and the tranche sizes, never huge, had amortised down further. Bonds that were loose were often in fast or smart money hands trying to work the same angles.

As time rolls on, more strategies become economically viable. Call and refinance still looks expensive on a pure coupon basis, but works given bond trading levels, and will flush out non-amortising reserve funds which might constitute 10% or more of the portfolio outstanding.

Then there are clean-up calls, which aren’t present in all deals of this era, but which in theory allow the servicer, directed by the trustee, to call a deal, and if it can’t fund its own call, to auction off a portfolio.

But again, there are several problems in actually putting these trades on. Call rights, in the form of residual notes, are hard to find. The crisis left hedge funds and investment bank securitised products businesses devastated, and the institutional memory of who owned what was shattered. Whole portfolios were dumped in bad banks or run-off vehicles, with a trader or two to unwind the worst stuff, then more or less abandoned.

Even if the call rights can be collected, size matters. Many of the deals are a fraction of their original size, and much too small for a standalone refinancing. The only funds with the expertise to play this kind of deal have better things to do with their time and their capital. It’s not something you could raise money for; it’s a trade or two, not a strategy.

But it’s an interesting example of the long tail of market inefficiency caused by the sharp break of 2008. There’s money here waiting to be picked up by someone, it’s just that actually picking it up is too difficult.

Toon and out

It’s a good day when there’s a debut issuer in the market.

2024 saw record numbers — 15 new issuers in European securitisation, across seven asset classes and three geographies — and some of these have already been back in 2025, for example Quantum Mortgages, LiveMore Capital, Plata Funding, and Vantage Data Centers.

That naturally makes it harder to surpass the debut issuer total in 2025, since the pool has to keep expanding, but it’s been a notably good year for the builders, with Nottingham and now Newcastle Building Society making their first forays into RMBS.

Newcastle Building Society’s Hadrian Funding 2025-1 was priced on Tuesday via arrangers BNP Paribas and BofA, hitting 50bps for a £650m senior tranche (of which £300m was retained), on a book which reached 2.4x at guidance of 52bps area.

Perhaps there’s some small standalone / debut premium — Coventry’s Economic Master Issuer 2025-1 started at 50bps area IPTs and should land inside that, while Nationwide’s SMI 2024-1 drop the week before (all three year trades) came at 47bps across a £500m tranche.

But hey this is three basis points we’re talking about! Going from a master trust veteran sponsored by the UK’s biggest building society to a brand-new standalone debut from a £6.2bn balance sheet minnow!

We also think it’s interesting to see the development of the master issuer structure in the UK market.

This is supposed to provide a simple, single SPV solution allowing lower hassle issuance, while maintaining the master trust advantages of scheduled amortisation. Coventry adopted it first, and has now done five deals (sixth currently in market) from its Economic Master Issuer shelf, while Yorkshire picked it up last year, mothballing standalone shelf Brass in favour of White Rose Master Issuer.

This year, Bank of Ireland (UK) made the switch as well, launching the debut offering from its Bowbell Master Issuer shelf in late May. It’s been a standalone issuer of UK RMBS since 2019, also under the Bowbell brand, with two deals placed with investors. Bowbell 3 from 2023 remains outstanding.

But neither Nottingham, with Lace Funding, nor Newcastle, with Hadrian, opted to head down this route.

Size and funding need may be partly behind it. Bowbell Master Issuer started off with £1bn of loans in the pool, while White Rose had £1.3bn (supporting placed deals of £350m in both cases). Coventry’s EMI vehicle has £2.46bn of loans in — getting on for half the size of Newcastle’s entire mortgage book.

Newcastle has a £355m in Bank of England TFSME maturities coming up through 2025, but it really doesn’t have a huge funding need; the need for greater execution efficiency probably only comes through for an institution that’s got substantial issuance plans.

Want to learn more about our newly launched subscription package, 9fin Asset-Based Finance? Email marketing@9fin.com for more information.